\' -* 
 
 THE LIBRARY 
 
 OF 
 
 THE UNIVERSITY 
 
 OF CALIFORNIA 
 
 LOS ANGELES
 
 This book is DUE on the last date stamped below 

 
 AN APPROACH 
 TO BUSINESS PROBLEMS 
 
 BY 
 ARCH WILKINSON SHAW, A.M. 
 
 LECTURER ON BUSINESS POLICY 
 IN HARVARD UNIVERSITY 
 
 EDITOR OF "SYSTEM" 
 
 CAMBRIDGE 
 HARVARD UNIVERSITY PRESS 
 
 LONDON: HUMPHREY MILFORD 
 
 OxFOBD Univebsitt Press 
 
 1920 
 
 7 :-> 1 7 8
 
 COPTRIGHT, 1916 
 
 BT 
 
 A. W. SHAW 
 
 1st impression of 1000 copies, Sept., 1916 
 2d impression of 1500 copies, Jan., 1917 
 3d impression of 1000 copies, Jan., 1917 
 4th impression of 2000 copies, Sept., 1918 
 5th impression of 1000 copies. May, 1920 
 6th impression of 2000 copies, Dec, 1920
 
 535 \ 
 i PREFACE 
 
 13EF0RE a science of business can take shape, there 
 '-^ must first be built up a system of business prac- 
 tice. The methods, plans, and rules of the most effi- 
 cient organizations must be brought together, tested, 
 and compared, and the most effective must be selected 
 c^ and coordinated. Some of this work has been done. 
 ^ Trade associations, universities, periodicals, and various 
 ■^^^ governmental agencies have been studying the details 
 of production and distribution. 
 
 But up to the present, no plan for the guidance of 
 , \ the student in the maze of business practice has been 
 r^ offered. To supply this deficiency, to discover a classi- 
 fication molded on the living activities of business, to 
 supply a uniform method of approach to business prob- 
 lems in whatever form they may arise, and to illustrate 
 the application of this method to typical problems, in- 
 cluding those involving the relations of business to 
 society — these are the purposes of this book. I need 
 hardly say that in all this the book makes no claim to 
 finality. If the classification and the uniform method 
 should do no more than suggest improvements upon 
 themselves, the work, I feel, would have been worth 
 while. But I venture to hope also that they will be of 
 some practical value in their present form: that they 
 will make it easier for business men and all thinking 
 men to see each activity of business in its relations to 
 others and to visualize and weigh all the significant 
 factors bearing on any business problem.
 
 PREFACE 
 
 The book is a development and rearrangement of ma- 
 terials used in a course of lectures begun by the writer 
 six years ago in the Graduate School of Business Ad- 
 ministration at Harvard University. It would be diffi- 
 cult to list all the individuals to whom I am indebted 
 for help with the lectures and the book — colleagues at 
 Harvard, members of the staffs of the magazines System 
 and Factory, and scores of business men with whom I 
 have debated the problems and policies here discussed. 
 In reorganizing and elaborating the lectures and in pre- 
 paring the present text, I have been aided particularly 
 by Mr. Daniel V. Casey, Dr. Selden O. Martin, Mr. R. 
 E. Coulson, Prof. Homer B. Vanderblue, and Mr. F. M. 
 Feiker. I can conceive no form of acknowledgment, 
 however, which would properly express my obligation 
 to Dean Edwin Francis Gay of the Graduate School of 
 Business Administration at Harvard. Without his in- 
 spiration and help, neither the lectures nor the book 
 could have the forms they have taken. 
 
 What have been the materials on which we have had 
 to draw for the lectures and the book ? To begin with, 
 there were the rules and guiding policies of practical 
 business men — the accumulated knowledge of years of 
 experience and observation. No man ever reaches a 
 position of power and responsibility without formulat- 
 ing certain general truths which he believes fundamen- 
 tal in the conduct of his enterprise. Let a new situation 
 come up, a fresh problem present itself, and he can 
 usually marshal this fund of information to an effective 
 settlement. Yet ordinarily he can contribute little that 
 is positive and general in application, outside his own 
 factory, office, or store. Men of larger mold, to be sure, 
 have viewed things nearer their real proportions and 
 
 iv
 
 PREFACE 
 
 have enunciated rules of action more broadly true and 
 applicable. But many of these precepts even, like that 
 of Marshall Field, " The customer is always right," 
 must be accepted with reservations, although they have 
 profoundly influenced merchandising ideals. 
 
 So with many business proverbs and maxims. One 
 frequently hears, for example, that " Goods well bought 
 are half sold." Yet this is misleading unless good buy- 
 ing comprehends a knowledge of all the complex con- 
 ditions which govern the market. Cheapness alone is 
 not a final test of good buying, for the loss from over- 
 stocking in order to secure price concessions may more 
 than counterbalance a price advantage. Applying the 
 proverb in a mere mechanical way, therefore, might 
 bring loss, if not disaster. 
 
 Of broader usefulness for our purpose have been the 
 generalizations contributed by economic theory, which, 
 as one of their greatest services, give a sense of direc- 
 tion to business thought. The important law of dimin- 
 ishing returns, for example, declares that beyond a 
 certain fixed point increased expenditure secures less 
 than a proportional increase in product. Cultivating 
 a market, that is to say, is much like cultivating a field. 
 Two " doses " of capital and labor may not bring twice 
 the returns of a single adequate "dose"; there is a 
 point at which the rate of return falls behind the rate of 
 increase in expense. Applying this principle to the field 
 of business, we find, for instance, that the larger the 
 number of salesmen in a given territory, the larger or- 
 dinarily will be the total number of orders. At some 
 points, however, the orders per salesman fall, although 
 the expense per salesman holds relatively constant. In 
 time, the point is reached where the return falls below
 
 PREFACE 
 
 the point of profitable expenditure. Other such gen- 
 eralizations have been drawn upon, and still others 
 proposed, in the body of the book. 
 
 But, after all, it is by observing the complex and 
 multiform activities of business itself that we can best 
 discover and verify the reactions and relations of all 
 these activities, and check and organize the knowledge 
 we already possess. 
 
 VI
 
 CONTENTS 
 
 INTRODUCTION 
 
 CHAPTER I 
 The AcTi^aTiES of Business 1 
 
 Every activity of business represents the application of motion to ma- 
 terials. WTien we inquire the purpose of each motion, we find it identifying 
 itself with one or three groups of activities, production, distribution, or ad- 
 ministration. The classification as an aid to finding useless motions, that 
 they may be eliminated, and determining others that may be usefully added. 
 Because production is concerned more with tangible factors and is better un- 
 derstood and systematized than either distribution or administration, this 
 book will develop and demonstrate a systematic approach to problems in the 
 first group and then apply it to those in the last two. Operating through all 
 departments of any business, however, will be found the fundamental prin- 
 ciples of interdependence and balance, giving rise to the manager's need of a 
 strategic position overlooking and controlling all departments but free from 
 routine concern with the details of any. Not only larger profit but also a 
 larger measure of individual and social freedom the reasonable effect of the 
 observance of these and other principles. 
 
 CHAPTER II 
 
 A Method of Approach to Business Problems . . 13 
 
 The activities of production divide according to their purpose into plant 
 and operating activities, the former concerned respectively with location, 
 construction, and equipment of the plant, and the latter with materials, 
 agencies, and the organization of factory processes. In administration and 
 in each of the two larger divisions of distribution, namely, demand creation 
 and physical supply, corresponding sub-groups of activities are found. This, 
 then, is a detail outline, a diagram, if you please, of business activities, at 
 every point of which practical problems arise. Qualities of thought which 
 help or hinder in the solution of these problems. Distinctive success largely 
 a matter of small advantages consistently cultivated and conserved. The 
 principle of cumulative differentials. The four steps in the systematic ap- 
 proach to business problems: (1) elimination — or at least recognition — of 
 the personal factor; (2) separation of the problem into its constituent prob- 
 lems; (3) listing the factors; and (4) taking a fresh point of view. Illustra- 
 tions of each. 
 
 vii
 
 CONTENTS 
 
 PART I 
 THE PROBLEMS OF PRODUCTION 
 
 CHAPTER HI 
 Location of the Plant 25 
 
 The systematic approach is applied to the problem of selecting a site for 
 a factory. A number of factors bearing on the suitability of any factory loca- 
 tion are indicated, including: accessibility of raw material sources and of 
 markets; control and cost of the more conveniently situated raw materials; 
 labor supply and ciu-rent wages in the locality; peculiar financial advantages, 
 if any; adaptability of the type of building to the business needs, to the site; 
 lighting conditions; water supply; drainage facilities; accessibility of a suit- 
 able residence district for employees; factory laws; taxes; insurance rates; 
 fire protection, etc. Analysis and listing of factors illustrated from the ex- 
 perience of a silverware factory and a drain tile plant. Relative advantages 
 of locations at and away from trade centers. 
 
 CHAPTER IV 
 
 Construction and Equipment of the Plant ... 40 
 
 The same approach is applied to the problems of choosing the best kind 
 of building and of selecting the most suitable equipment for it. Among the 
 considerations involved in the former are: the relative advantages of single- 
 story, multi-story, hollow square, oblong, and other styles of buildings, and 
 of brick, steel, concrete, mill, and other kinds of construction, taking into 
 account the funds available, the types of machinery to be installed, and the 
 site adopted. The availability of any type of building and of any kind of 
 building material will be influenced by the weights of raw materials and 
 manufactured parts to be moved, the cost of moving them from floor to floor 
 as compared with that of horizontal movement, the need of daylight illu- 
 mination, the exigencies of assembling, the probable growth of the building, 
 its salability, and other indicated factors. Illustrations from a New England 
 weaving mill, a Detroit stove foundry, and other plants. Factors bearing on 
 the choice of equipment, in turn, include: initial^cost and cost of upkeep, 
 output, patent rights, safety, etc. 
 
 CHAPTER V 
 
 Materials . 53 
 
 A factory's policy as to materials must be determined by analysis of con- 
 ditions having to do with (1) the kind of material to be used, (2) the quality 
 of material, (3) possible sources, (4) control, and (5) utilization in the fac- 
 tory. The kind selected should depend on future as well as present conditions.
 
 CONTENTS 
 
 on the probability of continued supplies of raw materials at reasonable prices, 
 and the possibility of introducing a satisfactory substitute in case of shortage 
 as well as on present salability and service. In determining the quality, the 
 management must be governed largely by financial and marketing consid- 
 erations — the selling price and the profit per unit of output. By control is 
 meant, particularly, the maintaining of proper quantities and balance of raw 
 stock on hand. The carrying of excessive raw stocks, according to efiiciency 
 engineers, is a common fault in American factories. The problem of utiliza- 
 tion, in turn, is finally solved only when the last by-product is used to the best 
 advantage. The value of laboratory standards. The systematic approach as 
 an aid in solving each of the five phases of the problem. 
 
 CHAPTER VI 
 Labor 64 
 
 The management's problem as to its factory workers is to overcome what- 
 ever indifference, suspicion, or antagonism they may entertain toward it and 
 to make clear that loyalty to the company is compatible with loyalty to 
 themselves and to labor organizations, where such organizations exist. The 
 most effective appeal is pay scientifically proportioned to individual effort. 
 In working out such a wage system the management must hold to its estab- 
 lished ratio of cost, quality, and service, cutting down cost where it consist- 
 ently can. The great defect of the traditional day rate is that it offers no 
 direct incentive to the individual to exert himself to his full capacity. The 
 ordinary piece rate, without at least an approximation to scientific standards, 
 gives too much play to the personal factor in fixing rates. Among the im- 
 provements on these methods the chapter discusses the Franklin, Towne, 
 Rowan, Halsey, Emerson, and Taylor systems. Industrial training as a 
 ftrrther incentive to the workman and as a policy profitable to the plant. New 
 outlook and opportunities opened to workmen by the training incidental to 
 scientific management. Various systems of instruction supplied by individual 
 plants alone and in cooperation with schools and universities. Eliminating 
 the personal factor is nowhere more desirable than in dealing with unions. 
 Thrift clubs, " welfare work," variation of employment, and other means for 
 reducing labor turnover. 
 
 CHAPTER VII 
 
 Organization 85 
 
 This chapter considers the various methods of delegating authority in a 
 factory, the management's broad problem here being to distribute authority 
 and supervision effectively and at the same time retain control. Three dis- 
 tinct types of shop organization are found in practice: (1) the line, (2) the 
 functional, and, midway between the two, (3) the line-and-staff. In line 
 organization, which is the traditional form, each department and sub-de- 
 partment head is held responsible for everything within his jurisdiction. His 
 functions are thus of many kinds: hiring the men, placing them, planning and 
 apportioning the work, fixing the payment, watching the machines, and so
 
 CONTENTS 
 
 on. The functional type, on the contrary, is the strict application of the 
 principal of the division of labor to shop management. First, it separates 
 planning from performance. Then each is subjected to subdivision of respon- 
 sibility. In a typical functional scheme, the planning is divided among an 
 order-of-work clerk, an instruction-card man, a time-and-cost clerk, and a 
 shop disciplinarian, while the direction of performance is divided among a 
 gang boss, a speed boss, a repair boss, and an inspector. A line-and-staff 
 organization may represent any one of a variety of compromises between the 
 line and functional types. Some one of the three types must be best adapted 
 to each individual business; the systematic approach helps to determine 
 which. 
 
 PART II 
 THE PROBLEMS OF DISTRIBUTION 
 
 CHAPTER VIII 
 
 Demand Creation and Physical Supply 99 
 
 Since the time when barter and sale by bulk ceased to be the universal 
 rule, distribution has embraced the two distinct functions of demand creation 
 and physical supply. The former is the process of communicating to prospec- 
 tive consumers ideas about the goods that w'll arouse desire for them and the 
 latter is that of transferring the goods to the consumer. The close relation of 
 the two functions again illustrates the principles of interdependence and 
 balance. To find markets rather than to supply those already active is in 
 general the more difficult problem of business to-day. Until within recent 
 years, the reverse was true. The long period during which the pressm-e was 
 upon production rather than distribution accounts in part for the fact that 
 distribution is now less systematized. The factors in distribution, including 
 the agencies of demand creation and physical supply, being more imperfectly 
 understood than those in production, it follows that the management must 
 give them more of its attention. The history of barter, sale by bulk, sale by 
 sample, and sale by description sketched. Improved machinery, educa- 
 tion, and means of communication among the factors in the development. 
 Broadly, the problem of distribution is to arouse the desired maximum of 
 demand at a minimum of expense, and to supply this demand with the least 
 possible leakage. 
 
 Section I — Demand Creation 
 
 CHAPTER IX 
 
 Location, Construction, and Equipment of the 
 
 Plant 115 
 
 The method of approach to the problems of locating, building, and equip- 
 ping a plant, developed in the production section of this book, is here applied 
 to the corresponding problems of demand creation. Shall the selling plant be
 
 CONTENTS 
 
 located at the factory, at the largest neighboring city, at the largest trade 
 market, or at the chief city of the country? TMiat type and size of quarters 
 shall it occupy? How shall they be equipped? All these questions are con- 
 sidered from the standpoints of the production, administrative, and distribu- 
 tive departments of the business. Illustrations from the furniture and other 
 trades. Construction and equipment problems, in particular, are governed by 
 considerations similar to those applicable to factory construction. Advertis- 
 ing value as a factor. 
 
 CHAPTER X 
 
 Materials 131 
 
 Following the order of the classification, this chapter takes up the first of 
 the operating activities of demand creation, those having to do with ma- 
 terials. The materials of demand creation are ideas which, when put into 
 circulation by certain available agencies, will cause consumers to want the 
 goods — the agencies being middlemen, salesmen, and direct and general 
 advertising. The ideas can be handled as definite, almost tangible things. 
 The effect they may be expected to produce can be determined with a degree 
 of accuracy not yet generally realized. Moreover, they can be classified, 
 indexed, and filed for future use. Like the materials of production, the ideas 
 must be considered from the standpoints of their (1) kind, (2) quality, (3) 
 sources, (4) control, and (5) utilization. Certain elements of good and bad 
 practice in these various phases of the problem are discussed. The problem 
 is further concerned, as in production, with the adoption and development of 
 laboratory standards. Practical methods of testing ideas by direct advertis- 
 ing and the law of averages are described and illustrated. Results of tests 
 uphold these methods. Similar laboratory tests through salesmen and other 
 agencies are practicable. 
 
 CHAPTER XI 
 
 Agencies of Demand Creation — The Middle- 
 man 154 
 
 Beyond obtaining the most effective ideas for demand creation, the man- 
 ager's problem is to determine which of the available agencies — middlemen, 
 exclusive salemen, and direct and general advertising — are adapted to trans- 
 mit the ideas most effectivelj' to possible purchasers. One, two, or even all of 
 the agencies may be used by a business. The tendency of many producers to 
 go around the middleman is considered in the light of the development of the 
 middleman system. The necessary functions the middleman performs: shar- 
 ing the risks; transporting the goods; financing the operations; selling or 
 creating demand; and assembling, resorting, and shipping the goods. The 
 rise of other agencies for performing these functions: insurance companies, 
 express companies and the parcel post, commercial banking, direct salesmen, 
 advertising mediums and agencies, factory warehouses and branches, etc. 
 The middleman's reluctance to reduce his percentage of profit as certain of 
 his functions have been assumed by other agencies — the function of demand 
 creation by producer advertising, for example — is one cause of the growth of
 
 CONTENTS 
 
 direct selling. His aversion (in many instances) to new brands, based often 
 on what from his point of view are sound enough reasons, is another. The 
 most efiFective answers of middlemen to the tendency have been combina- 
 tions for buying, the adoption of house brands, controlling factories, and 
 manufacturing on their own account. Summed up, the movement as a whole 
 is one of consolidation, on the part of trade as well as of industry, its natural 
 effect being the reduction of distributive wastes. 
 
 CHAPTER XII 
 
 Agencies of Demand Creation — Direct Sales- 
 men 178 
 
 Where the character, quality, or use of an article is not evident but re- 
 quires the demonstration or development of ideas imfamiliar to the trade and 
 the consiuner, the work of transmitting its selling points to the final user 
 through two or more middlemen becomes difficult and uncertain. Hence, 
 for an increasing variety of non-staple articles, the direct salesman is recog- 
 nized as an essential agent in demand creation. The producer's initial prob- 
 lem as to a direct sales force is to determine whether such a force is necessary 
 to create a maximum demand for his goods with a minimum outlay of money. 
 Once he has decided to use direct salesmen, he finds that his problems with 
 them, as with factory workmen, center about four chief points of contact 
 with the men: (1) hiring, (2) training, (3) paying, and (4) directing. Psy- 
 chologists and business men are not as yet agreed on anything approaching 
 scientific tests for salesmen, and so the average employer still depends on 
 personal " sizing up," together with records of past performances. While the 
 saying, " A salesman is born, not made," reflects a certain truth, it is more 
 significant that the effectiveness of salesmen can be multiplied by intelligent 
 training. Methods of training described. The value of standardized selling 
 points and processes reemphasized. The broad problem of paying and hand- 
 ling is to keep the salesman exerting his best efforts after he has achieved what 
 he conceives to be his standard. Standards based on close analysis of selling 
 areas are part of the solution. Sales contests are useful in some circumstances 
 and inadvisable in others. To pay a salesman too much or to pay him the 
 right amount in the WTong way may produce as bad results as paying him too 
 little. Relative advantages of salary and commission payment. As with 
 factory labor, again, the best payment is in general that which is most scien- 
 tifically proportioned to individual effort and results. But the management 
 can never have a satisfactory relation with its salesmen unless back of all its 
 methods is a genuine friendliness for the men. 
 
 CHAPTER XIII 
 Agencies of Demand Creation — Advertising . . 199 
 
 Advertising is an outgrowth of sale by description. It may be employed 
 either as a substitute for middlemen and salesmen, for pm-poses of demand 
 creation, or as an auxiliary to aid them in their exercise of the selling function. 
 It tends to displace these other agencies, in whole or in part, whenever it is 
 a less expensive or more effective means of commimicating ideas to the con- 
 
 xii
 
 CONTENTS 
 
 sumer. Illustrations showing how it justifies itself from the social as well as 
 the business point of view. When waste occurs in advertising, it is generally 
 due to one of five things : lack in the product of those elements of quality and 
 service which appeal to the consuming public's need or desire, ignorance of 
 the true function of advertising, blundering application of recognized prin- 
 ciples, failure to develop laboratory standards for ideas, or neglect to keep 
 in operating balance the other essential agencies of distribution. The char- 
 acteristics which peculiarly commend advertising are its expansive reach, its 
 economy, and the complete control it allows the management over the pres- 
 entation of ideas to prospective buyers. Factors bearing on the formulation 
 of a policy as to advertising. 
 
 CHAPTER XIV 
 
 Organization of Demand Creation — Analysis of 
 
 THE Market 219 
 
 The problems incident to organizing for demand creation fall into three 
 groups, which are concerned respectively with analysis of the market, fixing 
 a price for the goods, and the combination and coordination of agencies, each 
 of the three groups being an important factor in the others. The business 
 man's broad problem in analyzing his market is to find where his product 
 can be marketed profitably, and in approximately what quantities. In ap- 
 proaching the problem, he finds an indefinite body of possible purchasers, 
 widely distributed geographically, living under the various conditions of 
 town and country, and exhibiting various degrees of purchasing power and 
 intelligence, various tastes and beliefs and both conscious and unconscious 
 demand. Every locality in which it is proposed to distribute his product 
 should be examined with respect to these and still other conditions. The 
 method of examination illustrated. Tests of typical localities give typical 
 reactions, and thus show, at relatively small cost, the desirability or unde- 
 sirability of using the same policy in a larger field. How a market analysis 
 made it possible to equalize the seasons of a department store. A suggestive 
 list of factors to be considered in analyzing a market. 
 
 CHAPTER XV 
 
 Organization of Demand Creation — Price Poli- 
 cies 234 
 
 Not the cost of production but the requirements of the market are the 
 primary consideration in fixing prices. The manager has the choice of three 
 general price policies: (1) selling at the market minus, (2) selling at the mar- 
 ket par, and (3) selling at the market plus. The first policy aims to increase 
 volume of sales by reducing price. Ordinarily, it does not involve difiFerentia- 
 tion from other stock products of like nature. Selling at the market minus is 
 the easy method of securing distribution, although other than economic mo- 
 tives govern sales in many instances and the middleman's indifference be- 
 comes an obstacle unless it is turned into active interest by the prospect of 
 larger profit on the new article. Subject to certain restrictions, the policy 
 tends not only to draw purchasers away from competitive goods but to bring 
 
 ziii
 
 CONTENTS 
 
 new purchasers into the market. To sell at the market par, the merchant 
 must, in general, differentiate his product from those of his competitors and 
 build up a particular demand for it. Under this policy, new purchasers may 
 be drawn into the market by giving the product an added importance in their 
 eyes. Selling at the market plus is perhaps the most characteristic feature of 
 modern distribution. It makes the severest demands on the ability of the 
 distributer and offers correspondingly great rewards for success. Usually the 
 product is more effectively differentiated, more closely adapted to human 
 needs or desires, and the large profits it sometimes brings — not infrequently 
 out of all proportion to the added cost of manufacturing the article — may 
 be regarded as the price society pays for this special service. 
 
 CHAPTER XVI 
 
 Organization of Demand Creation — Combination 
 
 OF Agencies 255 
 
 The agency that is most effective for demand creation in one geographic 
 or economic or social division of the market may not be so in another. The 
 manager's problem in this field, therefore, is to deiermine not merely which 
 agency is most effective in any single section or which will serve passably for 
 all sections, but what combination of agencies is most effective for the entire 
 market. The method of sale — by bulk, sample, or description — as a factor 
 in the choice of agencies. Facilities and difficulties offered by nuddlemen. 
 In directly assuming the burden of demand creation, the manufacturer has 
 the choice of two agencies, an exclusive sales force and direct or general ad- 
 vertising. Combination of the two is often desirable, because, while adver- 
 tising has the broader reach, the direct salesman can adapt himself more 
 precisely to individual cases. And while advertising usually allows the more 
 careful arrangement of ideas, the salesman has the advantage of spontaneity 
 and personal presence. The salesman's function is often to follow up " leads " 
 that result from advertising. Salesmen and direct advertising may also be 
 used separately or together in combination with middlemen. When all are 
 used, it may be preferable for the middleman to devote himself altogether to 
 physical supply or to cooperate with the others in demand creation. The 
 interest of each agency no less than of the purchasing public is served best 
 when each agency assumes the functions which it can most efficiently perform. 
 
 Section II — Physical Supply 
 
 CHAPTER XVII 
 
 Plant and Operation Policies 275 
 
 The broad problem of physical supply is to satisfy the maximum of aroused 
 demand with minimum demand leakage and expense, present and future 
 sales both being considered. If minimum leakage of demand is to be attained 
 — and it is extremely important — the principles of interdependence and
 
 CONTENTS 
 
 balance must be observed consistently as between production, physical sup- 
 ply, and demand creation, that the first two may keep pace with the last. 
 The activities of physical supply, like those of production and demand crea- 
 tion, break up into plant and operating activities, and these, in turn, into 
 corresponding sub-groups. Warehouses and their attendant shipping facili- 
 ties are the plant. The goods themselves are the materials. And the agencies 
 may be: (1) functional middlemen in the transportation field, including rail- 
 roads, express companies, and the parcel post; (2) areal middlemen, including 
 wholesalers and retailers; or (3) agencies of direct supply, including branch 
 houses and exclusive agencies or retail stores. The organization sub-group 
 embraces those persons and arrangements through which physical supply is 
 immediately directed and supervised. The systematic approach indicated 
 for each sub-group, and characteristic factors suggested. 
 
 PART III 
 THE PROBLEMS OF ADMINISTRATION 
 
 CHAPTER XVIH 
 
 Plant and Operation Policies 295 
 
 The facilitating operations of business, including those having to do with 
 finance, purchasing, employment, credits, collections, accounting and audit- 
 ing, suggest unit handling under a common executive, with authority corre- 
 sponding to that of the factory superintendent and sales manager within their 
 fields, or at least recognition of their equality as a group with the other two 
 major divisions of business. This view simplifies the maintenance of inter- 
 departmental balance and contributes to efficiency. The activities of the 
 administrative group break up in accordance with the general classification 
 and yield themselves to the systematic approach. The nature and conditions 
 of the various sub-groups of activities considered. The character and scale of 
 the business help to define their scope. The manager's position, aloof from 
 routine concern with facilitating activities as well as those of the other de- 
 partments, leaves him free to sense and solve the larger activities of his busi- 
 ness. 
 
 Conclusion 
 
 CHAPTER XIX 
 
 The External Problems of Business 321 
 
 The business executive of to-day is not the autocrat of a private under- 
 taking, able to base his policies on personal whim or desire or an entirely 
 selfish conception of business. Society is asking more seriously than ever 
 before, what besides the market minimum of value and service at the price 
 he is going to supply as its profit on its investment of community machinery 
 and opportunity. Besides the internal problems of business, therefore, the 
 
 XV
 
 CONTENTS 
 
 manager has also a pressing set of external problems to solve. These problems 
 have to do not merely with the shaping of legislation, but with public opinion, 
 upon which the laws are formed. His task is not only to influence public 
 opinion, but to be influenced by it in the formulation of his policies, when- 
 ever justice and prudence dictates that course. Recent business legislation 
 considered. Mutual understanding between business and society and co- 
 operation in behalf of the common interest is the great solution. 
 
 XVI
 
 INTRODUCTION
 
 AN APPROACH TO BUSINESS 
 PROBLEMS 
 
 CHAPTER I 
 THE ACTI\aTIES OF BUSINESS 
 
 WHEN a workman in a factory directs the cut of a 
 planer in a malleable steel casting, he is operating 
 on a piece of raw material for the purpose of changing 
 its form. 
 
 When a clerk in a store passes over the counter to the 
 consumer a package of factory-cooked food, his opera- 
 tion is one that results in change of place. 
 
 When a typist at his desk makes out an invoice cov- 
 ering a shipment of merchandise, he is operating, not to 
 change the form of matter or the place of commodities, 
 but to facilitate these changes. 
 
 Isolate any phase of business, strike into it anywhere, 
 and the invariable essential element will be found to 
 be the application of motion to materials. This may 
 be stated, if you will, as the simplest general concept 
 to which all the activities of manufacturing, selling, 
 finance, and management can ultimately be reduced. 
 
 Starting with this simple concept, then, it becomes 
 evident that we have an easy and obvious basis for the 
 classification of business activities. With the philo- 
 
 1
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 sophical aspects of the concept itself I am not here con- 
 cerned. SuflBcient that it gives us a simpHfying, unify- 
 ing principle from which to proceed, instead of a mere 
 arrangement by kind or characteristic, of the materials, 
 men, operations, and processes which we see in the 
 various departments of a business enterprise. 
 
 The nature of the motion does not of itself supply the 
 key to a usable classification. For while the action may 
 be characteristic of one part of a business and not dupli- 
 cated elsewhere, like the pouring of molten metal in a 
 foundry or the making up of a pay roll, it may, on the 
 contrary, be common to all the departments into which 
 the organization is divided, like the requisition of a 
 dozen lead pencils or a box of paper clips. It is not until 
 we single out the common fundamental element and 
 inquire " What is the purpose of this motion? " that we 
 find the key. 
 
 I do not wish to exaggerate the importance of this 
 simple and apparently obvious idea; but for me it has 
 opened a way to locate the activities of business and 
 disclose their relations to one another and to their com- 
 mon object, and so has proved a thing of daily use. For 
 the final function of the classification, as it is the prac- 
 tical problem of all business, is to find those motions 
 which are purposeless, so that they may be eliminated, 
 and to discover new motions of sound purpose, that 
 they may be introduced. 
 
 If you study an individual motion or operation in it- 
 self and in relation to the other activities surrounding 
 it and no satisfactory answer can be found to the ques- 
 tion, " What is its purpose.'* " you have strong grounds 
 for assuming that it is a non-essential and useless mo- 
 tion. It may have the sanction of house tradition or 
 
 2
 
 THE ACTIVITIES OF BUSINESS 
 
 trade custom, but its superfluous character persists and 
 the wisdom of eliminating it becomes plain. Conversely, 
 a motion proposed for adoption may never before have 
 been tried in the trade, but that alone is not an argu- 
 ment against it. Purpose again is the decisive test. 
 From the social standpoint, any motion which has no 
 valid purpose is economically useless and wrong. The 
 effect of such a motion in business, like the effect of 
 omitting a useful motion, is to keep down profits when 
 they might reasonably rise. 
 
 From the manager's point of view, then, the purpose 
 of the analysis is not alone to place the activities of 
 business and trace out their relations, but to order his 
 thought so that he can more readily see what activities 
 he should discontinue and what others he should add. 
 
 This does not always mean a reduction in the total 
 number of motions. In our roundabout system of pro- 
 duction,^ with its minute subdivision of labor, it is pos- 
 sible to make a greater number and variety of motions 
 and distribute them over a longer period of time, yet 
 increase the eventual output or decrease the cost 
 through the group effectiveness of all the motions. 
 
 In the three operations mentioned above — those of 
 the factory workman, the retail clerk, and the oflSce 
 typist — each application of motion was for an eco- 
 nomically valid purpose and each instance was typical 
 of one of the three great groups of business activities: 
 
 1. The activities of 'production, which change the 
 form of materials. 
 
 2. The activities of distribution, which change the 
 place and ownership of the commodities thus produced. 
 
 3. The facilitating activities, which aid and supple- 
 
 1 See E. V. Bohm-Bawerk, " Positive Theory of Capital," pp. 17-20.
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ment the operations of production and distribution. No 
 adequate descriptive term suggests itself ; facilitation is 
 awkward, administration is inexact because it suggests 
 all the relations of the management to a business rather 
 than a separate group of functions toward which the 
 manager should maintain the same sort of relations as 
 with those of production and distribution. I shall, how- 
 ever, for convenience use the term administration in 
 this narrower sense. 
 
 Whatever the nature or kind of any business activity, 
 its final effect is one of the three just indicated. Such a 
 categorical grouping is necessary in any schematic ap- 
 proach to the problems of business; for out of the rela- 
 tions of these activities, each with the others and with 
 the materials affected, emerge those problems with 
 which this book is concerned. 
 
 Before we take up the study of the individual prob- 
 lems, it may be well to make a brief general survey of 
 the three groups and note certain likenesses and differ- 
 ences that characterize these groups of activities. 
 
 Production is relatively well standardized. In any 
 scrutiny of practical business, it becomes clear that the 
 manager concerns himself less with the activities of pro- 
 duction than with those of distribution and administra- 
 tion. The problems of the factory as a rule demand, on 
 his part, less of analysis and synthesis, less of study and 
 experiment and constructive thought because produc- 
 tion has been brought much nearer to standardization 
 in its operations and policies. 
 
 For upwards of a century economic conditions di- 
 rected the attention of the ablest business minds to the 
 perfection of manufacturing processes rather than the 
 development of sales methods. The market expanded
 
 THE ACTIVITIES OF BUSINESS 
 
 fast enough to absorb the increased offerings: the big 
 prizes went to the men who could speed up production 
 without adding to the expense of a unit of product. Im- 
 provement of factory methods and processes, too, was 
 the manifest way to economy and profit. The factors 
 involved — machines, materials, the actions of men — 
 were all concrete, visible things which daily thrust them- 
 selves on the attention of the manager. Short cuts al- 
 most suggested themselves. Lost motions and other 
 inefficient applications of human or mechanical energy 
 made themselves apparent to men of open minds and 
 demanded study until the wastes were reduced and 
 better methods inaugurated. Because every operation 
 was easily distinguishable from every other one, the ad- 
 vantages attending the division of labor were discovered 
 and the application of the principle extended. This 
 took place, however, without affecting the processes of 
 distribution in any marked degree. 
 
 As a consequence of this century of invention, de- 
 velopment, and refinement, production is, in point of 
 systematized economy, far more advanced than dis- 
 tribution. Moreover, there have developed, on the 
 factory side, recognized classes of specialists — tech- 
 nically trained superintendents, engineers, chemists, 
 designers, foremen — who are capable of assuming, and 
 frequently do assume, all the responsibilities of pro- 
 duction. 
 
 In distribution and administration there enters a dif- 
 ferent set of factors, more largely psychological. Men 
 meet not only as units in a business organization — as 
 colleagues, subordinates, and superiors — but also in the 
 delicate personal relations of seller and buyer. Produc- 
 tion problems are usually problems involving the cost 
 
 5
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 of materials, of labor, of equipment, and so on. In dis- 
 tribution, on the contrary, the factor of cost is over- 
 shadowed by that of price, and the latter is determined 
 largely, at times, by changing conditions over which the 
 seller has very little control. Competition, encountered 
 to some extent in the field of production, in the hiring 
 of men, the buying of materials and the like, is in the 
 field of distribution almost ever-present and frequently 
 acute. 
 
 Finally, there are few trained experts in distribution 
 and administration, and few absolute standards of prac- 
 tice, except perhaps in the special field of accounting. 
 The approach to the problems of these groups, there- 
 fore, as contrasted with those of production, is still 
 largely empirical. 
 
 It would seem, then, the logical plan to consider first 
 the activities of production, since here conditions, reac- 
 tions, and results have been reduced to something like 
 accepted standards; then, after developing and demon- 
 strating in this known and charted field a method of 
 analysis, a standard classification and a systematic ap- 
 proach to business problems, to apply this approach and 
 analysis to distribution and administration, about which 
 so much less is known and the factors of which are all 
 less tangible, less understood, and more variable. 
 
 The function of the general manager of a business is 
 to coordinate and direct these three groups of activities. 
 Therefore it is practical as well as convenient that we 
 should approach them from his point of view, the course 
 of our approach itself indicating what that point of 
 view should be. This will bring out and establish the 
 permanent strategic position he should occupy toward 
 all three. Further, in treating general policies involving 
 
 6
 
 THE ACTIVITIES OF BUSINESS 
 
 coordination of the problems arising under each, a uni- 
 form method of approach will be outlined and applied. 
 
 Every manager maintains and must maintain certain 
 definite policies toward production, distribution, and 
 administration. In determining what these policies 
 shall be, there are at least two basic principles of which 
 he must never lose sight. The first of these is the inter- 
 dependence of all business activities. Each has rela- 
 tions more or less direct with all other activities. No 
 problem can be solved, therefore, without making al- 
 lowances for the connection the solution may have with 
 other problems. 
 
 In their broader aspects, this mutual interdependence 
 extends to all the activities of production, distribution, 
 and administration. No business problems are strictly 
 intra- departmental. They all have implications reach- 
 ing out and affecting activities in other departments. 
 The extent and importance of these is unobserved by 
 many managers at present; hence, most businesses are 
 out of adjustment. 
 
 Every man of experience will recognize this interde- 
 pendence, I think, as a fundamental principle in bus- 
 iness. Likewise the corollary principle, that a balance 
 must be maintained in these relations. The ratio of cost, 
 quality, and service observed in the actual manufacture 
 of a product, for example, must be the same as that put 
 forward by the sales force. Otherwise customers will be 
 dissatisfied, internal friction will develop, and efficiency, 
 good will, and profits will suffer. If, on a scale of 100, 
 the stress on quality stands at 60 in the production de- 
 partment, it must stand at 60 in the distribution and the 
 administration departments. The credit policy must 
 balance with the production policy, the purchasing with 
 
 7
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 the selling, and the selling, again, with the production; 
 and so on. If production is under pressure to manufac- 
 ture at low cost with relative indifference to quality, 
 while distribution bases its appeals upon quality and 
 service, the business is clearly out of adjustment. The 
 production force is not delivering the kind of thing that 
 the distribution force is trying to sell. 
 
 The lack of proper adjustment and balance in many 
 businesses is not surprising when we remember that the 
 average business exists because the man behind it is a 
 specialist. He has capitalized his buyer's instinct for 
 values, his salesman's tact and enthusiasm, or his ability 
 and patience as a shop organizer. He has built on this 
 special talent and comes to depend on it too exclusively. 
 This departmental bias, if permitted to run its course, 
 carries the grave consequences of a failure in efficiency. 
 The department on which the manager concentrates 
 must carry the burden of those he neglects ; as a conse- 
 quence, the growth of the business is impaired. 
 
 Perhaps the worst effect of departmental bias is seen 
 in what might be called the tangential development that 
 takes place in certain cases of joint or composite man- 
 agement, where, for instance, a man strong in selling 
 takes for a partner a man strong in production, and the 
 two proceed to manage the business together, each 
 stressing his specialty without regard for the principle 
 of interdependence. Instead of a straight parallel pull 
 toward a common goal, such as observance of the prin- 
 ciple of interdependence would insure, their pull be- 
 comes more and more tangential, their angles with the 
 business become wider and wider, until finally each is 
 pulling squarely against the other, and the business 
 comes to a standstill. 
 
 8
 
 THE ACTIVITIES OF BUSINESS 
 
 I once heard a production engineer in New York speak 
 of selling as a " necessary evil." The owner of a great 
 business who was present disagreed with him and said 
 quite as emphatically that selling was the one vital 
 function. Knowing that the sales enthusiast had built 
 up a remarkable business from a small beginning, I took 
 the trouble to check his statements about his company's 
 experience. What I found was this. The owner per- 
 sonally concentrated on selling and honestly believed 
 that selling was the function which contributed most to 
 the success of a business. But he had instinctively 
 gotten hold of the principle of interdependence, and had 
 surrounded himself with men in other departments who 
 were strong in their lines and who supplemented his own 
 specialty without straying into tangential development. 
 Thus the activities of production and administration 
 were properly coordinated with those of distribution. 
 The business was a success because the owner had un- 
 consciously observed the principle of interdependence 
 without formulating it or perhaps without being aware 
 that it was something to be formulated. 
 
 A very commonplace illustration will show how in- 
 timately the principles of interdependence and balance 
 apply in the mere routine detail of department policy, 
 and also show how rapidly the manager's thought must 
 pass from considerations involving the one principle to 
 those involving the other. Take the case of a concern 
 proposing to install an automatic conveyor in its fac- 
 tory. Here are some of the questions which a consider- 
 ation of these principles should bring at once to the 
 mind of the manager : 
 
 Is the estimated charge for interest, upkeep, and de- 
 preciation more than the present cost for trucking? If 
 \ 9
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 so, of how much importance is this factor of added ex- 
 pense? To what degree does the additional tax on capi- 
 tal operate against the change? If speedier work can be 
 done in the shop, quicker delivery and a more rapid 
 turnover may cut the overhead to counterbalance the 
 added expense. How important, then, is quicker de- 
 livery as a factor in distribution? 
 
 What will be the effect on the rest of the production 
 process? Will the automatic conveyor facilitate or re- 
 tard the progress of parts or partly finished goods from 
 machine to machine or department to department? 
 Will a fixed rate of progress for work lead to carelessness 
 and hurry in any section and the possible sacrifice of 
 quality? Or will the conveyor, by reason of the more 
 even flow of work in process, contribute to both qual- 
 ity and service? Will there, finally, be an advertising 
 advantage in the use of the conveyor, like the " un- 
 touched-by-human-hands " argument which in the case 
 of a food product might be a distinct factor in the pop- 
 ular estimate of its desirability ? 
 
 Here we may plainly see to what extent an apparently 
 intra-departmental problem spreads out and becomes 
 inter-departmental; and also how the constituent fac- 
 tors should be judged with due regard for a balance be- 
 tween cost, quality, and service. 
 
 Since so much depends on strict conformity to these 
 two principles, the manager should be placed where he 
 can observe them impartially. He should occupy a 
 detached, supervisory position over all departments, 
 free from routine concern with the details of any. Or, if 
 the business is small and it is necessary for him to serve 
 also as a department oflficial, he should cultivate the 
 habit of withdrawing mentally to the position here in- 
 
 10
 
 THE ACTIVITIES OF BUSINESS 
 
 dictated when performing his managerial functions. I 
 was careful to say that the manager should be free from 
 routine concern with departmental details, because he 
 will probably spend the greater part of his time dealing 
 with symptomatic details whose importance is not to 
 be measured by their size or the amount involved. A 
 customer's complaint involving thirty-five cents may 
 in itself be a trifle. If it points the way to a weak link 
 in distribution, however, and this in turn shows a cor- 
 responding lack of coordination in production or ad- 
 ministration, it becomes significant and presents a 
 problem for the manager. 
 
 In the position described, aloof from the various 
 groups of his business activities, the manager can more 
 readily free himself of the tendency toward bias, which 
 is exceedingly strong when he keeps a routine connec- 
 tion with any one department, or involves himself any- 
 where in non-symptomatic detail. He can keep a clear 
 vision and true perspective of the whole field. He can 
 coordinate the functions of the three departments, 
 shaping them all in accordance with the relations which 
 he wishes to maintain between his business and the 
 public. It is only from such a detached position, in- 
 deed, that he can see clearly what these relations should 
 be; can take from the public the suggestions and in- 
 centives which he needs, and can gain a broad and com- 
 prehensive view of the workings of his own organization 
 and the possibilities of the markets that it serves. 
 
 I have said of the principles of interdependence and 
 balance that at least these two should be observed by 
 every manager. Are there not other principles just as 
 simple and profitable as these, some already known to 
 business practice but others yet to be detected and de- 
 ll
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 fined? Is it too much to hope that the movement of 
 analytical business thought may in time distinguish a 
 whole body of business principles — principles now 
 operating unnoticed wherever there is a factory, a store, 
 or an office? Not, it seems to me, until we gain by 
 some such process a clearer understanding of the forces 
 with which we are dealing can we realize the fullness 
 of individual and social freedom into which industry is 
 capable of bringing us. 
 
 12
 
 CHAPTER II 
 A METHOD OF APPROACH TO BUSINESS PROBLEMS 
 
 WE have seen that every activity of business, what- 
 ever its nature, will find itself in one of three great 
 groups of business activities — production, distribu- 
 tion, or administration. On closer examination, it is 
 seen that the activities of each group yield themselves 
 to a further classification, and that this classification is 
 almost identical for each group. 
 
 Thus the activities of production split according to 
 their purpose into those having to do with the prepara- 
 tion of a ready-to-run factory, which for convenience 
 we may call plant activities, and those having to do 
 with the operation of the factory, which will be termed 
 operating activities. Plant activities divide into those 
 concerned with (1) location, (2) construction, and (3) 
 equipment of the plant. Operating activities are those 
 relating directly to (1) materials, (2) the agencies proc- 
 essing the materials, with labor as the first to be con- 
 sidered, and (3) the coordination and direction of factory 
 processes. 
 
 In distribution, two distinct functions dictate a broad 
 preliminary division. One of these, demand creation, 
 is concerned with the making of a market for the goods, 
 through the transmission of ideas about them to pos- 
 sible purchasers. The other function, that of physical 
 supply, consists of the operations necessary to put the 
 buyer in possession of the goods. But the activities of 
 
 13
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 each of these groups break up into two sub-groups like 
 those of production : plant activities, which are occupied 
 with location, construction, and equipment, and operat- 
 ing activities, which relate to materials, agencies, and 
 organization. The materials of demand creation are 
 ideas about the goods; in physical supply they are the 
 actual merchandise delivered. 
 
 Again, in the third major division of business, the ac- 
 tivities of administration group themselves along the 
 same lines as in production and distribution. On the 
 plant side, they closely parallel those of demand crea- 
 tion, because of the physical likeness of the functions 
 performed. Operating policies, however, although they 
 have to do with the usual three groups of activities — 
 materials, agencies, and organization — are governed 
 in part by factors which do not appear elsewhere. So 
 far as its chief function goes, each administrative de- 
 partment has direct relations with the manufacturing 
 or selling department whose work it facilitates, as well 
 as with the other units of its organization group. In 
 addition, the materials to which motion is applied are 
 paper representations of the activities facilitated. 
 
 Here, then, are the three great general groups of busi- 
 ness activities, with numerous subordinate groupings 
 under each. All along the lines of the classification, at 
 this point and at that, arise the practical problems of 
 business. Above and outside the groups stands the 
 manager. Keeping constantly in mind the vital prin- 
 ciples of interdependence and of balance, his task is to 
 solve these problems. 
 
 Before developing a definite method of approach to 
 their solution, however, something may be said about 
 the qualifications and general state of mind of the suc- 
 
 14
 
 A METHOD OF APPROACH 
 
 cessful business man. Calm, deliberate analysis is an 
 important factor; imagination, judgment, courage, and 
 executive ability are others. To quote Professor Taus- 
 sig: " Executive ability is probably less rare than the 
 combination of judgment with imagination. But it is by 
 no means common. It calls, on the one hand, for intel- 
 ligence in organization; on the other hand, for knowl- 
 edge of men. The work must be planned, and the right 
 man assigned to each sort of work." ^ 
 
 How does the able manager differ from the ordinary 
 business man.? Or the conspicuously successful concern 
 from that whose margin of profit is never very wide or 
 safe? 
 
 I am inclined to think that the difference between 
 moderate and distinctive success in business is in the 
 main just a sum of individually small advantages. The 
 average employer, for instance, contents himself with 
 the run of the labor market, and pays approximately 
 the market price. The able manager, on the contrary, 
 recognizes that there are many grades of executives, 
 salesmen, mechanics, and clerks, and that the difference 
 in productive capacity between competent and mediocre 
 workers is almost always greater than the difference in 
 the cost of their labor. He makes it his business, there- 
 fore, to pick out and hire the best individuals available 
 for each type of service required, going beyond the 
 local market if that supply is unsatisfactory. 
 
 From this careful selection of help, he gets his first 
 differential of profit on labor. But it is only the first in 
 a series. He has learned that even skilled and indus- 
 trious men are capable of much greater effort than, of 
 themselves, they are likely to put forth. His part, then, 
 
 ^ F. W. Taussig, " Principles of Economics," Vol. II, page 164. 
 15
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 is to supply the conditions and the stimulus which will 
 tap these reserve powers and bring them to bear on the 
 day's work. All along the line, such further differen- 
 tials come to him: from his control of operations, the 
 quality of the training and supervision he gives his men, 
 the class and condition of his equipment, the conven- 
 ience and certainty of his system of work-supply, his 
 method of paying for work accomplished, and a host of 
 indirect factors like the lighting, heating, cleaning, and 
 ventilation of his workrooms and offices. 
 
 Moreover, all these advantages tend to augment and 
 multiply one another. Greater output is not the only 
 effect of the able manager's discriminating labor policy. 
 He gets also a more even, if not a superior, product. 
 This builds up his " good will " and facilitates sales. 
 Increasing sales and growing prestige in turn invite 
 favors from raw material supply houses, banks, and 
 transportation companies; and these favors tend pro- 
 gressively to greater volume of business, smaller unit 
 costs, and larger profits. So the small advantages, if 
 consistently cultivated and conserved throughout a 
 business, make in the long run an enormous difference. 
 
 Here is indicated a further principle, that of cumu- 
 lative differentials, which no manager can well ignore. 
 For who will say that even the notable innovators, men 
 like McCormick, Field, Henry Ford, the Butlers, and 
 Montgomery Ward, could have developed great ideas 
 into mighty businesses without constant observance of 
 this principle. 
 
 But how characterize the unsuccessful? Broadly 
 speaking, there are at least two classes of business men 
 whose inefficiency is due to faults of temper. There is 
 the excitedly busy man who attempts to know every 
 
 16
 
 A METHOD OF APPROACH 
 
 detail, and so loses what is significant in the mass of 
 what is not. His desk is piled with papers. He always 
 insists that no step, however trivial, be taken without 
 his immediate supervision. It is of this kind of ineffi- 
 ciency that Bagehot ^ was thinking when he wrote that 
 if the head of a large business " is very busy, it is a sign 
 of something wrong." Another class of inefficients is 
 composed of the trouble borrowers. Here the typical 
 man anticipates his problems. He expends his energies 
 in worrying about questions which never arise, or when 
 they do arise, solve themselves almost automatically. 
 
 As Dean Bailey of Cornell once said in a talk to his 
 students: "Many problems solve themselves if left 
 alone. I suggest that each man give himself lessons in 
 the gentle art of keeping cool." When the crucial time 
 comes, the solution can be worked out in the light of 
 actual conditions rather than of probabilities. Business 
 changes too quickly for the solution of to-day to be of 
 much service to-morrow. Styles, the market, labor con- 
 ditions may shift. Indeed, the example of Napoleon, 
 who is said to have filed his letters thirty days ahead 
 before answering them, has its moral for many business 
 men. Not that such a policy could be taken over bodily 
 into business. The illustration is extreme. But it serves 
 to emphasize my point. The business man should not 
 go out to meet problems which are not yet due and may 
 never arise. " The great art is to be governed by time," 
 the Corsican genius declared; " that which ought not 
 to be done until 1810 cannot be done in 1801." The 
 energy of the executive must be conserved for the big 
 problems actually before him, which demand the con- 
 centration of his best efforts. 
 
 ^ Walter Bagehot, " Lombard Street," Chapter VIII. 
 17
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 The trouble with a great many men in business is that 
 they have no definite and ordered way of going at their 
 problems. When a new situation or emergency arises, 
 they take counsel with this man and that, waver from 
 one viewpoint to another, and allow themselves to be 
 influenced in one direction or another by the opinions 
 and snap judgments of advisers who have, perhaps, an 
 entirely erroneous conception of the issue involved. 
 Instead of analyzing the problem and locating all the 
 activities which affect and are affected by its solution, 
 they see it only from what appears to be the principal 
 function involved. Lacking a systematic method of 
 approach, they come to a conclusion which sacrifices 
 more advantages, perhaps, than it conserves. 
 
 The first step in this approach is the elimination — 
 at least, the recognition — of the personal equation. 
 Immediately a new proposition is brought up, this ques- 
 tion almost always intrudes: " How will this affect me 
 personally? " The first reaction is almost sure to be 
 based on personal convenience. Will extra time or 
 extra effort be required.'^ Yet if the business man is to 
 solve his problems scientifically, he must look at them 
 as a scientist looks at his — objectively. This may at 
 first thought seem simple. It is in fact very difficult. 
 Most men cannot detach themselves. Sometimes it re- 
 quires a determined, conscious effort. 
 
 Then the problem must be resolved into its constitu- 
 ent problems, for nearly every problem which arises in 
 business is, as it were, a bundle of distinct minor prob- 
 lems, which together make up the whole. But if the 
 composite problem is analyzed, and the individual units 
 in the bundle are put into their proper places, the broad 
 policy can be seen in its true proportions. Many of the 
 
 18
 
 A METHOD OF APPROACH 
 
 minor problems can be solved, sometimes almost at a 
 glance. And the solution of the big problem will become 
 more and more simple as these individual solutions can- 
 cel off. So it is true that a problem thus analyzed often 
 solves itself. 
 
 Take such a simple case as that of inaugurating a sys- 
 tem of approval shipment. Some of the sub-problems 
 are indicated in the following questions: What will be 
 the effect on collections and on the cost of shipment? 
 What is to be the credit policy ? Will the stock in trans- 
 it or in the hands of customers reduce the number of 
 turnovers per year.'* Will the risk of damage to re- 
 turned goods be great enough to jeopardize the reg- 
 ular profit .f^ Will the increase in sales more thaji 
 offset any added cost in the administration depart- 
 ments.'^ Yet no consideration of this broad problem 
 would be complete without taking into account the 
 psychological factors in selling, such as customers' curi- 
 osity and caution. 
 
 After a problem has been separated into its constitu- 
 ent problems, the factors entering into the solution of 
 each should be assembled, perhaps listed, especially 
 where a large number of varying elements is to be con- 
 sidered. The advantage of listing the factors is clear. 
 It is always simpler to deal with a situation if you have 
 all its elements before you in black and white. Proper 
 weighing and cancellation can be done more accurately 
 and deliberately than if the attempt is made mentally to 
 keep track of the factors and at the same time deter- 
 mine their values. In fact, when all the elements are 
 set down in logical order, it may be possible to reduce 
 them to something approaching a mathematical basis 
 of values. The measure of each factor must be more or 
 
 19
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 less an approximation, but usually each can be expressed 
 in terms of percentages. 
 
 The last step is to take a fresh point of view — to 
 look at the problem as a new problem, one never solved 
 before. It is impossible to overemphasize the impor- 
 tance of the fresh point of view. I do not mean that ex- 
 perience is of no value, but that it is not to be relied 
 upon to the exclusion of experiment and initiative. 
 The value of this open-mindedness is well expressed in a 
 letter from a firm of efficiency engineers to a manufac- 
 turer of national reputation. It is well, they said, " to 
 put your business under the microscope of the produc- 
 tion engineer for these reasons : We analyze your busi- 
 ness from a new viewpoint. We are not swayed by 
 sentiment. We are untrammeled by routine or the 
 customs of your particular line." 
 
 The last point is probably the most important. Busi- 
 ness has long been hampered by tradition. This is to 
 be expected from the nature of its development. Even 
 when a new idea is established in one branch of industry, 
 the man in another almost invariably says: " My busi- 
 ness is different; that idea would not work under our 
 conditions." Between buggies and men's collars, for 
 example, there is no apparent analogy. Yet the man- 
 ager of a vehicle works found in the cutting room of a 
 collar factory a refinement in method which saved him 
 thousands of dollars when applied to the shaping of his 
 leather trimmings. 
 
 His cutters, working on narrow tables, were accus- 
 tomed to " size up " each hide as they unrolled it, then 
 apply their patterns to a section of it and slice out the 
 pieces needed. Five minutes' study of the collar cut- 
 ters led the manager to substitute at home tables wide 
 
 20
 
 A METHOD OF APPROACH 
 
 enough and long enough to accommodate a full hide 
 flat. On this each cutter was taught to shift his pat- 
 terns about until every possible inch of leather was util- 
 ized before touching knife to it. The time required to 
 cut up a hide was doubled and new cutters had to be 
 employed; but more than half the saving of ten thou- 
 sand dollars in costly raw material remained to be added 
 to the company's net profit. All because a manager 
 with an open mind recognized in an alien industry an 
 operation fundamentally identical with one of his own 
 processes and forthwith readjusted his viewpoint on 
 this process. 
 
 Other instances might be cited of the value of a new 
 angle of vision in a business. The rearrangement of an 
 office, suggested by a subordinate, saved daily for his 
 superior half a mile of unnecessary walking. And a 
 casual visitor at a large mail-order house, where it was 
 thought every unnecessary motion had been eliminated, 
 pointed out wasted effort in the order department. In 
 filling out the blanks, the notation " P. O." was used for 
 post-office money orders, " E. O." for express orders, 
 and so on. Since ninety per cent of the 30,000 remit- 
 tances received every day were by post-office order, 
 there was a waste represented by the time used in 27,- 
 000 unnecessary notations. By tagging only the 3,000 
 orders on which remittances were made in some other 
 manner, postal orders could be distinguished without 
 further effort. 
 
 The adoption of a new point of view was the starting- 
 place in the development of the Taylor system of shop 
 management. If scientific management means any- 
 thing, it means an entire divorce from tradition. Old 
 methods of work must be tested, and, if found lacking, 
 
 21
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 must be replaced by a standardized operation deter- 
 mined upon after analysis and time-study of all the 
 motions involved. Scientific management may be noth- 
 ing but applied common sense, as some of its critics 
 have declared. It is important, however, because it is 
 common sense applied in a new way. 
 This systematic method of approach: 
 
 (1) Elimination of the personal equation, 
 
 (2) Separation of the problem into its constituent 
 problems, 
 
 (3) Listing of the factors involved, 
 
 (4) Taking of a new point of view, 
 
 is not intended to be a conscious process. Only when it 
 has become habitual to the user is it most effective. A 
 man should not pause in the process of thought to con- 
 sider the method of his thinking. But when a problem 
 is complex and many factors are involved, the formal 
 use of this method may be of benefit. The assembling 
 in written form of the constituent questions and their 
 chief factors, with the value and importance of each 
 approximated, should at least clarify the situation and 
 increase the chances of a satisfactory solution.
 
 PART I 
 THE PROBLEMS OF PRODUCTION
 
 CHAPTER III 
 
 LOCATION OF THE PLANT 
 
 rriHE activities of production, we have seen, are con- 
 ■*• cerned either with the providing of a ready-to-run 
 factory, its location, construction, and equipment; or 
 with its subsequent operation, the processing of ma- 
 terials through the agency of labor, and the organization 
 of these processes. Before taking up the first group of 
 plant activities, those having to do with location, let me 
 explain that what follows here is to be an illustration of 
 a definite and systematic approach to business problems 
 rather than a complete analysis of the conditions. It 
 would not be in accord with the plan of this book, in- 
 deed, to attempt anything like comprehensive treat- 
 ment of a subject susceptible of so many combinations 
 of varying influences. 
 
 According to one engineer, there are twenty-six fac- 
 tors to be taken into account in selecting a location for 
 a factory. He includes with major items like access to 
 markets, sources of materials, and supplies of labor, 
 indirect factors which contribute to the satisfaction 
 of the working force, like climate and water, schools 
 and churches, housing and local transportation, parks, 
 theatres, and other community attractions. This view 
 may seem extreme, almost finical. Contrasted with 
 the usual method of choosing a location, however, it 
 emphasizes the importance of the first step in the sug- 
 gested method of approach — the elimination, or at 
 
 £5
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 least the recognition, of the personal equation in con- 
 sidering your problem. 
 
 One of the notable large factories of the United States 
 supplies a negative case in point. In plan, arrangement, 
 construction, equipment, and working conditions this 
 plant could hardly be improved upon. It employs sev- 
 eral thousand men and consumes merchant steel and 
 iron, lumber, coal, and other supplies by the train load. 
 For many years, this great bulk of materials had to be 
 trucked nearly two miles from the city freight yards to 
 the factory and the product in turn hauled back to the 
 freight houses for shipment. Recently, after a long 
 campaign for an enabling ordinance, the company se- 
 cured direct railroad connections which allow delivery 
 of its heavier materials in the factory yards. But it is 
 still obliged to maintain a fleet of nearly forty motor 
 trucks to supplement the limited railroad service which 
 its isolated situation permits. 
 
 Going back to the origins of the plant, you find that 
 the personal equation entered largely into its location. 
 Outgrowing his first small shop, the owner — whose 
 commercial foresight in many other respects has been 
 remarkable — permitted sentiment and convenience to 
 guide his selection of a new site. The second factory 
 was erected on his father's farm, remote from the rail- 
 roads and because of its topographical surroundings 
 difficult to reach with a connecting line. Possession of 
 the land was one reason for this choice; but family pride 
 and fondness for the neighborhood were admittedly the 
 compelling motives. Twenty -five years ago, of course, 
 a switch track was not considered essential to a factory ; 
 while the most active imagination might have failed to 
 conceive the need, in time, of forty motor trucks as 
 
 26
 
 LOCATION OF THE PLANT 
 
 traffic attendants on the little new plant in the corn- 
 field. 
 
 Convenience and sentiment still control in the placing 
 of countless industries. In starting a business, a man 
 ordinarily selects his home town for its headquarters. 
 From the viewpoint of finance, his first market and the 
 labor supply, choice of the city in which he is known 
 and commands friendly interest is, perhaps, the part of 
 wisdom. Under certain conditions — limited capital, 
 for instance — any other action might be impossible ; 
 but here the personal equation would not actually in- 
 fluence the decision. Convenience and sentiment, in- 
 deed, have more to do with the location of plants in 
 particular sections of the city, which the owners prefer 
 because they are easy to reach from their residences or 
 because they want to be associated in the public mind 
 with the districts chosen. To indulge such groundless 
 preferences is to handicap the success of the under- 
 taking, just as mistaken loyalty to his native town or 
 a desire to live in a certain city blinds many a capable 
 business man to the advantages of other more strategic 
 centers. 
 
 The surest way to overcome the personal equation is 
 to take the second step in the approach already outlined 
 and resolve the main problem into its constituent prob- 
 lems. Thus the suitability of any factory site breaks 
 up into a number of unit questions : How near and how 
 accessible is the market for your product? What are 
 the transportation facilities and rates, and the chances 
 of alternative service during car shortages.'' What 
 sources of materials are at hand.^^ Under what control.'* 
 What will materials cost, compared with costs at com- 
 peting points? Is there an adequate supply of labor 
 
 27
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 of the kind required at the prices you can pay? Are 
 local conditions such as will help or hinder in the secur- 
 ing, retaining, and handling of the working force? Will 
 any financial advantages accrue from location at this 
 particular point, either in the saving of capital invested 
 or the securing of outside money needed? What will 
 an adequate and satisfactory site cost? 
 
 Will the type of construction best suited to the present 
 and future requirements of the business comply with 
 physical and legal conditions? Will the necessary out- 
 lay for buildings be greater or less than at other avail- 
 able locations? Will the tax rate be lower or higher? 
 Will the situation of the factory add to the fire hazard 
 and the insurance rate? Will lighting conditions be sat- 
 isfactory? The water supply adequate? The drainage 
 and other facilities for disposal of wastes suflBcient? 
 Are the surroundings likely to affect manufacturing 
 processes or the comfort or health of employees at work? 
 Will the plant be within walking distance or at least not 
 diflficult to reach from the homes of employees? Will 
 the location contribute any selling or advertising value 
 to the product? Will it involve any excessive legal 
 limitations on the activities of the business? Unusual 
 circumstances or local conditions may give special im- 
 portance to certain factors in these problems. The 
 drastic provisions of employers' liability laws in New 
 York state, for instance, have trebled the former cost of 
 casualty insurance. Compared with a New Jersey or 
 Connecticut site, the other advantages of which were 
 approximately equal, a New York location at present 
 might be a poor investment. Yet in the long run it 
 might provide a higher type of laboring force. 
 
 The splitting up of the main problem and the listing 
 
 28
 
 LOCATION OF THE PLANT 
 
 of the factors in this way not only prevent the giving of 
 undue weight to any single factor, but also reduce the 
 analysis of competing sites to something like a math- 
 ematical operation, in which the advantages and dis- 
 advantages of each are compared with those of the 
 others. 
 
 The removal of a silverware plant from New York 
 City to a Connecticut factory town several years ago 
 presented a typical problem in location, and the com- 
 pany's method of approach is an excellent illustration 
 of just such an analysis and assembling of the signifi- 
 cant factors as are indicated above. 
 
 As in so many successful industries, the original fac- 
 tory had been outgrown. Looking about for a new lo- 
 cation, the directors found that all available New York 
 sites had one drawback in common. The high cost of 
 land would make it necessary to limit the ground floor 
 area of the new building to the space requirements of 
 the power plant, the shipping rooms, and the general 
 offices. Three additional floors would be needed for 
 current manufacturing operations; further expansion of 
 the business must be taken care of by superimposing 
 other stories. This would mean the probable division 
 of one important department between two floors and 
 would certainly not conduce to economical production. 
 Elevator service for the transfer of stock in process and 
 workmen from floor to floor would increase overhead 
 expense; inquiry showed that in some cases the cost of 
 such service, including repairs, power, operators, and 
 the time of foremen and mechanics using it amounted 
 to about two per cent of the total pay roll. 
 
 Outside New York, in any one of several industrial 
 towns, the same investment would secure a site large 
 
 29
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 enough to permit all heavy manufacturing processes to 
 be concentrated on the ground floor, in the most effec- 
 tive sequence and with the minimum of trucking for 
 stock and of movement for employees. Construction, 
 too, could be of a much cheaper and more flexible type 
 in a two-story detached factory in the country than in a 
 four, six, or eight story structure wedged in between 
 buildings in a city district where the fire hazard was 
 great, insurance rates correspondingly high, and the 
 city building ordinances most exacting. Cheapness and 
 flexibility in construction were vital considerations, for 
 the reason that invention of new machines or new proc- 
 esses might, at any time, necessitate the complete re- 
 arrangement of one or many departments. Safety and 
 a low insurance rate were matters of account; so also 
 were daylight illumination of workrooms and freedom 
 from dust and soot. On all these items, the small town 
 site was greatly to be preferred. 
 
 Certain advantages were inherent, however, in a 
 metropolitan location. New York offered the best mar- 
 ket for raw materials and supplies. It was the sales 
 center of the silverware trade; during the seasons when 
 customers came in to buy, quick and easy access to the 
 plant and the general offices had a definite selling in- 
 fluence on visitors. Direct trucking communication 
 between factory and salesrooms allowed of closer co- 
 operation in filling orders and considerable economies 
 in the handling and refinishing of goods. As a shipping 
 center, also. New York was beyond comparison. Ninety 
 per cent of the company's forwarding was done by ex- 
 press. New York offered a choice of several carriers and 
 direct, immediate dispatch in the case of rush orders; 
 while service in the smaller town might be limited to a 
 
 30
 
 LOCATION OF THE PLANT 
 
 single express company and be delayed by the necessity 
 of transfer at New York. Still, the almost prohibitive 
 cost of a city site or the alternative drawbacks in con- 
 struction, arrangement, and operation determined the 
 directors to canvass the small towns thoroughly before 
 settling the question. 
 
 The problem finally narrowed down to choice between 
 a New York location and one in a particular industrial 
 center in Connecticut. Half a dozen other cities in 
 Connecticut, New York, and New Jersey — all wuthin 
 a radius of fifty miles — were considered. Sites at a 
 greater distance were eliminated because nothing they 
 had to offer would compensate for the added difficulty 
 in reaching them. Customer-visits to the plants, it was 
 decided, could not be jeopardized by making the jour- 
 ney long and tiresome. For analogous reasons, various 
 smaller towns were dismissed because hotel accommo- 
 dations were lacking and train service, both for passen- 
 gers and express shipments, not so good. In addition 
 to the railroad facilities of the town finally selected, 
 three steamboat lines to New York City provided cer- 
 tain assurance against delays and made a complete tie- 
 up of communication practically impossible. This last 
 consideration illustrates the interdependence of produc- 
 tion problems and problems of distribution (already 
 discussed in the first chapter) and the necessity of fore- 
 seeing the influence of each upon the other. 
 
 Besides the typical small-city advantages as an in- 
 dustrial site, the chosen town possessed other attractive 
 features. The prices of gas and water were below the 
 average. There would be a substantial saving in taxes 
 for two reasons, the rate was lower and the real estate 
 investment would be much less than in New York. 
 
 31
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Power could be produced more cheaply than in New 
 York. The water freight on steam coal from the West 
 Virginia fields was nine cents higher than to New York, 
 but this excess would be more than made up by the 
 saving in trucking charges from dock side to boiler 
 room and in disposal of the ashes. The saving, in fact, 
 would amount to nearly seventy -five cents a ton on all 
 coal burned, in itself a considerable economy. 
 
 Conditions were favorable also to the easy handling 
 of labor. The company wanted to carry with it to the 
 new factory all its skilled men and many of its unskilled. 
 City-bred mechanics, however, frequently have a dis- 
 taste for life in a small town. The employees of one 
 large manufacturer recently voted against removal of 
 the plant from a city noted for its parks, beaches, and 
 public institutions, insisting that they could get more 
 pleasure and relaxation for nothing where they were 
 than they could buy with a week's wages in the other 
 town. The silverware company, on the contrary, was 
 able to present its new site as a better place in which to 
 live and bring up families. Its nearness to New York 
 would enable them quickly and cheaply to run into the 
 city whenever they desired, thus maintaining such con- 
 nections and associations as they valued. The available 
 sites for the new plant were within walking distance of 
 districts where desirable homes could be bought or 
 rented at figures much below the New York levels for 
 corresponding environments. The saving in time spent 
 in crowded street cars would be at least an hour daily 
 for each workman, to say nothing of the carfare. As 
 for schools, churches, parks, bathing beaches, and other 
 accessible community utilities, New York offered noth- 
 ing more. Even working conditions would be bettered 
 
 32,
 
 LOCATION OF THE PLANT 
 
 by the removal to a place where daylight and fresh air 
 were abundant. 
 
 Thus, item by item, the company assembled all the 
 factors bearing on its location problems. Certain of 
 these, of possible prime importance to a young business 
 with limited resources, could be disregarded. For ex- 
 ample, no question of capital or finance had to be con- 
 sidered. The company was able to pay any reasonable 
 sum for a site, could erect and equip a factory without 
 outside aid. Banking facilities were a minor item, since 
 New York connections could be maintained. Sources 
 and prices of raw materials would not be affected. And 
 so with other factors not analyzed above and ranged as 
 favoring or discouraging removal, they could practically 
 be ignored. And having brought together all the ele- 
 ments that counted, 'pro or con, and balanced them in 
 accord with their relative values, the Connecticut town 
 was decided upon as the place for its new factory. 
 
 In coming to this decision, the company consciously 
 or unconsciously illustrated the final step in the method 
 I have suggested for approaching business problems — 
 the taking of a fresh point of view. It has been made 
 clear, I think, that they attacked the question of loca- 
 tion as a new problem, surveying all the conditions and 
 circumstances without bias of any sort, either personal 
 prejudice or regard for trade customs or traditions. It 
 is the usual thing, for example, to locate new mills in 
 districts where successful mills of the same class or 
 character are in operation. In a particular case, it is 
 quite possible that this customary procedure is the right 
 one to follow; again there may be found reasons for an 
 opposite course. These reasons, however, can be de- 
 veloped only by assuming a viewpoint independent of 
 
 33
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 traditions. As in the case of this silverware company, 
 all the conditions which make for failure or success 
 should be observed and carefully analyzed from the 
 standpoint of a new enterprise. Nothing should be 
 taken for granted. 
 
 Certain general advantages, to be sure, reside in a 
 location where other factories in the same line or al- 
 lied lines are concentrated — as shoe making in Lynn, 
 Brockton, and St. Louis; flour milling at Minneapolis 
 and St. Paul, and cotton manufacturing at Fall River 
 and New Bedford. In such centers, each individual 
 industry, because of the importance of its group, com- 
 mands resources and service, and profits by various 
 external economies which the isolated establishment 
 is likely to miss. At Detroit, for instance, where the 
 production of motor cars is carried on extensively, sub- 
 sidiary factories have multiplied and now supply eco- 
 nomically and without delay the special machinery and 
 tools, the specialized parts, materials, accessories, and 
 supplies which the parent industry requires. The mag- 
 nitude of the market open for these specialized prod- 
 ucts tempts initiative and ingenuity to undertake detail 
 manufacturing tasks which the maker of motor cars 
 prefers not to assume himself, and thus brings about a 
 cooperative effort profitable to everybody. 
 
 For like reasons, purchasing conditions are more 
 favorable at the focus of an industry; prices are closer, 
 quality of deliveries is more carefully watched, and de- 
 liveries themselves are more apt to be on schedule. 
 Bankers, too, have studied underlying conditions more 
 intensively and the sound enterprise is more easily 
 financed. Public opinion is usually more favorable to 
 the basic community industry in which a large number 
 
 34
 
 LOCATION OF THE PLANT 
 
 of citizens are interested. And finally, where many- 
 minds are directed upon the same technical problems 
 and there is interchange of information, both privately 
 and through trade association channels, advances in 
 production practice are more frequent and more quickly 
 shared. 
 
 Where the factories gather, also, labor congregates. 
 Mechanics of the highest skill or specialized training 
 avoid the isolated plant for definite reasons. If a dull 
 selling season brings about a shut-down or half-time pro- 
 duction, the chance of employment elsewhere is small; 
 while a disagreement with a foreman nearly always 
 means removal to some other city. So general is this 
 feeling against the factory at a distance from the greater 
 labor markets, that even the most stable of businesses, 
 when situated thus, find it hard to induce skilled labor 
 to join their forces. One specialty company in Ohio, 
 for example, when redesign of its machine tools was 
 determined upon and twenty additional tool-makers 
 were needed, paid an average of $180 each to start the 
 new men at work. Most of these had families, and 
 their moving expenses were paid on the theory that the 
 families would help to anchor them to their new jobs. 
 In line with this belief is the experience of a plant su- 
 perintendent in central Illinois who is content if he 
 can keep unmarried operators of screw machines three 
 months before they make their next jump to St. Louis 
 or Chicago. In an isolated plant like this, the loss of 
 a few skilled workmen in a processing department might 
 cripple production until they could be replaced. 
 
 Unskilled labor — no less a necessity of large-scale 
 production — has a similar tendency to concentrate in 
 the big communities, particularly since so many of our 
 
 3d
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 pick-and-shovel men have been recruited from southern 
 and eastern Europe. As a result of this common drift 
 on the part of both skilled and unskilled, our cities have 
 become huge labor markets from which any number of 
 extra workers of well nigh any class can be drawn when 
 needed. So long as an industry is given to decided fluc- 
 tuations in sales and output, it needs such a dependable 
 labor pool. Hence we witness the building up of manu- 
 facturing districts like that surrounding the Baldwin 
 locomotive works in Philadelphia, whose scores of ma- 
 chinery houses have taken care of business expansion by 
 adding upper stories to their plants, and sacrificed a 
 certain amount of everyday efficiency in order to insure 
 a flexible labor supply. 
 
 These are all general advantages, however, and should 
 not be emphasized to the exclusion of possible special 
 benefits which, in a particular case, might balance or 
 outweigh them. Nearness to supplies of raw materials 
 and cheap labor, for instance, have made it possible 
 for southern cotton mills to compete favorably with 
 New England establishments, especially on the coarser 
 fabrics. This, despite New England's better transpor- 
 tation, lower freight rates, cheaper power, highly organ- 
 ized markets, and notable concentration of the industry, 
 with its attendant advantages. It must also be taken 
 into account that in mdustrial centers, skilled and com- 
 mon labor are both likely to carry organized effort to 
 limit production and enforce relatively high wage scales 
 much further than in communities where class con- 
 sciousness has not been fostered. 
 
 For the sake of showing how definite, almost arith- 
 metical, the approach to a business problem can be 
 made, the steps in the location of a drain tile factory in 
 
 36
 
 LOCATION OF THE PLANT 
 
 the Middle West some years ago may be recounted here. 
 The owners made nearness to market their first consid- 
 eration, though the other salient factors were not over- 
 looked. Their product would be bulky and of low value 
 relative to weight; the plant with the shortest haul 
 would enjoy a real selling advantage in its lower freight 
 rates. The plant was located, therefore, near the center 
 of a district where land levels were low and drain tile in 
 constant demand ; the extra cost and added risk in proc- 
 essing the inferior raw materials found there were ac- 
 cepted for the sake of the advantage in transportation. 
 As the customer paid the freight, the saving appealed 
 directly to his pocket. The new company did not need 
 to cut prices and thus provoke disparagement of its 
 product, while the difference in rates emphasized its 
 character as a " home " industry. 
 
 There was also a plentiful supply of native labor, with 
 sufficient intelligence to master quickly all but the most 
 technical processes of manufacture. \Mien the balance 
 was struck, the new location scored on four principal 
 counts — transportation, two points; labor, one point; 
 finance, one point (a free site was offered and additional 
 deposits of raw materials could be purchased cheaply) ; 
 advertising and selling, one half point. Against these, 
 three disadvantages appeared: banking facilities were 
 limited, subtracting one point; material was inferior, 
 one half point; and sources of fuel were distant, increas- 
 ing the cost and deducting another half point. The net 
 result was a count of four and one half points to two in 
 favor of the central situation. 
 
 The problem of plant location, however, does not stop 
 with the selection of a particular city. The actual plac- 
 ing of the factory to satisfy and make the most of local 
 
 37 
 
 / :^ * i H
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and internal conditions is a matter which can be settled 
 only by the same process of careful analysis and weigh- 
 ing of essential factors. Where in the general district 
 (the traffic zone surrounding a large city, for instance, 
 its boundaries marked by the area to which the city 
 freight tariffs apply) is the plant to be built? What 
 section of the chosen town is most suitable? 
 
 Here enter many of the factors which we found impor- 
 tant in determining the main problem: transportation 
 (direct railroad connections), power (no neighborhood 
 restrictions on its production), labor (accessibility from 
 residence districts and freedom from objectionable 
 conditions), advertising and selling values, and so on. 
 The physical characteristics of competing sites must 
 also be examined. Can suitable foundations, for ex- 
 ample, be secured at average cost? What is the slope 
 of the ground? Can this slope be used in planning the 
 layout of the buildings? How does it affect drainage or 
 the water supply? Will the excavation furnish mate- 
 rials for construction, as gravel or sand in a building of 
 reinforced concrete? 
 
 In the local placing of the drain tile plant just men- 
 tioned, transportation again governed. The shale beds 
 and free site lay some distance from the village, along- 
 side one railroad but a mile and a half from its junction 
 with another road. To build here would require no 
 investment in a site, while the material could be de- 
 livered to the machines at minimum cost. The draw- 
 back would be the lack of connections with the other 
 railroad. Many shipments which might secure com- 
 petitive service and rates if both carriers were accessible 
 would get only routine attention at the maximum rate. 
 Every car shortage might result in serious delays and 
 
 38
 
 LOCATION OF THE PLANT 
 
 possible cancellation of contracts. To insure certainty 
 in deliveries, therefore, the management paid fifteen 
 hundred dollars for three acres at the junction and 
 erected its plant there, preferring to pay freight on its 
 material rather than save at the possible expense of sales. 
 Some weight was attached also to proximity to the vil- 
 lage; workmen would have a mile less to walk mornings 
 and evenings and could go home to their noonday din- 
 ners. The decisive factor, however, was transportation 
 facilities and service in this instance, just as the lower 
 cost of shorter hauls had been the determining cause in 
 locating the plant in its geographical relation to the 
 market. 
 
 At the risk of appearing prolix, analysis of the prob- 
 lem of location has been carried out at some length in 
 this chapter, first in order to indicate with quantities 
 having standard and recognized values the scheme of 
 analysis which later will be applied to the less familiar 
 activities of distribution and administration, and also to 
 illustrate how a systematic approach will facilitate the 
 determination of any business question. In succeeding 
 chapters, the effort will be made to sketch rather than to 
 develop the outlines of each problem and the factors 
 affecting its solution.
 
 CHAPTER IV 
 CONSTRUCTION AND EQUIPMENT OF THE PLANT 
 
 IN the construction of a factory the main considera- 
 tion, of course, is its efficiency as a producing unit. 
 Size, type, shape, arrangement, cost, number of stories, 
 and other physical characteristics are essential factors 
 in achieving this result. The determination of each, 
 therefore, is a problem to be solved by the unbiased and 
 systematic approach illustrated in the preceding chap- 
 ter on location. In their broad aspects, indeed, location 
 and construction, as well as equipment, are parallel and 
 interdependent problems. 
 
 The best site for a plant (within the district which sat- 
 isfies material, labor, transportation, and other primary 
 requirements) is that tract of land which lends itself to 
 the erection of the particular kind of building or group of 
 buildings in which the processes of manufacture can be 
 carried on most effectively. Conversely, any plant falls 
 short of its highest efficiency unless, in its planning and 
 building, advantage has been taken of every favorable 
 feature of the site. And neither the location nor the 
 construction can be intelligently or profitably considered 
 apart from the equipment which is to be accommodated. 
 In general, it would be better, from the standpoint of 
 efficiency, to plan and erect the building around the 
 equipment than to arrange the machinery after the 
 building was completed. 
 
 Preliminary to the analysis and balancing of fac- 
 
 40
 
 CONSTRUCTION AND EQUIPMENT 
 
 tors, merely personal considerations must again be dis- 
 counted. The owner's preference for a special type of 
 building or a certain kind of construction must be elim- 
 inated unless the preference rests on practical engineer- 
 ing or commercial grounds. He may fancy his business 
 housed in one of the imposing ivy-covered buildings 
 which are pictured at intervals in the magazines; but 
 if his product can be turned out more economically in a 
 sprawling, one-story plant with a saw-tooth roof, he will 
 do well to dismiss his day dream. The selection, corre- 
 lation, and placing of the equipment are the things that 
 count in starting a factory; and the plan, appearance, 
 and construction of the building should all be subor- 
 dinated to the eflSciency of the machines. 
 
 This does not mean that the structure should be an 
 eyesore; a pleasing exterior or unusual interior finish 
 may possess considerable publicity value for a product 
 with quality of cleanliness as a sales argument, to say 
 nothing of the effect of clean and cheerful surroundings 
 on the working force. The point to be observed in this 
 connection is that eflficiency of operation should not be 
 sacrificed to satisfy a personal whim or prejudice. 
 
 When we come to break up the problem of design and 
 construction into its constituent problems, we find our- 
 selves concerned with three primary questions: What 
 type shall our building be — single-story or multi-story, 
 oblong, hollow square, or some other standard shape.'* 
 What character of construction shall be adopted — fire- 
 resisting or slow-burning, brick and steel, concrete or 
 mill construction? What financial considerations are 
 involved — initial cost and upkeep? As in every re- 
 lated group of business problems, we find the principle 
 of interdependence in control, making it impossible to 
 
 41
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 find the right answer for anyone without taking the 
 influence of the others into consideration. 
 
 The type of building to be erected (or purchased, as 
 it is frequently profitable to do) should be dictated first 
 by the character and sequence of the manufacturing 
 processes to be sheltered. Where the parts have great 
 weight and bulk, as in a plant producing heavy-duty 
 steam engines or mining machinery, the single-story 
 building appears to be the most practical type. Where 
 the raw materials can be elevated in the first place to 
 the top floor and the successive operations applied to 
 them as they move onward and downward, as in the 
 making of men's hats, the building of several stories 
 suggests itself as more economical. In such a plant a 
 gravity-conveyor system is often used for the transpor- 
 tation of materials in process. 
 
 Where the product is an assembly of many relatively 
 small parts, such as a sewing machine, an electric fan, or 
 a motor car, the type of building is of less importance if 
 the routing of materials can be arranged so that all the 
 processing departments feed without lost motion or 
 interference toward a central assembling room or stock 
 room. A simple rectangular building, of one story or 
 many stories, may prove a satisfactory type. Or when 
 several classes of raw materials, such as lumber, pig 
 iron, sheet steel, and brass enter into the make-up of the 
 final product, a building with projecting wings to allow 
 separate but synchronous processing of these materials 
 may be chosen. The same type is likely to prove best 
 for a business making a line of goods diverse in nature. 
 Or these wings may be developed into unit buildings to 
 accommodate wood-working, foundry, machining, and 
 other special operations. In factories of any size, in- 
 
 4S
 
 CONSTRUCTION AND EQUIPMENT 
 
 deed, the best practice gives the foundry and wood- 
 working departments buildings of their own for much 
 the same reasons that make the power house a separate 
 unit. 
 
 Apart from the relative cost of single-story and mul- 
 tiple-story buildings, transportation, arrangement of 
 equipment, and providing of sunlight illumination for 
 the greatest number of operations possible are the chief 
 considerations in determining between the types, except 
 where the heavy character of the parts produced dic- 
 tates choice of the first. Here transportation is clearly 
 the prohibiting cause; even in factories without this 
 limitation, the high cost of handling parts from floor to 
 floor as compared with horizontal trucking is an impres- 
 sive item. One production engineer, after time studies 
 of the trucking and elevator service in a table factory, 
 discovered that the horizontal time equivalent of an 
 elevator trip to the second story and return was one 
 hundred feet, with an additional fifty feet on the level 
 for every extra story. 
 
 For daylight illumination, the saw-tooth roof gives 
 the one-story building indefinite powers of expansion, 
 subject only to the maximum floor spaces allowed in one 
 inclosure by the rules of the insurance underwriters. In 
 any multi-story structure, on the contrary, the effective 
 widths range from fifty to seventy feet, dependent on 
 the height of the ceilings and the proportional area of 
 windows. There are thousands of factories, it is true, 
 which ignore these standards; and other thousands 
 which accept artificial light as a necessary accompani- 
 ment of production, even in country situations. But 
 it is equally true that the most efficient modern plants 
 conform to them and rank perfect daylight illumination 
 
 43
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 of work practically as inflexible a rule in construction as 
 solid foundations or a non-leaking roof. Some of them 
 concentrate on the top floors the operations involving 
 very small parts, even providing saw-tooth roofs with 
 northern exposure as insurance against eye strain for 
 workers and spoiled parts for the scrap bins. 
 
 There is another motive for adopting that standard 
 type and size which satisfies your production require- 
 ments. Not only is such a building less costly to erect, 
 because it utilizes materials of standard market specifi- 
 cations and because contractors are familiar with such 
 construction, but it will also be more salable in an emer- 
 gency or in the event that removal proves desirable. 
 
 For the same reason, the more closely its layout con- 
 forms to standard practice, the simpler will be the task 
 of selling it. To be easily disposed of, a plant of any 
 size must have a railroad siding at its shipping platform. 
 Even though his business is of a kind that has little use 
 for direct rail connections, a prospective buyer will not 
 offer so much for a building without a rail outlet. There 
 are many exceptions to this rule, of course, and any 
 number of cases where a switch track would add to the 
 cost of the site more than its use would justify. The 
 damage a millinery house would suffer in its stock, for 
 instance, would wipe out any saving in trucking its 
 light-weight materials. Any light manufacturing busi- 
 ness, unless on an unusual scale, could well avoid the 
 extra investment a switch track would demand and use 
 the money for developing a wider market. 
 
 This factor of salability should also be considered in 
 the problem of design and construction. It should not 
 be the major consideration, but rather the corrective 
 measure which brought up or down to recognized stand- 
 
 44
 
 CONSTRUCTION AND EQUIPMENT 
 
 ards the specifications which might otherwise result in 
 a building costly beyond any reasonable need, or too 
 light in design for any use except that for which it was 
 erected. Both are extremes to be avoided in fireproof or 
 mill construction; when the structure is of concrete, 
 however, the added value of floors capable of sustaining 
 maximum loads is out of proportion to the slight extra 
 cost. Here, bringing construction up to recognized levels 
 of strength also insures the future of the business. Even 
 in light manufacturing, methods and processes are con- 
 stantly changing. To keep pace with possible develop- 
 ments in machinery, the factory may require all the 
 margin of safety its floors possess. 
 
 More important factors in determining the character 
 of construction are the legal requirements of the com- 
 munity in which the plant is located (particularly as 
 these bear on the degree of fire resistance demanded) 
 and the classes of material available, with the relative 
 market price and the comparative cost of handling each 
 kind. Financial considerations must be taken into 
 account as well, since the initial cost and the upkeep of 
 various classes of buildings vary in inverse ratio. 
 
 For the business with large resources and long experi- 
 ence, the problem narrows down to choice between 
 different kinds of substantial firepoof or slow-burning 
 construction. This is not alone to keep down insurance 
 premiums and repairs, but also to provide against pos- 
 sible destruction of factory and machinery at the height 
 of a selling season. The owner or manager of a young 
 business, on the contrary, usually has a minimum cap- 
 ital to work with and must figure plant investment 
 down to the lowest point. Frequently he starts in rented 
 quarters none too well adapted to his purposes and buys 
 
 45
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 his power from an outside source at a higher price to 
 avoid the installation of a power plant. In the main, 
 however, these problems of design, construction, initial 
 cost, and upkeep are technical questions which must be 
 settled in individual cases after consultation with a fac- 
 tory engineer or architect and, perhaps, with a building 
 contractor. There are too many possible combina- 
 tions of conditions to be analyzed or even enumerated 
 here. 
 
 The technical expert should have a part, too, in the 
 next step, the assembling and weighing of all the factors 
 that count in construction of this individual factory. 
 The tendency of the architect or engineer to lay stress 
 on the technical aspects of building must be corrected 
 by the practical viewpoint of the owner or manager. 
 An enthusiasm for some special type of construction, 
 for instance, may cause the architect to overlook the 
 peculiar advantages of another and contrasting type 
 for the particular industry under the special set of con- 
 ditions which may exist. Here the experience and the 
 broader perspective of the manager come in to reduce 
 each factor, technical and practical, to its relative pro- 
 portions and to strike a balance between these opposing 
 but interdependent elements. ^^ 
 
 A good architect and a competent production engineer 
 are quite capable of providing an ideal factory and an 
 ideal layout for almost any given industry. But a going 
 business nearly always has an individuality, organiza- 
 tion habits, and special demands which must be consid- 
 ered if the new building is to fit its needs and hamper 
 none of its activities. The owner or executive, there- 
 fore, can depend upon his technical advisers or aids for 
 the suggestion and carrying out of technical details; but 
 
 46
 
 CONSTRUCTION AND EQUIPMENT 
 
 the relative weight to be given each of the general fac- 
 tors must be determined primarily by the management. 
 
 Here the fresh point of view has great value. The 
 fact that certain types of buildings and certain kinds 
 of construction have become standards in your district 
 or your line of industry is not a guarantee that either is 
 the one best suited to your purpose. Conformity to 
 this standard will make for salability, but may at the 
 same time seriously affect the efficiency of production. 
 
 The furniture factories at Grand Rapids, for example, 
 usually have been of mill construction, with walls of 
 white brick and only the usual quota of windows for 
 lighting. A seasoned furniture man, if he were erecting 
 a new building, might incline to this traditional type, 
 but a man whose experience had been in a machinery 
 or automobile plant, if he entered the furniture field, 
 would be likely to build a factory with windows as wide 
 and high as his walls would permit, using the steel or 
 concrete construction with which he was familiar to 
 balance the extra fire hazard caused by his raw ma- 
 terials. The furniture man, before he decided what 
 type his new building should be, probably would profit 
 by advising with construction specialists and visiting 
 modern machinery and motor-car factories. 
 
 The importance of the fresh point of view in construc- 
 tion is suggested by the experience of a New England 
 weaving mill which replaced a three-story building hav- 
 ing a width of seventy feet with a one-story structure 
 one hundred and five feet wide and covered with a saw- 
 tooth roof. The improved natural lighting and increased 
 facility in handling materials and product brought about 
 the remarkable reduction of one half in the cost of weav- 
 ing and two thirds in the cost of artificial light. 
 
 47
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 The foundry of a Detroit stove works affords another 
 striking illustration. Indifferent lighting and foul air 
 are usually accepted as necessities of foundry operation; 
 working conditions are generally worse than in any other 
 processing department. This stove company, however, 
 determined to erect a foundry which would eliminate 
 discomfort and disorder. Except for a six-foot base of 
 brick, the walls are entirely of steel sash and glass. By 
 opening tilting sections of these walls, it is possible to 
 provide satisfactory ventilation in summer and to clear 
 the building of vitiated air within fifteen minutes after 
 each pouring of metal. In contrast with this departure, 
 which allows resumption of work after a short delay, 
 many foundries plan only two pourings a day and turn 
 their men out when these have been completed, with 
 mutual loss of time and output. As foundry capacity 
 in a stove works sets a limit for the whole plant, the 
 value of the fresh viewpoint is indicated by the effect 
 on production. 
 
 Like all other business problems, the problems of con- 
 struction cannot be solved without taking into account 
 the principle of interdependence and the principle of 
 balance. The geographical location of the plant and 
 the actual site have much to do with choice of building 
 materials and the kind of construction adopted. The 
 type of building, on the other hand, cannot be intelli- 
 gently determined without keeping always in view the 
 character of the manufacturing processes, the space, 
 power, and safety requirements of the equipment and 
 materials, and the physical necessities or desires of the 
 working forces. And again, the size, type, and class of 
 factory erected or leased may rightly be subordinated 
 to considerations which have to do solely with sales or 
 
 48
 
 CONSTRUCTION AND EQUIPMENT 
 
 with finance and other administrative activities. No 
 business problem is a simple equation to be settled off- 
 hand without regard for other activities which may be 
 advanced or jeopardized by the decision. 
 
 Closely bound up with construction, as we have al- 
 ready seen, is the problem of equipment. Having deter- 
 mined what type, size, and shape your building shall be 
 and what class of construction is best adapted to the 
 necessities and possibilities of your business, the broad 
 outlines of your machinery scheme are already clear, 
 since the general sequence of your manufacturing opera- 
 tions was a factor in your construction planning. For 
 this reason and for the additional reason that knowl- 
 edge of the development and standardization of equip- 
 ment is so general, it will be enough here to make the 
 briefest possible application of our method of approach 
 to the problem. 
 
 This fact of standardization gives, in the first place, 
 very little play to the personal factor. Assuming that 
 the owner or manager has a reasonable acquaintance 
 with current practice in his field and a competent su- 
 perintendent or production engineer to advise him, his 
 choice of basic equipment will probably lie between 
 machines designed by rival manufacturers to achieve 
 approximately the same result. Patent control of spe- 
 cial machinery held by his competitors may limit this 
 choice and challenge his initiative and ingenuity to find 
 other ejB&cient and economical methods of manufactur- 
 ing. It is only in his decision on such general questions 
 as that of the type of drive he will use (whether he will 
 have individual electric motors or a belt-driven plant), 
 and in his choice of auxiliary equipment for heating, 
 ventilation, sanitation, and lighting (still in the experi- 
 
 49
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 mental stage, though of extreme importance because 
 of the legislation which is being brought to bear upon 
 them), that the personal prejudices or prepossessions of 
 the owner, if arbitrarily indulged, may run him into 
 mechanical difficulties or saddle the business with dis- 
 proportionate expense. 
 
 In breaking up the problem of equipment into its 
 constituent problems, the concrete character of the 
 various factors helps to define them for the manage- 
 ment. Financial considerations come first, initial in- 
 vestment needing to be balanced against upkeep costs, 
 though both are subordinate, of course, to the prime 
 question of how much money can be devoted to the pur- 
 chase of machinery. Next in order are the quantity and 
 quality of output required, balanced against the kind 
 and class of raw materials to be used; then safety against 
 both fire and accident; finally, the matter of design. 
 This sub-problem of design will be influenced by all the 
 factors just named and by the additional factor of the 
 capacity of the labor available. 
 
 In the end, within the limits set by capital, output, 
 and labor considerations, the manager's problem will 
 narrow down to picking out the particular model of 
 punch press or turret lathe, milling machine or drill 
 press, which will handle the materials he must use or the 
 parts he must process to the best advantage. This is 
 only an outline suggestion of the manner in which one 
 equipment problem might shape itself; it does not offer 
 an ironclad analysis generally applicable. In some 
 cases, one or more of the factors listed above might 
 settle the question out of hand and make the others 
 negligible; in other instances new factors might arise 
 and dominate. But in every case analysis of the main 
 
 50
 
 CONSTRUCTION AND EQUIPMENT 
 
 problem into its constituent parts will go far toward 
 suggesting the right solution. 
 
 In listing and weighing the factors in equipment, the 
 advice of the production specialist is valuable. The 
 owner or executive, unless he happens to be factory- 
 trained, is not necessarily familiar with technical de- 
 tails, like the most effective speed of screw machines on 
 various classes of work or the formulae of power trans- 
 mission. For such information he should call on his 
 superintendent or production engineer. His own func- 
 tion is to assemble and balance all the factors bearing on 
 the general problem, including those furnished by his 
 expert counsel, and to determine the relative weight 
 each is to have in the final decision. 
 
 An open mind, a fresh point of view, is nowhere 
 more essential than in determining equipment policies. 
 Changes and improvements are of almost daily occur- 
 rence; usage and tradition, therefore, must never be 
 allowed to block the installation of the most efficient 
 types of machinery. It is generally believed, for exam- 
 ple, that the economies brought about in other indus- 
 tries by the use of automatic machinery are not equally 
 profitable in the wood-working groups. As a matter of 
 fact, automatic machines have been successfully em- 
 ployed in a number of wood-working operations. Their 
 further development and application will depend largely 
 on the initiative and open-mindedness of the men en- 
 gaged in the industry, or on the invasion, perhaps, of 
 the wood-working field by men from the metal-working 
 groups who think about any manufacturing problem in 
 terms of automatic machinery. 
 
 This completes our outline and analysis of the plant 
 activities — the policies which govern the location, con- 
 
 51
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 struction, and equipment of the factory. As the purpose 
 was chiefly to illustrate the application of a systematic 
 method of approach to any specific set of business prob- 
 lems, no attempt has been made at an exhaustive sur- 
 vey. It was enough to indicate the importance of a 
 definite method of approach and to make clear the steps 
 of that approach in anticipation of its use later in the 
 analysis of problems whose factors are less tangible. 
 
 52
 
 CHAPTER V 
 MATERIALS 
 
 OVER against the plant activities of production — 
 the placing, building, and equipping of the fac- 
 tory — is balanced the group of activities concerned 
 with operation. Their importance hinges on the fact 
 that when the wheels begin to turn the business ceases 
 to concentrate on itself and deals with factors which 
 in many instances are difficult to control, unless control 
 has been considered and provided for in the location and 
 equipment of the works. These outside relations have 
 to do first with materials, their kind, quality, and the 
 freedom with which they can be secured; and second 
 with labor, the extent and character of the market for 
 men. Together with the organization of manufacturing 
 processes, these make up the operating activities of 
 production. As another demonstration of the principle 
 of interdependence which runs through the whole struc- 
 ture of business, their essential relations can be briefly 
 and graphically illustrated in the experience of a group 
 of business men in a small city in middle Illinois. 
 
 The first adventure in production made by this group 
 was the manufacture of barbed- wire fencing in the later 
 nineties, before the regulation of big business had been 
 undertaken by the federal government. Standard- 
 gauge wire, the material used by the factory, was bought 
 from the nearest mill of a big rival, which presently 
 adopted a policy of purchase and consolidation of 
 
 53
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 smaller competitors whenever these became annoying 
 factors in sales. After running three or four years with 
 satisfactory returns, the smaller plant suddenly was con- 
 fronted with advances in the price of wire which wiped 
 out its margin of profit. The manager had foreseen such 
 a possibility and had urged the stockholders to appro- 
 priate for a wire-drawing mill. They had refused, how- 
 ever, and when the last advance in materials destroyed 
 the profit margin, there was no time to erect and equip 
 a wire department. Surrender on fairly generous terms 
 was the outcome; the stockholders banked their money 
 and watched the removal of the equipment to another 
 plant. The factory itself was offered for sale. 
 
 The manager had learned a drastic lesson in produc- 
 tion policy — that you must make sure of the source 
 and price of your raw materials. When the creative in- 
 stinct stirred in him again, therefore, his systematic ap- 
 proach to his new venture differed only in form from 
 that which has been set forth in previous chapters — 
 recognition of the personal equation, analysis of the 
 general problem, the listing of the factors, and the tak- 
 ing of a fresh point of view. 
 
 He concentrated on materials, on the building of an 
 industry which could riot be crippled by failing supplies, 
 since his location, construction, and labor problems 
 were already solved if only he could bring his material 
 and equipment problems into accord with them. A 
 factory and power house could be had cheaply, with 
 outlets on three railroads and advantageous freight 
 rates to several merchandising centers. Labor, after a 
 long season of unemployment, could be hired reason- 
 ably and could be depended upon for energetic and in- 
 telligent cooperation. The men were machine-tenders, 
 
 54
 
 MATERIALS 
 
 ** handy men," who could not easily be developed into 
 mechanics. Their capacities had to be considered in 
 deciding what to make and how to make it. Repeat 
 operations by semi-automatic machinery on some un- 
 controlled raw material were the essentials of success 
 in production. 
 
 Three raw materials stood out as possibilities. Plas- 
 tic shales were abundant locally for the manufacture of 
 sewer pipe, fire brick, and fireproofing ware. Merchant 
 steel was offered by independent sources at fair prices. 
 Lumber and timber were to be had cheaply and in any 
 quantity, particularly soft woods from the great mar- 
 kets on the Mississippi River. 
 
 Checking the advantages and disadvantages of each 
 material against the others, the manager chose to enter 
 wood-working. There was a local clay industry whose 
 moderate dividends would stand in the way of interest- 
 ing capital in another competing plant; besides, the 
 available factory was ill adapted to the making of clay 
 products. Steel- working offered no special advantages 
 such as did the presence of the low-priced timber supply. 
 Wood- working, therefore, was the logical field; but 
 what kind of product? In turn, the manager took up 
 and dismissed the manufacture of kitchen furniture, 
 wooden toys, cheap trunks, incubators, variety store 
 goods, and a number of similar lines. No one of these 
 would command a ready market or possess imperative 
 selling arguments. 
 
 By a process of elimination, the manager finally deter- 
 mined to make matches. Not by the patented auto- 
 matic process controlled by an organization of national 
 scope, but by the German method which broke match- 
 making up into a dozen or more operations by spe- 
 
 55
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 cialized machines. The speed and cheapness of the 
 American process were partly neutrahzed by the cost of 
 the cork pine required as material. The German ma- 
 chines, on the contrary, would chew almost any soft 
 wood up into match blanks and turn out an acceptable 
 product. Their operation was simple enough to be 
 learned in a day. The match produced had its own good 
 points, besides the powerful sales argument that it was 
 " not made by a trust." In addition, all the neighboring 
 cities were wholesale grocery centers, which would in- 
 sure a ready market distribution. To sum up in a sen- 
 tence, the new business was able to double its original 
 capital out of the profits of the first year because its 
 manager had analyzed his plant problems, coordinated 
 them with his operating activities, and worked out 
 a well-balanced scheme of production and distribu- 
 tion. 
 
 In solving his material problem, this match-maker 
 was not influenced in the least by personal preference or 
 prejudice, but based his decision entirely on logical con- 
 siderations. As a rule, indeed, the personal equation 
 does not enter into the choice of raw materials to any- 
 thing like the same extent as in the determination of 
 other plant policies. The character, quality, and price 
 of the product fix in a large measure the kind and qual- 
 ity of the materials to be used. Yet this is not always 
 true. One publisher, for instance, has such a strong 
 liking for a certain blue-white paper that he uses it for 
 the printing of his magazine, though he might make a 
 substantial saving by using a cheaper natural-white 
 paper which would meet every printing requirement 
 and which probably would be equally satisfactory to the 
 magazine's readers. 
 
 56
 
 MATERIALS 
 
 The manager's broad policy on materials must be 
 determined by his analysis of conditions from the stand- 
 point of (1) the kind of material to be used; (2) the 
 quality of material; (3) sources from which material 
 can be drawn; (4) the control of material; (5) the utiliza- 
 tion of material in the factory. 
 
 Within reasonable limits, it is clear that he must indi- 
 cate the kind of material to be used. In making filing 
 cabinets, for example, he would at least decide whether 
 wood or metal cabinets were to be produced. He would 
 be governed here by consideration not only of current 
 but of future conditions. Is the present supply of raw 
 materials adequate and reasonable in price.'* Does it 
 threaten to diminish or run out? In such an event, are 
 other sources available, or can a substitute be found at a 
 cost which the price of the product will allow? In cer- 
 tain of the wood-working industries, for instance, pine 
 lumber, when its cost became prohibitive, was replaced 
 by basswood; in like manner when the hardwoods used 
 for furniture appreciated in value, recourse was had to 
 a veneer construction which allowed combination with 
 various cheaper grades of lumber. 
 
 In prescribing the quality of the materials, the man- 
 ager must be governed largely by financial and market- 
 ing considerations, the selling price and the profit per 
 unit of output. When the product is a specialty with 
 a wide margin between the factory cost and the selling 
 price, the definite policy may be established of using 
 only the highest grade of materials even though cheaper 
 grades might be employed without sacrificing any of the 
 service rendered the consumer. This probably has been 
 the policy followed by the manufacturers of the more 
 expensive safety razors. Certainly it is true in general 
 
 57
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 that the making of a " quality " article requires " qual- 
 ity " materials. But the management should never lose 
 sight of the fact that the substitution of cheaper, though 
 not less efficient, materials might enable it to lower the 
 price to a point where a marked increase in sales would 
 result. 
 
 An instance of the substitution of materials against 
 trade custom is furnished by a manufacturer of vacuum 
 cleaners. He found that the use of pressed steel instead 
 of aluminum gave him a cheaper, lighter, and stronger 
 product. The first cost of the new dies was a strong 
 argument against the change, but it was outweighed, 
 according to the manufacturer's analysis, by the saving 
 that would follow. 
 
 The policies influencing the control of materials are 
 dictated not alone by technical or intra-departmental 
 considerations but by those which look to the general 
 good of the business. By control is meant, particularly, 
 the disposition of materials after they have been placed 
 in stock, rather than the insurance of adequate sources 
 of supply. The strategic position of the manager, suf- 
 ficiently withdrawn from departmental affairs to see 
 them all in their right proportions, enables him to deter- 
 mine policies which will make the most of his resources 
 and result in the greatest general efficiency of the busi- 
 ness as a whole. In his control of materials, for example, 
 he must strike a balance between an excessive inventory, 
 representing idle capital, and a possible shortage of ma- 
 terials which would mean a loss due to idle machinery. 
 In order to establish a balance between these two fac- 
 tors, his supply of materials must be brought into equi- 
 librium with the needs of his production departments. 
 
 It is here that the assumption of a fresh point of view 
 
 58
 
 MATERIALS 
 
 and the discarding of traditions to make way for de- 
 cisions based on actual conditions demonstrate their 
 value. According to the records of eflBciency engineers, 
 overinvestment in materials, the carrying of excessive 
 raw stocks, is a common fault in American factories. 
 The inventory is usually the first problem they address 
 themselves to, because it is at this point that the quick- 
 est and most convincing savings can be developed. 
 
 Instances could be cited almost without end : like that 
 of a Massachusetts company which projected a bond 
 issue in order to finance further extensions. Investiga- 
 tion of its assets discovered that at least $200,000 worth 
 of materials were in storage in excess of probable require- 
 ments, and the funds needed for expansion were released 
 by reduction of the inventory. Or that of an Indiana 
 manufacturing concern, too young and possessed of too 
 little in the way of tangible assets to attempt a bond 
 issue, which was able to finance a twenty-five per cent 
 increase in sales by strict coordination of its production 
 and purchasing program. 
 
 Unbalanced accumulation of stock and supplies is 
 generally the explanation of an excess inventory. One 
 department head is uneasy unless he has materials on 
 hand for six months' requirements; another finds stores 
 for three months entirely adequate. Unless it be more 
 difl&cult to secure prompt deliveries of the materials 
 used by the first department, or unless other considera- 
 tions, such as market fluctuations, are involved, it is 
 evident that any stock maintained in excess of three 
 months' needs involves a waste. Use of the capital tied 
 up in the surplus inventory and the space occupied are 
 lost for productive purposes. The importance of main- 
 taining an executive policy of control over stores is ob- 
 
 59
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 vious; without such general control, overinvestment and 
 unbalanced accumulation are almost inevitable. 
 
 The question of utilization also comes up for final de- 
 cision by the executive, particularly in cases where by- 
 products can be manufactured out of scrap material. 
 The packing industry affords, perhaps, the best known 
 illustration of utilization of raw materials. Even here, 
 however, development has been slow, and the manufac- 
 ture of by-products as a production policy was enforced 
 by economic necessity rather than adopted deliberately 
 as an essential of constructive business. 
 
 In the main, raw materials of every class have been 
 so cheap and abundant in the United States that there 
 has been no such pressure to secure complete utilization 
 as is felt in Germany, France, and England. Effort has 
 been concentrated on manufacturing processes and on 
 the turning out of the finished product rather than on 
 economy of materials. 
 
 Solution of these problems of control and utilization, 
 even more than problems of the kind and quality of 
 materials, is facilitated by a fresh point of view on the 
 part of the management. That a business man has 
 prospered, for instance, in the face of incomplete stock 
 records, haphazard buying, and slovenly control of ma- 
 terials in process, is not proof that stock records are 
 negligible or an inflexible system of requisitions for ma- 
 terials unimportant. It may simply mean that leaks in 
 the factory have been covered up by the high profit 
 margin of the business and that these losses, if detected 
 and checked, might have added materially to the net 
 returns or allowed price reductions to the consumer. 
 Too often stock records are regarded merely as insur- 
 ance against dishonesty, while their larger and more 
 
 60
 
 MATERIALS 
 
 useful function as a stimulus to efficiency is over- 
 looked. 
 
 So also with utilization. Modern production stand- 
 ards, it may fairly be said, are based on laboratory 
 studies. Essential in those factories which have to do 
 with highly technical or complex processes, this prac- 
 tice of establishing absolute standards may be overdone 
 or misdirected, but it is normally a most effective appli- 
 cation of a new point of view. No progressive manu- 
 facturer, for instance, takes chances any longer with 
 his glue, varnish, enamel, paint, wood, or coal. 
 
 The day of crude, rule-of -thumb experimentation has 
 passed. Producers now buy their materials on the basis 
 of exact knowledge ; they do not buy merely because of 
 somebody's casual recommendation or because the local 
 agent is a friend or because the first cost is low. The 
 furniture maker no longer buys twelve-cent glue if he 
 has found that eighteen-cent glue goes further and holds 
 best. Though higher in first cost, in final cost it may be 
 much the cheaper. Moreover, it saves trouble in the 
 factory and dissatisfaction on the part of customers, the 
 money value of which he can measure only indirectly. 
 To insure that he is getting the best quality of glue 
 available for his money, and that the quality of the 
 adopted brand does not subsequently deteriorate, he 
 often maintains a laboratory or employs a competent 
 chemist on a fee basis, by whom not only glue but all 
 his other materials of manufacture, including his fuel, 
 are systematically tested. This, with an efficient cost 
 system to show him what materials work up to the best 
 advantage and with the least waste, enables him to pur- 
 chase on a basis of true economy. 
 
 Standardization of materials can affect the sizes and 
 
 61
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 shapes in which these materials are bought as well as the 
 quality. A manufacturer of steel folding couches, for 
 instance, was making his products seventy-four and a 
 half inches long. The long pieces in the framework of 
 the couch were made of extra heavy material. The 
 nearest stock length in this weight left nearly two feet 
 of scrap which could not be used for any other purpose. 
 By reducing the length of the couch one and one half 
 inches, however, a slightly lighter angle iron that cut 
 with practically no waste could be utilized, making a 
 double saving of a cheaper material and less waste. 
 Before the change was made, investigation showed that 
 the demand for the seventy -three-inch couch was likely 
 to be just as great as for the longer model. 
 
 Somewhat similar was the experience of a large manu- 
 facturer of farm implements. At various intervals, as 
 models were added to the line, special designs were made 
 for all the parts, without any attempt at standardiza- 
 tion or the establishing of relations with other models. 
 Patterns for certain similar parts varied only an eighth 
 or a quarter of an inch in length; there were literally 
 hundreds of parts which might be made interchange- 
 able with other parts by means of minor changes. The 
 manager, bringing a fresh point of view to bear on a pol- 
 icy which had become traditional, ordered the redesign- 
 ing of the whole line with a view of standardization and 
 the purchase of raw materials in stock dimensions. 
 Thus the same parts were made available for use in more 
 than one model, and the same lengths and weights of 
 steel rods, bars, and angles became standard for many 
 similar parts. The savings were substantial: buying 
 was simplified and prices were generally lowered; the 
 inventory was reduced; less storage space was required; 
 
 62
 
 MATERIALS 
 
 the number of different sizes of raw stock carried was 
 cut in half; while the standardization of parts lowered 
 production costs and greatly reduced the stock of fin- 
 ished parts in storage. 
 
 Such radical departures from routine are likely to be 
 found only when the management has preserved its 
 strategic position, aloof from departmental details and 
 able, therefore, to view problems like this one of ma- 
 terials from the standpoint of general efficiency. Close 
 contact with processes and machinery and familiarity 
 with existing methods breed in foremen and others 
 directly identified with the handling of materials a 
 habit of acceptance. It is by the manager, then, with 
 his unbiased attitude and his ability to analyze condi- 
 tions that the opportunities for standardization can best 
 be observed and the most effective relations between 
 operations and materials can be established.
 
 CHAPTER VI 
 LABOR 
 
 WHEN the paymaster walks through the factory or 
 oflSce with his box of envelopes, he completes a 
 week's transactions between management and men. 
 The cash or checks are exchanged for labor. How much 
 money for how much labor depends on many conditions. 
 The unit of payment may be the hour or day, with 
 loosely defined minimum and maximum results as the 
 commodity delivered by the worker. Again, it may be 
 a swift repeat operation on a standard part or a complex 
 assembling process requiring many hours to finish. It 
 may be any one of several combinations or variations of 
 time units or result units. But in every case the purpose 
 is the same — to secure for each dollar expended the 
 largest regular day by day and year by year return in 
 productive effort. 
 
 To the employee the money is the essential factor in 
 the transaction. How far this pocket appeal can be in- 
 creased by the method of fixing and paying wages or 
 can be supplemented by cultivating the pride, ambition, 
 latent skill, or unawakened intelligence of the worker 
 depends upon the policy of the management in ap- 
 proaching and handling its labor problems. Low average 
 wages frequently occur in shops where production costs 
 are far above the normal; and thirty-dollar mechanics 
 may predominate in a factory similarly equipped where 
 goods are turned out at bed-rock figures. The re- 
 
 64
 
 LABOR 
 
 verse of both statements may also be true; the ratio 
 between wages and efficiency varies widely, though 
 there is a marked tendency toward leveling up and the 
 establishment of standards in certain groups of related 
 industries. 
 
 The reason for this variation is plain. In the hiring 
 and directing of labor, the human factor counts for 
 much more than in any other activity of production. 
 Assuming that the manager has been broad and wise 
 enough to minimize his own prejudices and preposses- 
 sions (reckoning with trade unionism, for instance, 
 more from the viewpoint of the efficiency of the whole 
 business and less from his personal, social, or sentimen- 
 tal standpoint) he will have to face, and find a way to 
 neutralize, the personal equations of his men. They 
 will concentrate on wages, hours of labor, and, perhaps, 
 recognition of the unions which have exerted so power- 
 ful an influence on both wages and hours. His problem 
 is likely to take the form of the query: What more 
 besides wages and hours? These are essential for mere 
 time-serving. Real efficiency must be bought at a higher 
 price and paid for either in added money or applied 
 intelligence of management. 
 
 His problem, like all business problems, can be solved 
 more easily and with greater certainty by breaking it 
 up into its constituent parts. Analysis, indeed, will 
 develop two groups of problems. The first has to do 
 with the classification of workers according to their 
 functions, the temperaments and characteristics of the 
 individuals usually exercising those functions, and the 
 motives which appeal most effectively to each group. 
 This classification is basic, for the second group, the 
 process-problems of labor (hiring, training, paying, and 
 
 65
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 directing) , will be found to differ in considerable degree 
 for each class — factory workers, factory foremen, oflBce 
 employees, salesmen, and general executives. Because 
 we are dealing as yet only with production policies and 
 need not go beyond the factory to illustrate a systematic 
 approach to the general problem of labor, any reference 
 to the factors which must be considered in the hiring and 
 managing of men outside the production departments 
 will be only incidental. 
 
 Viewed broadly, factory workers have certain class 
 characteristics which make the problem of managing 
 them difiFer materially from the problems which arise 
 in the managing of salesmen or oflBce help. In common 
 with all other groups of intelligent workers, they have 
 ambition; but it is usually ambition tempered by a cer- 
 tain distrust of the employer's motives and held in 
 check by what amounts to a class philosophy. It is the 
 manager's task to overcome this suspicion and to make 
 clear that loyalty to the company, interest in its prog- 
 ress, and pride in its product are entirely compatible 
 with loyalty to labor organizations, where such organ- 
 izations exist. He must keep in mind that the narrow 
 education of the worker and the constant pressure of 
 his environment combine in urging him to treat the 
 employer as an impersonal force whose interests are 
 selfish and therefore hostile, and to bargain with him 
 on the basis of giving as little and getting as much in 
 return as circumstances will allow. 
 
 We may as well admit that this is the attitude of 
 skilled and semi-skilled workmen in every large labor 
 market. To escape it many firms have chosen to move 
 to country towns, sacrificing whatever advantages the 
 large city offered to escape disturbing labor conditions. 
 
 66
 
 LABOR 
 
 Other concerns have joined issue on the question, de- 
 claring for the open shop on the ground that unionism 
 is the chief factor in antagonizing workers. Still other 
 managers have taken for granted a certain indifference 
 on the part of employees and have directed their efforts 
 toward changing this attitude to one of confidence in 
 the management, comprehension of what it is trying to 
 accomplish both inside the organization and outside, 
 interest in the day's work, however monotonous, pride 
 in the quality or quantity of product, and cooperation 
 for mutual advantage. This is an ambitious program, 
 but it is these latter executives who have made the most 
 important advances in the methods and philosophy of 
 employment, and it is their experience which is drawn 
 upon in the shaping of this chapter. 
 
 When a manufacturer analyzes his labor problem into 
 its elements, the question of balanced output comes 
 first and looms largest in his thoughts. How can he 
 secure a reasonable maximum in quantity, the essential 
 accompaniment of low production costs, yet maintain 
 requisite standards of quality and avoid more than a 
 minimum of spoiled parts? Bearing on this, his main 
 objective, are the personal traits of his employees, their 
 honesty and industry, their skill and sense of respon- 
 sibility, and also their mental attitude toward their 
 work and toward his management. Personal traits and 
 qualities are fundamental considerations in hiring men, 
 though the good ones may be developed by subsequent 
 training and the harmful ones corrected or eliminated. 
 Assuming a factory force of average spirit and caliber, 
 however, its mental attitude is a problem of daily prac- 
 tical concern. How can it be influenced.'' How can it 
 be brought into accord with the purposes of the manage- 
 
 67
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ment and be made a constructive force, instead of one 
 merely negative or antagonistic? 
 
 The broad basis of appeal, he will find, is pay scientifi- 
 cally proportioned to individual efiFort. Ease, conven- 
 ience, and safety are convincing arguments. Caution, 
 as to retaining place, service credits, preferred work, 
 or other privileges, is a powerful motive to invoke. Pride 
 in workmanship, in the company's standing, in the per- 
 fection and value of its products, is a sentiment which 
 can be stirred only when conditions are generally satis- 
 factory, that is, when the worker is convinced that he is 
 receiving a square deal. Praise of a man's ability or 
 achievement soon loses its savor unless it " gets him 
 something " in the way of increased pay or promotion. 
 
 Money and material utilities are the solid foundations 
 on which alone efficiency can be built. The workman's 
 natural tendency is to strike a balance between the ad- 
 vantages the manager offers and those held out by other 
 employers. No appeals less tangible are allowed to out- 
 weigh these necessities. Certainty of employment on 
 full time or nearly full time is an essential for men who 
 have families and others who want to get ahead. Self- 
 interest, in fact, is the only compelling motive which can 
 be enlisted to secure cooperation in the betterment of 
 existing methods or the mstalling of improved systems. 
 If he can adopt the easier or more direct method without 
 losing any of his pay per unit of production, the average 
 factory worker will welcome the change. Unless he gets 
 a share of the profit coming from his increase in output, 
 however, suspicion that he is being " speeded up " is 
 likely to follow. Confidence in the manager and his mo- 
 tives must precede any successful attempt to increase 
 the productive capacity of an organization. 
 
 68
 
 LABOR 
 
 Paying labor, then, is the crucial problem, though 
 training, placing, and directing are vital questions in any 
 eflSciency program. Hiring, a preliminary operation, will 
 not be taken up here because, at any rate in the large 
 plants, it is usually an administrative function and will 
 be reserved for discussion with other activities of ad- 
 ministration in a later chapter. Any plan of payment, 
 we have seen, should furnish the worker with an incen- 
 tive to increased production through addition to his 
 income. The best wage system, indeed, is that which 
 makes clear to the worker the exact relation between his 
 effort and the amount which he receives for it. At the 
 same time, it must not add to unit production costs. The 
 manager must hold at least to his established ratio of 
 cost, quality, and service, while the prime object of any 
 change, of course, is to cut down the first element in this 
 business equation. This to the management is no less 
 a test than self-interest to the employee, of any new 
 method of wage payment. Does it actually lower pro- 
 duction costs? Personal prejudice, sentiment, or con- 
 venience should never be permitted to govern or even 
 cloud the decision. 
 
 In his approach to this problem of payment, the 
 manager has for his guidance the experience of other 
 employers, many of whom have conducted extended 
 investigations and long series of experiments to deter- 
 mine the most effective wage plan. He must under- 
 stand, however, what his real problem is; how it divides; 
 what are the factors that count in the solution of each 
 sub-problem; and how he can apply in his own factory 
 the methods which have been more or less successfully 
 demonstrated by other manufacturers and production 
 specialists. 
 
 69
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Dividing the problem into its constituents and con- 
 sidering each, the manager finds himself facing a choice 
 among several methods of paying men. There are the 
 systems long in use, the day rate, the piece rate, and 
 the contract system, all relatively simple, but all having 
 the common weakness that the personal factor is large 
 in determining how much productive effort labor shall 
 deliver for each dollar received. Where the straight day 
 wage is maintained, output is often a compromise be- 
 tween customary trade standards plus the man's idea 
 of what is fair and the ability of the management to 
 speed up its machines and workmen without causing 
 revolt. Whether this compromise basis is above or 
 below the average depends chiefly on the personality 
 of the manager, on his sympathy, magnetism, and other 
 qualities of leadership, and in lesser degree on the size 
 and character of the force itself. 
 
 The great defect of the traditional day rate is that it 
 offers no direct incentive for the individual employee to 
 exert himself to his full capacity. The piece rate is an 
 advance on this primitive method of measuring energy 
 in terms of time. Where many repeat operations are 
 performed and where time studies or a series of obser- 
 vations and experiments have established trustworthy 
 standards, the piece rate supplies an incentive to higher 
 efficiency and a fuller utilization of the plant. When 
 such standards have not been worked out, however, it 
 gives too much play to the personal equation in setting 
 rates and in the majority of shops leads the men to put 
 a secret but practical limit on individual production. 
 Moreover, without an adequate inspection system, 
 quality may be sacrificed to output. In a word, the 
 interests of manager and men are still tangential in- 
 
 70
 
 LABOR 
 
 stead of parallel. The contract system is simply the 
 piece rate applied to a group or cycle of operations in- 
 volving a number of workmen, and is subject to the 
 same drawbacks and limitations. 
 
 From the simple forms of both day rate and piece 
 rate various systems of payment have been developed 
 to overcome their basic weakness, that is, lack of an 
 imperative incentive for the workman to employ time 
 and equipment to their fullest productiveness. The 
 straight piece rate, which pays one price per unit of 
 work done whether the rate of eflSciency be high or low, 
 has been supplemented in many factories by the fixing 
 of a daily or weekly task which the workman must equal 
 in order to hold his place. The graduated or differential 
 piece rate modifies this appeal to caution by offering a 
 high rate for performance of the task in a given time 
 and a lower rate if the workman falls short of this stand- 
 ard eflSciency. An interesting variation is the Franklin 
 quality piece rate which puts the emphasis on perfect 
 work by paying the maximum when spoilage is elim- 
 inated and reducing the rate paid as the proportion of 
 spoiled parts increases. 
 
 The development of the day rate has taken shape in 
 bonus or premium systems, which are based on the idea 
 of rewarding the workman for saving time and cutting 
 costs by sharing with him the money thus saved. The 
 Towne system was one of the earliest of these; it made 
 an annual pro-rated distribution of the saving in labor 
 costs over those of the best preceding year, the company 
 taking its reward in the increased production per fac- 
 tory unit. The Halsey premium plan went a step fur- 
 ther. It made the workman's reward immediate and 
 individual by fixing a definite time limit for each opera- 
 
 71
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 tion (based on past performances), paying the worker 
 his regular hourly rate for the time spent in executing 
 it, but allowing him a bonus if he finished in less than 
 the specified time. This premium is usually half the 
 labor cost thus saved. The Rowan system differs from 
 the Halsey plan by making the workman's premium 
 a percentage of the standard labor cost equal to the 
 percentage of the standard time which his saving 
 represents. 
 
 Both the Towne-Halsey and the Rowan systems 
 showed flaws in operation — the first by putting the 
 fixing of the standard times entirely in the hands of the 
 workmen, the second by cutting the premium paid for 
 any saving of more than fifty per cent. They are ob- 
 jected to further because as the workman's production 
 increases, his earnings per piece decrease. To avoid 
 these faults, the Emerson system based its standard 
 times on workmen's performances corrected by expert 
 motion-and-time studies, demanded two thirds of this 
 standard eflBciency of every workman, and paid gradu- 
 ated premiums for every increase above the minimum. 
 When the workman equaled the standard time, his 
 bonus amounted to twenty per cent of his wages; above 
 the standard time the premiums increased one per cent 
 for each per cent of increase in eflSciency. 
 
 The Taylor differential piece rate puts forward a 
 final incentive to workmen by providing scientifically 
 determined standard conditions and times for work, 
 and offering higher piece rates as the standard time is 
 approached and passed and lower rates as the work- 
 man drops down in the scale of eflSciency. The task 
 and bonus plan evolved by H. L. Gantt in connection 
 with the Taylor system guarantees the regular hourly 
 
 72
 
 LABOR 
 
 rate and offers the man a bonus for attaining a previ- 
 ously determined high level of efficiency, each man's 
 bonus increasing as his output above the prescribed 
 standard increases. The ratio of the bonus to the regu- 
 lar hourly rate varies among different trades and occu- 
 pations, according to the character and difficulty of the 
 work. The regular hourly rate is paid while the man 
 is under instruction. 
 
 There is yet another approach to this task of pro- 
 viding an incentive to workers — profit-sharing. In 
 the broader sense, it amounts to any collective pay- 
 ment of bonuses for faitliful service, predicated on sat- 
 isfactory net earnings for the employer. It may take 
 the Henry Ford form of decidedly higher day wages, 
 which divide with the workers every week a prede- 
 termined amount of the estimated profits of the com- 
 pany's current operations; though this might be a 
 dangerous proceeding for the average business venture, 
 which is subject to the ups-and-downs of the consuming 
 market. The more familiar form is an arbitrary sum 
 set aside by the directors at the end of every prosper- 
 ous year for distribution among employees, the size of 
 the lump sum and the amount of the individual dividend 
 depending on the measure of the company's prosperity, 
 thus establishing an incentive for each worker. 
 
 The more sagacious plans do not leave the size of the 
 worker's dividend for the management's decision at the 
 end of the year. The amount to be shared is predeter- 
 mined; or at least the percentage or proportion of the 
 net profits to be shared is announced beforehand. In 
 fact, in the strict technical usage of the term, profit 
 sharing means a method of remuneration by which the 
 employees receive in addition to standard or normal 
 
 73
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 wages a share of the profits fixed in advance. The 
 manner of distribution varies. The A. W. Burritt 
 Company, for instance, after paying six per cent on 
 capital, divides its net profit between the stockholders 
 and the workmen in the ratio of capital invested to 
 wages earned, each workman receiving an amount 
 proportionate to his own wages. The workmen agree 
 also to share similarly any net losses, up to ten per 
 cent of their wages. The Farr Alpaca Company pays 
 the same dividend on a dollar of wages as is paid on 
 a dollar of capital. The R. F. Simmons Company, 
 which began by sharing with its workers from eight 
 to twelve per cent of the dividend voted on its capital, 
 has gradually increased the amount so as to insure its 
 representing four or five per cent of the employees' 
 annual wages. The Simplex Wire and Cable Com- 
 pany determines in January the percentage of profits 
 to be paid at the end of the year to workers who 
 have been connected with the company for twenty-six 
 months or more. 
 
 This latter emphasis on continuity of service is com- 
 mon to nearly all profit-sharing plans, which fix a min- 
 imum period for eligibility and increase the dividend in 
 proportion to the time of employment. Profit-sharing 
 programs range widely, indeed, in character and effec- 
 tiveness; their weakness is that the collective bonuses 
 they offer fall short of providing the constant, specific, 
 and measurable incentive to individual workers which a 
 scientific system of wage payment should supply. 
 
 Facing these successful wage systems (outlined merely 
 in principle here 0> the manager can make a wise deci- 
 
 ^ For a fuller discussion, see C. B. Thompson, " Scientific Management," 
 and the works cited in the Bibliography beginning on page 863 of that book. 
 
 74
 
 LABOR 
 
 sion only by keeping a tight grip on all the factors in- 
 volved. His product may be of such a type that the day 
 rate is the only possible method of maintaining quality 
 or satisfying his production conditions. His organization 
 may be of such size or character that the cost of instal- 
 ling and maintaining a differential piece rate, with all 
 the conditions involved, would be prohibitive. Be- 
 tween these extremes there are a score of alternatives. 
 The conditions in different departments of the same 
 business, in fact, often call for different wage systems. 
 In choosing the one or more than one which fit his situ- 
 ation, the executive must keep his principle of balance 
 always in mind. The fresh point of view, too, is of more 
 than usual importance, since the wage systems of most 
 industries are traditional compromises or guesses devel- 
 oped like all other rules of thumb. Its value, indeed, 
 has been generally recognized : witness the recent wide- 
 spread interest in the Taylor system of shop manage- 
 ment and other efficiency methods, and the associated 
 movements in many industries to reduce labor costs and 
 conditions to something like common and reasonable 
 standards. With the overhead charges on factory oper- 
 ations averaging as much as the direct labor costs, the 
 rewards are for the manager who can hasten produc- 
 tion and thereby reduce the proportion of rent, light, 
 heat, power, and other fixed charges which each unit of 
 product must bear. There, in countless plants, the 
 zone of profit lies. 
 
 But a wage incentive, however alluring, is not enough 
 to get results from factory labor. With the best will in 
 the world, a workman must have the knowledge and 
 skill his job demands, or his output will suffer in quan- 
 tity or quality. This knowledge and skill can frequently 
 
 75
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 be hired in the larger markets; with a certain amount 
 of " breaking in," a new workman becomes capable of 
 handling the tasks for which he was employed. Quite 
 as often competent mechanics or machine operators are 
 not to be had; or, if available, they cannot be engaged 
 except at rates higher than conditions in the business will 
 justify. The only alternative, then, is for the manager 
 to train unskilled labor to the level of eflficiency required. 
 
 Every factory organization confronts this situation 
 at more or less regular intervals. For at least a genera- 
 tion the apprentice system, limited in its application 
 by the rules of the unions and neglected by self-centered 
 employees, has been inadequate to provide the skilled 
 workers our industrial expansion called for. Hundreds 
 of thousands of English and German mechanics, emi- 
 grating in answer to the call, helped to supply the lack. 
 The division of labor and the development of automatic 
 machinery of endless variety and countless functions 
 further served the need. But the burden of getting and 
 retaining men was not lifted from the manufacturer. 
 Each machine required an operator; and its output in 
 no small degree was measured by the thoroughness of 
 the training grafted on his natural intelligence and 
 dexterity. 
 
 This fact gives the Taylor system and similar ef- 
 ficiency programs their chief significance. Their stand- 
 ardizing of materials, machines, processes, and methods 
 by means of time and motion studies is all preliminary 
 to the final task of teaching each worker how to perform 
 his operation with the least possible outlay of time and 
 energy. They approach each process, each separate 
 motion, indeed, from the engineering viewpoint. Is it 
 necessary? Is it the most effective way of doing the 
 
 76
 
 LABOR 
 
 thing? By observation, analysis, and experiment they 
 arrive at a standard method. This method they dem- 
 onstrate to the worker, helping him to master it. 
 They watch him try it time after time, day after day, 
 if necessary, correcting his mistakes, explaining the 
 how and why of each detail, holding him to the exact 
 technique they have determined as the best. As we 
 have seen, they get the man's cooperation by paying 
 him his full wage while he is under instruction and show- 
 ing him how he will profit by following the new proced- 
 ure. The increase in wages, so graduated that the 
 highest premium can be gained only by the fullest 
 cooperation with the management, supplies the incen- 
 tive to effort. But the fixing of the standards and the 
 training of the workmen are indispensable to the success 
 of the plan. 
 
 Under the best systems, indeed, the training of the 
 worker never ends. He is not allowed to earn the maxi- 
 mum bonus one week and slump the next. An instruc- 
 tor-foreman, having a limited group of men in charge, 
 keeps close watch on his output and exerts all his knowl- 
 edge, tact, and authority to keep it up to standard. 
 As the instructor's own bonus is based on the success 
 of his men in earning theirs, the efficiency level is usually 
 maintained. In addition to the instructor or gang boss, 
 there are other functional foremen, such as speed and 
 repair bosses, inspectors and planning-room men, each 
 of whom contributes his special aid in achieving stand- 
 ard efficiency. 
 
 One of the indictments brought by sociologists and 
 humanitarians against the modern factory system of 
 production is that the minute subdivision of labor 
 dwarfs the development of the worker and makes him 
 
 77
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 a one-operation specialist rather than an all 'round me- 
 chanic. At first sight efiiciency systems would seem to 
 exaggerate this tendency. On close observation, how- 
 ever, the opposite would seem true. The fresh view- 
 point which standardization proposes to the workmen, 
 the mental exertion required to master the technique of 
 the new methods, the processes of analysis and syn- 
 thesis which take place under his eyes, all open new 
 channels of thought and transform him from a creature 
 of habit to a man awake, with a mind open to new 
 things. If he has average intelligence and ambition, 
 after mastering one process or series of operations he 
 " gets the hang " of the new idea and finds it easy to 
 master a second and a third or any number of opera- 
 tions which shop exigencies may propose. 
 
 This should eliminate monotony of occupation and 
 supply something like an equivalent of the training an 
 apprenticeship or a trade school might have given him. 
 From certain factories where the Taylor system has 
 been installed come reports of individual progress of 
 this sort which would indicate the possibility of all 
 'round training for employees without any sacrifice of 
 output while their education was going on. It is said 
 that at the Tabor Manufacturing Company, Phila- 
 delphia, two young men who gave promise of executive 
 ability were " passed through " all the departments of 
 the factory in two years, reaching during that period 
 the eflBciency standards of every operation in the plant. 
 
 This development of capable, " all 'round " men, first 
 for the performance of single processes and finally for 
 the supervision of groups of workers engaged on a whole 
 cycle of operations, has long been recognized as an im- 
 portant phase of the management's labor problem. The 
 
 78
 
 LABOR 
 
 subdivision of labor as carried out in a great many in- 
 dustries has meant that the apprentice or the unskilled 
 man learns only to tend certain special machines which 
 perform a limited range of operations on an unending 
 succession of similar jobs. His work becomes auto- 
 matic; there is nothing to cultivate skill outside his 
 narrow field, to develop his thinking power, or to help 
 him to see his own detail in its relations to the general 
 production scheme. Take him off his machine or away 
 from his bench and he must start all over again. 
 
 Coupling this tendency with a general breakdown 
 of the apprenticeship system, many of the older and 
 larger industrial organizations, like R. Hoe & Com- 
 pany and the General Electric Company, have estab- 
 lished apprentice schools in which selected boys are not 
 only taught a trade, in the former sense of the phrase, 
 but are given regular class-room instruction in shop 
 arithmetic, geometry, algebra, drawing, physics, me- 
 chanics, strength of materials, and subjects directly 
 related to the technique of the industry or calculated 
 to fit them for understanding all its shop processes. 
 This class work, which occupies usually from three to 
 twelve hours a week, is partly or wholly on the em- 
 ployer's time, and the apprentice is required to give it 
 the same attention and show the same progress in it 
 as in his shop work. 
 
 At the General Electric works, the courses are 
 planned to produce expert machinists, pattern makers, 
 iron molders, core makers, and draughtsmen, though a 
 few boys in every class are taught blacksmithing, tin- 
 smithing, and steam fitting. The course lasts four 
 years, during the first three of which the boys are given 
 experience on the various machine tools and are put to 
 
 79
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 work on as many different processes as possible. In the 
 fourth year of service they are rated as competent work- 
 men, so far as quality of work is concerned, though still 
 under direct charge of the superintendent of appren- 
 tices. The draughting apprentices serve their second 
 year in the shops doing the same class of work, under 
 the same direction, as the machinist apprentices. Anal- 
 ogous courses for " student engineers " are also con- 
 ducted by this same company with the object of giving 
 graduates of technical schools, when they enter its 
 employ, experience in all its production departments 
 and an intimate grasp of their activities. 
 
 Because the cost of equipping and maintaining effec- 
 tive schools for apprentices puts them beyond the reach 
 of the ordinary business, the manufacturers of Fitch- 
 burg, Massachusetts, have adapted to their local needs 
 the " half-time " plan of education conceived by Dean 
 Herman Schneider for the University of Cincinnati's 
 engineering school. 
 
 Under the Schneider plan, seventy-five or more in- 
 dustrial plants, construction companies, and transporta- 
 tion lines cooperate with the university in the training 
 of the engineering students. The latter work in pairs, 
 one man attending the university while the other works 
 for wages at one of the cooperating plants. They alter- 
 nate bi-weekly at study and service in their common 
 job and thus get an opportunity to " hitch their theo- 
 retical knowledge to practical things and bring a prac- 
 tical viewpoint to their studies." Incidentally some of 
 them earn enough on the various jobs they have, during 
 their sequential training through the departments of 
 the cooperating concern, to keep them in school, and 
 they all have the valuable experience of turning out 
 
 80
 
 LABOR 
 
 commercial products under actual working conditions. 
 The manufacturers secure high-grade intelligence prac- 
 tically and theoretically trained at the wages which 
 they pay any one else for the same class of work; also 
 they have a chance to pick possible executives, de- 
 signers, salesmen, and other high grade employees 
 among the student graduates. At the same time the 
 university is spared the expense of equipping its lab- 
 oratories with expensive machinery and of maintaining 
 a staff of machinists and instructors to direct students' 
 practice work. 
 
 Fitchburg's manufacturers have a similar coopera- 
 tive arrangement with their city high school. The first 
 year of the four-year course there is no interchange 
 between school and shop; but during the last three 
 years the pairs of pupil-apprentices serve alternate 
 weeks under teachers and foremen. The school does 
 not attempt to teach anything about practical shop 
 work; but concentrates on mathematics, mechanical 
 drawing, elementary physics, and chemistry, and other 
 branches which give the pupil a grip on theory and in- 
 terpret shop practice, broadening his outlook and in- 
 creasing his capacity. It tries to answer his " why " 
 questions and leaves the " hows " to the mechanics and 
 foremen of the different factories. 
 
 Some manufacturers have brought trade schools into 
 their schemes of training labor. Knowing that the 
 conditions under which factory production goes for- 
 ward make it difficult for the apprentice or young 
 helper to get more than a smattering of the underlying 
 trade, not a few pick out at intervals boys of more than 
 average character, force, and intelligence and help them 
 to put themselves through a trade school in order to 
 
 81
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 master their chosen craft. The purpose, of course, is 
 to secure workmen of wider experience, skill, and 
 knowledge than can be developed in their own plants 
 and to train possible executives. There are numerous 
 cases where specialists of various sorts, chemists, fac- 
 tory engineers, and the like have been developed from 
 the ranks by employers observant enough to notice 
 unusual aptitudes in factory workers and keen enough 
 to give these exceptional men the backing necessary to 
 secure them special technical training. 
 
 A most important general policy governing the prob- 
 lem of labor has to do with the manager's attitude 
 toward trade unionism. As already suggested, it is 
 in this connection that the personal factor is most diffi- 
 cult to discount. Opposition to organized labor, to col- 
 lective bargaining, to contracts secured and enforced 
 by implied or expressed threats of strikes may be said 
 to be one of his normal habits of mind. For the or- 
 ganization of his employees on occupational lines is 
 certain to bring about interference with his free hand 
 in running the business. In some businesses, however, 
 compensating factors have arisen from recognition of 
 the unions and a surrender of a portion of the em- 
 ployer's power over shop discipline and conditions. 
 In a notable clothing concern in Chicago, for ex- 
 ample, the establishment of machinery for adjusting 
 disputes is said to have brought peace, stability, and 
 production betterments to an industry in which antago- 
 nism and unstable conditions have always obtained. 
 From the viewpoint of the manufacturer, it may be 
 worth while to surrender part of his authority and con- 
 trol in order to gain a spirit of contentment and coopera- 
 tion in the factory. 
 
 82
 
 LABOR 
 
 Frequent '* turnover " of labor is the outward ex- 
 pression of another serious labor problem. Manufac- 
 turers in both small town and city experience much 
 difficulty in holding competent operators of certain 
 standard and widely used machines, screw-making ma- 
 chines, milling machines, and so on. These are men 
 above the unskilled in intelligence, though lacking 
 exact knowledge of a trade, and therefore more sensitive 
 to the monotony of repeat operations indefinitely con- 
 tinued. Relief from this monotony is denied usually by 
 factory conditions and the lack of training; it is found 
 by moving on to the next city where there is a steady 
 demand for this kind of labor. 
 
 Efforts to hold these " floaters," usually most restless 
 in seasons of maximum production, have taken many 
 forms. Company " thrift clubs " and other plans for 
 encouraging and financing the purchase of homes is one 
 avenue of approach. So-called " welfare work " de- 
 signed to make the factory more attractive from the 
 standpoint of comfort, convenience, and social and 
 working conditions is another. An eastern manufac- 
 turer comes closest, perhaps, to the root of the difficulty 
 by studying methods of restoring variety to the day's 
 work. Instead of putting through work in quantities as 
 large as is commercially possible, he takes the monot- 
 ony factor into consideration and manufactures in lots 
 small enough to allow frequent shifts to new tasks for 
 his machine operators. This plan, he has found, reacts 
 as well on other production difficulties, lessening to a 
 large extent the proportion of spoiled parts in each 
 factory order. 
 
 With many managers, it is a settled policy to try by 
 other means to hold workers who have made their hon- 
 
 83
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 esty, industry, and good will evident by service. Be- 
 fore a man can be paid off, even when leaving on his 
 own initiative, either the manager or some one in his 
 confidence has a talk with the employee about his mo- 
 tives for leaving. If the reason is a condition which 
 can be corrected, the change is made. If a transfer to 
 another department or another task will remove the em- 
 ployee's dissatisfaction, the change is made whenever 
 possible. In many successful organizations, indeed, 
 discontent on the part of the worker is anticipated 
 and it is made one of the chief duties of the employ- 
 ment official or department to discover the capacities, 
 ambitions, and desires of all workers and to endeavor 
 to place them, as soon as practicable, in the positions 
 which will bring out the fullest measure of their in- 
 terest and productive powers. 
 
 After adjustment of the wage factor — which may or 
 may not include profit-sharing — the placing and train- 
 ing of workers are the two policies most vital in the 
 building up and holding of an efficient factory force. 
 Give a man the work he wants to do; help him to fit 
 himself for its effective performance; pay him fairly for 
 what he does, and the other factors in the equation will 
 almost lose themselves; so strong is the creative, the 
 productive instinct, so real the joy in doing the work 
 that fits. 
 
 84
 
 CHAPTER VII 
 
 ORGANIZATION 
 
 OTARTING with a factory, a working force, and a 
 ^ supply of raw materials, the manager faces a final 
 production problem, that of effectively organizing his 
 operating activities. He has a definite end in view. To 
 the smallest necessary amount of wood, metal, clay, or 
 fiber he wants to add the fewest motions required to 
 turn out a properly balanced unit of product. To do 
 this he must so coordinate and direct the application of 
 these motions that there shall be no duplication or loss, 
 no waste of time or stock or energy. And he must fur- 
 ther provide a permanent method of controlling both 
 motions and materials in order that this equilibrium of 
 means and results shall be preserved. 
 
 Inefficient management involves either the applica- 
 tion of too much power or labor to a given unit of mate- 
 rial, or the use of too large or too valuable a quantity 
 of raw stock in the manufacture of a standard unit of 
 product. In flagrant cases it may mean that all the 
 elements which enter directly into manufacture — ma- 
 terial, labor, power, machinery, space, light, and so on 
 — are entirely out of proportion to the result which is 
 obtained. Efficient production, on the other hand, is 
 always the consequence of establishing and maintaining 
 a balance of all the elements that count in the making 
 and, to a lesser degree, in the marketing of an article. 
 
 The factory head must so organize his plant and proc- 
 
 85
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 esses that he can turn goods out economically. Yet 
 in his attention to cost he must not lose sight of qual- 
 ity, accuracy, durability, beauty, or whatever is the 
 characteristic which measures the service to customers. 
 In the building of machinery, for instance, it may be 
 possible to increase output substantially by permitting 
 small deviations from standards in cutting gears, ream- 
 ing out bearings, and like finishing processes. But lax- 
 ness in this respect is almost certain to be followed by 
 the development of lost motion when the machine is 
 operated by the purchaser, and lost motion always 
 means greater consumption of power and a shortened 
 life for the machine itself. This inevitably reacts on 
 sales and sometimes brings disaster, as in the recent 
 experience of a great industrial concern whose chief 
 product was sold on long-time payments. 
 
 In an effort to bring output up to a quota based en- 
 tirely on selling and financial considerations, this fac- 
 tory was enlarged and its working force increased to a 
 point where the old plan of organization was inadequate 
 to insure efficiency. Relaxed or incompetent inspec- 
 tion allowed many faulty parts to be incorporated in 
 machines. Tried out by the purchasers, so many of 
 these were thrown back on the agents' hands that the 
 company's reputation suffered, a financial crisis fol- 
 lowed, and complete reorganization was necessary to 
 protect creditors and stockholders. Other factors, be- 
 sides too rapid expansion of the working force and 
 scamped workmanship, undoubtedly contributed to the 
 failure. The same emphasis on volume led to general 
 overselling by the field force and a breakdown of the 
 plan for financing deferred payments on orders. Both 
 of these weaknesses might have been corrected, how- 
 
 86
 
 ORGANIZATION 
 
 ever, if the plant had shipped standard and satisfactory 
 machines to all buyers. It was the flood of returns 
 which shook confidence in the company's products, 
 neutralized its selling efforts, and ultimately destroyed 
 its credit. Because its factory organization lost control 
 of production and failed to maintain the right ratio be- 
 tween cost, quality, and service, an industry of almost 
 national scope collapsed like a house of cards. 
 
 Control of operations, then, is the culminating func- 
 tion of production. The man at the head of a small 
 plant has need only of a personal system or a personal 
 routine, perhaps, which will bring him into contact at 
 frequent intervals with all his workmen and all the jobs 
 they are engaged on. If he has a real knowledge of the 
 technical processes and a certain amount of personality 
 — the typical " one-man " business seems to be built 
 largely on these two elements — he can at the same time 
 control the quality and cost of output, hasten pro- 
 duction, and accommodate it to sales or financial exi- 
 gencies to a degree which few large businesses can hope 
 to equal. But when he finds it necessary to delegate 
 authority because important details have so multi- 
 plied that he can no longer handle all of them, a more 
 formal organization becomes essential if his plant is not 
 to lose efficiency. As Alfred Marshall senses it, the 
 crucial point in business expansion arrives when orders 
 must be written instead of being spoken. 
 
 To keep control of operations while delegating su- 
 pervision and authority is the manager's broad prob- 
 lem in organization. Once he has determined the lines 
 on which the latter can be divided and has established 
 his underlying policies in accord with the principle of 
 balance and the principle of interdependence, he can 
 
 87
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 hire or train men to whom he can turn over the direc- 
 tion and the responsibility of all or nearly all his 
 operating activities. Experience indicates that this is 
 naturally the first step taken toward the division of 
 labor in the field of management. This, it has already 
 been shown, is for two reasons : first, because production 
 (while much is still to be done) has so largely been stand- 
 ardized and so many of its processes and reactions re- 
 duced to generally accepted and measurable terms; 
 second, because classes of specialists have been devel- 
 oped who are competent to take charge of operations 
 and hold them to any predetermined balance of cost, 
 quality, and service. When the growth of his business 
 or the initial size of his undertaking forces the executive 
 to substitute organization for personal supervision, then 
 it is in the factory that he finds it easiest and most 
 profitable to share his functions and responsibility with 
 subordinates. His problem, therefore, is to discover 
 how best to delegate his authority in order to retain 
 effective control. 
 
 When he consults the experience and practice of other 
 manufacturers, he finds in use three distinctive types 
 of shop organization. These differ in the manner of 
 delegating authority and control, in the extent of the re- 
 sponsibility fastened on individual executives and fore- 
 men, and most of all in the philosophy of business which 
 informs them. Widest apart are (1) the line type of 
 organization and (2) the functional type. Midway be- 
 tween these and having characteristics in common with 
 both is (3) the line-and-staff. In considering which 
 kind of organization is likely to prove most applicable 
 and most efficient in his own factory, there is hardly 
 need to say that the factors which determine his choice 
 
 88
 
 ORGANIZATION 
 
 should not be personal in their nature or relations and 
 that the decision should not hinge solely on the man- 
 ager's experience or familiarity with a particular type 
 of organization or on the usage of the district or the 
 industry in which he is engaged. 
 
 In breaking up his problem of organization into its 
 constituent problems and assembling the factors which 
 enter into the solution of each, he will find it necessary 
 to analyze and consider from the viewpoint of his own 
 business the principles and the working of the three 
 types of control just named. In the line organization 
 — the traditional form, and until our own generation 
 the only common type — the authority and responsi- 
 bility are delegated throughout. The head of each de- 
 partment and sub-department is held responsible for all 
 that happens within his jurisdiction. It follows that his 
 functions are of many kinds and large in number. He 
 must hire his men, place them where their capacities 
 will be most useful, determine what is a fair day's work 
 for each and what payment should be made for it, watch 
 his machines to see that they are in proper condition, 
 plan his work in such a way and keep material so mov- 
 ing in the shop that the men and equipment are always 
 producing. He is expected to be able to show a work- 
 man how an operation should be carried out to secure 
 maximum results with the least expenditure of time 
 and energy, and in an emergency to give exact instruc- 
 tions for meeting it. 
 
 This list of duties, incomplete as it is, illustrates both 
 the advantages and disadvantages of the line type of 
 organization. It unifies work by concentrating author- 
 ity and definitely fixes responsibility upon certain in- 
 dividuals. As the shop grows larger, however, the 
 
 89
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 demands made upon the foreman's time, skill, intelli- 
 gence, and patience, increase to a point where it is diffi- 
 cult to satisfy them. He may carry the hardest of his 
 problems up to the superintendent or even to the head 
 of the business, while his responsibility for groups or 
 cycles of operations is frequently divided by putting 
 these in charge of job or section bosses. But the fore- 
 man's mental and physical limitations remain; and the 
 efficiency of the individual workman and of the factory 
 as a whole suffers unless the size of the shop or the 
 special character of the operations allows each boss to 
 master the technical details and oversee the perform- 
 ance of all the processes for which he is held responsible. 
 
 If an industry is to satisfy these conditions and best 
 lend itself to economical application of line organi- 
 zation, it seems obvious that the processes must be 
 virtually standard and continuous, one operation fol- 
 lowing another in regular order and all of nearly routine 
 character, while the machinery must be largely auto- 
 matic. There must be no extraordinary demands upon 
 the intelligence or initiative of the foreman and his 
 duties must include little outside keeping his men at 
 work, preserving discipline, and seeing that materials or 
 parts in process move according to schedule. But in 
 factories where assembling operations form an impor- 
 tant part of the production cycle, the line organization 
 offers few advantages. It is in the assembling indus- 
 tries, indeed, that the functional system and the dif- 
 ferent systems of line-and-staff shop organization chiefly 
 have been developed. 
 
 The functional system is simply the application of the 
 principle of the division of labor to shop management, 
 while the line-and-staff may be considered either as a 
 
 90
 
 ORGANIZATION 
 
 modified form of the functional scheme or as a grafting 
 on the traditional line organization of functional prin- 
 ciples. In succeeding chapters we shall see this same 
 tendency toward the division of labor in management 
 revealing itself in the gradual division of the activities of 
 distribution along functional lines, and the unrecorded 
 and almost unnoticed development as such of functional 
 middlemen, lilve bankers, insurance underwriters, and 
 transportation and express companies. For the mo- 
 ment, however, we are concerned only with factory 
 activities. 
 
 The first functional division of production is a sepa- 
 ration of planning from performance, with the aim of 
 relieving the foremen of the duty of arranging and rout- 
 ing materials through the shops. A further functional 
 division then takes place within the planning depart- 
 ment. In the Taylor system of management, which 
 is a typical functional scheme, the work is divided 
 among an order-of-work clerk, an instruction-card man, 
 a time-and-cost clerk, and a shop disciplinarian. The 
 names of these foremen, for that is their real rank, in- 
 dicate the duties which each performs. Auxiliary to 
 the planning department are the time-and-motion- 
 study men, upon whose investigations, analyses, and 
 experiments are based the standards which govern all 
 the shop activities, whether these are ordered by the 
 planning room or actually applied under the eye of the 
 foreman in immediate charge of the operations. 
 
 The function of the order-of-work clerk is to plan the 
 route which each piece or lot of material shall take 
 through the shop. But his task is not simply to indi- 
 cate the sequence of the movements. He determines 
 what processes the material shall pass through and in 
 
 91
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 what order, and maps his conclusions on route sheets 
 which illustrate this order graphically and chronologi- 
 cally. With these charts before him, the instruction- 
 card man works out the details of each operation, even 
 to the particular machine and tool to use, the speed to 
 be observed, and the order in which the cuts are to be 
 made. All this from standards already set by prelim- 
 inary time-and-motion studies and kept at hand for 
 instant reference in the case of new or unfamiliar work. 
 
 The instruction cards have two purposes: they are 
 work orders for the proper functional foremen out in 
 the shop, and they also give exact directions for the 
 mechanic to follow in his work. After the latter has 
 finished the job, a return of the time consumed in both 
 labor and machine-hours goes to the time-and-cost 
 clerk, whose function it is to make up the pay roll, in- 
 cluding the bonuses earned, and to determine the cost 
 of each separate operation performed on the job. Lastly 
 there is the shop disciplinarian (frequently the head of 
 the planning room as well) to deal with any dispute or 
 insubordination which the other bosses are unable to 
 handle satisfactorily. 
 
 In direct contact with the men, the machines, and 
 the work in process are four other functional foremen, 
 the gang boss, the speed boss, the repair boss, and the 
 inspector. 
 
 The gang boss is more a teacher than an executive. 
 It is his part to interpret the planning room's instruc- 
 tions on unfamiliar work, to demonstrate if necessary 
 the standard way of carrying on the process as indi- 
 cated on the instruction card, and to continue this 
 demonstration and supervision until the workman is 
 able to turn out the job in the standard time., More- 
 
 92
 
 ORGANIZATION 
 
 over, on familiar or repeat operations in which the 
 workman is likely to lose interest and efficiency, it is 
 the duty of the gang boss to hold him to the efficiency 
 level where he will earn his own bonus, and allow the 
 gang boss to earn the premium gained when his group 
 makes more than a specified minimum of individual 
 bonuses. No need to say that the workman's bonus is 
 an imperative part of the functional system of man- 
 agement; without such an incentive to effort, the eflS- 
 ciency level could not be permanently maintained. 
 
 The function of the speed boss is to see that every 
 machine moves at the exact speed called for by the 
 instruction card. This indicated speed is the one which 
 has been shown to be the most effective in the tests and 
 experiments carried on when this particular operation 
 was submitted to time-and-motion studies and thus 
 standardized for all future repetitions. The business of 
 the repair boss, naturally, is to keep all machines, tools, 
 power transmissions, and the like in such perfect order 
 that processes can be performed in the standard times. 
 The inspector, last of the four shop bosses, has jurisdic- 
 tion over the manner of performing the different opera- 
 tions as well as over the finished results. The idea here 
 is to prevent mistakes rather than to penalize them by 
 rejecting parts or assemblies which fall below specifi- 
 cations. 
 
 In some plants, where modified forms of functional 
 management are in use, the work of the speed boss is 
 assumed by the gang boss or instructor. Sometimes, 
 too, the repair boss and inspector have diminished func- 
 tions and the gang boss is restored to much the same 
 authority as that which he holds under line organiza- 
 tion. It is here, indeed, that the line-and-staff system 
 
 93
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 emerges as a virtual compromise between line and func- 
 tional management. Keeping the line organization to 
 supervise operation, a staff of technical experts is added 
 to analyze every factor and detail in production — 
 materials, machines, methods, routing, arrangement, 
 and the like — to determine standards for each of these 
 and to secure the adoption of these standards by the 
 officers of the regular line organization. 
 
 In every sense of the word, these specialists are func- 
 tional foremen who do not come into direct contact 
 with the working force except when they are making 
 their time-and-motion studies of processes, or demon- 
 strating the standard methods of performing an op- 
 eration. The greatest advantage of the line-and-staff 
 system, perhaps, is that a beginning can be made in 
 the betterment of shop practice without disturbing the 
 regular factory organization and that the betterments 
 can be carried forward at whatever rate the manager 
 deems advisable. In this it differs from functional 
 management, the advocates of which declare that it 
 must be adopted or rejected as a whole. Both systems 
 are as one in insisting on the bonus method of wage 
 payment, a factor in efficiency which was discussed at 
 some length in the previous chapter. 
 
 To the business man approaching the subject of func- 
 tional management from the outside, the apparent 
 amount of clerical work required, the opportunities 
 offered for mistakes and for the shifting of responsi- 
 bility, the amount of preparatory work demanded, and 
 the long interval before marked results can be expected 
 should not be absolute deterrents. Functional man- 
 agement, unabridged, may be beyond the present scale 
 or needs of his business as he is conducting it or pro-
 
 ORGANIZATION 
 
 posing to start it. But the principle behind both the 
 functional and the line-and-staff systems of manage- 
 ment — the finding and using of the " one right way " 
 under present conditions of doing things — cannot be 
 ignored if his business is to meet competition grounded 
 on this fundamental of efficiency. 
 
 In this discussion of production little attention has 
 been given to strictly intra-departmental problems. 
 As suggested in earlier chapters, plant and operating 
 activities are relatively well organized, and technically 
 trained factory experts are available to carry the burden 
 of these internal problems. The purpose, indeed, has 
 been less to outline the policies of production and the 
 principles on which they are based than to suggest in a 
 known and standardized field a logical classification of 
 the activities of business and to illustrate how a syste- 
 matic approach may be applied to the problems dis- 
 closed as the classification develops. 
 
 95
 
 PART II 
 THE PROBLEMS OF DISTRIBUTION
 
 CHAPTER VIII 
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 WHEN economic life was organized on a basis of 
 barter, both traders inspected and took over 
 the actual goods involved before the transaction was 
 complete. One or the other carried his surplus of 
 food or skins or weapons to the point where the second 
 man had need of it and a store of some desirable thing 
 to exchange for it. Or both transported their merchan- 
 dise to a recognized market place where the bargain was 
 made and the transfer effected. For a long time, indeed, 
 after the first broad division of labor along functional 
 lines had brought the merchant into existence and the 
 evolution of a medium of exchange or " money econ- 
 omy " had displaced barter, the buyer had personal con- 
 tact with his purchases at all stages of the transaction 
 and viewed, handled, and tested them before assuming 
 possession. 
 
 This was distribution in its simple phase — literally, 
 the application of motion to change the place and owner- 
 ship of material which production had already changed 
 in forvi. The two chief functions of marketing, de- 
 mand creation and physical supply of the merchandise, 
 were all but performed simultaneously. There was 
 no necessity of distinguishing between them or of con- 
 sidering their activities separately, since the sale proper 
 and the delivery of the goods merged into one operation. 
 
 Development of the range and means of distribution 
 
 99
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 however, has enlarged the origmal scope of these two 
 functions. Each has given rise to a distinct group of 
 related activities, with corresponding problems for the 
 man who must coordinate and control them. The 
 primitive producer solved all problems when he con- 
 veyed his cattle or corn or tanned skins to the market 
 place. For the most part, his goods were common util- 
 ities, mere sight of which stirred desire for ownership, 
 even among those already adequately supplied. Under 
 normal conditions, therefore, he was sure of a market 
 at the prevailing market price. If this were unsatis- 
 factory, he could take his merchandise home again and 
 wait for a more favorable occasion. 
 
 The problem of distribution to-day is more complex. 
 Broadly speaking, it divides into two sub-problems 
 closely related and interdependent but each having to 
 do with a different set of factors and reactions. The 
 first takes shape in the question: Given a particular 
 article, how can a demand for it be created of sufficient 
 volume to make its production and distribution projBt- 
 able.^* The second is: Through what channels can the 
 article itself be conveyed from the factory warehouse, 
 where it is of least value, into the hands of those con- 
 sumers who will pay the most profitable price for it, 
 though this price may not be the highest at which a 
 more limited volume could be sold.'* 
 
 The activities of demand creation focus on the con- 
 sumer. Their purpose is to communicate to his mind 
 such ideas about the product as will arouse desire for it 
 and cultivate willingness to pay the price and make the 
 effort required to secure possession of it. This aroused 
 demand would have no commercial or economic value, 
 however, unless provision were made for satisfying it by 
 
 100
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 actual transfer of the goods to the consumer through 
 one or more of the agencies available. 
 
 The relations between the activities of demand crea- 
 tion and of physical supply, in fact, illustrate again the 
 persistence of the two principles of interdependence and 
 of balance. Failure to coordinate any one of these ac- 
 tivities with its group-fellows and also with those in 
 the other group, or undue emphasis or outlay put upon 
 any one of these activities, is certain to upset the equi- 
 librium of forces which means efficient distribution. 
 Nor must it be forgotten, in organizing these activities 
 and establishing the policies which shall guide them, 
 that both groups present situations which involve the 
 production activities discussed in previous chapters and 
 the activities of administration to be taken up in an- 
 other section of this book. 
 
 In approaching the problems of distribution, greater 
 consideration must be given by the manager to strictly 
 departmental policies than was necessary in dealing 
 with production. In the first place, the standardization 
 of processes and materials, which has been carried to 
 such length in the factory, has hardly touched the ma- 
 jor operations of distribution. Relatively little progress 
 has been made, for instance, in observing and compiling 
 the essential facts about demand creation, in classifying 
 them, coordinating them, and establishing their mutual 
 relations; and in tracing and defining the broad tend- 
 encies and principles which analysis of a sufficient num- 
 ber of cases would disclose. 
 
 It is true that many intensive studies have been made 
 of various functions, processes, and reactions concerned 
 with distribution. But these investigations generally 
 have been made from the viewpoint of a single business 
 
 101
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 or a group of similar businesses. They have concen- 
 trated in the main on promising territories and avoided 
 those where the chances were unfavorable. And as a 
 consequence, the results are so subject to qualification 
 that few significant rules capable of general application 
 have been formulated. Promising beginnings have been 
 made — the investigation of retail selling by the Bureau 
 of Business Research at Harvard University, the exper- 
 iments conducted by the schools of commerce at various 
 universities, the constant gathering, sifting, and organ- 
 izing of practical methods by business magazines, the 
 trade press and trade associations. By comparison with 
 the work yet to be done, however, they are only begin- 
 nings, trial efforts valuable chiefly because they point 
 the way and demonstrate the feasibility of an adequate, 
 organized study, perhaps under the national govern- 
 ment's leadership, of all the activities of business. 
 
 The present lack of standards is due largely to the 
 diflSculty of analyzing a structure so intricate as our 
 system or distribution, complicated further by geo- 
 graphical, seasonal, and other physical conditions, and 
 by the presence of the changing human factor in every 
 individual transaction. Consider, for a moment, how 
 many and various are the elements presented by the 
 United States as a consuming market. Here are about 
 one hundred million people distributed over an area 
 of more than three million square miles. Some are 
 gathered in the large cities, where millions jostle elbows. 
 Some are scattered over great areas with considerable 
 distances between them and their neighbors. Some 
 daily pass hundreds of retail stores; some must ride 
 miles to reach the nearest store. Wide extremes in pur- 
 chasing power exist. Millions have a purchasing power 
 
 102
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 barely sufficient to obtain for themselves the necessi- 
 ties of life. A few can satisfy the most extravagant 
 whims of the imagination. Between these extremes lie 
 all degrees of purchasing power, the number in each 
 class becoming greater as you descend in the scale of 
 purchasing power. 
 
 Their desires are as varied as their purchasing power. 
 Besides the great colonies of immigrant folk in every 
 large city — Poles, Bohemians, Hungarians, Italians, 
 Scandinavians, and Russian Jews, each race with trans- 
 planted tastes and standards and a language barrier to 
 make more difficult the approach — the native-born 
 consumers present a complex problem, with environ- 
 ment, education, social customs, individual habits, and 
 all the variations of body and mind tending to make 
 their wants diverse. 
 
 In each individual certain conscious needs are con- 
 stantly gratified by the purchase of goods produced for 
 such gratification. Then there are the conscious needs 
 which go ungratified because of the limitations upon 
 buying power and the necessity of satisfying other needs 
 of greater felt importance. And finally, there are the 
 unrecognized needs which fail of expression because the 
 individual is ignorant of the existence of goods which 
 would gratify them. Twenty years ago, to illustrate, 
 there existed in the farmer, far from a barber shop and 
 clumsy of touch, an unformulated need for a safety 
 razor. To-day a score of manufacturers bring to his 
 attention the existence of such devices and the recog- 
 nized need finds expression in efiFective demand. 
 
 The existing system of distribution was built up along 
 the line of least resistance, aiming always at the satis- 
 fying of known wants. As suggested in a previous 
 
 103
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 chapter, the capacity of the market to absorb goods 
 has generally exceeded the ability of manufacturers 
 to produce them. This at least was true from the 
 introduction of power-driven machinery into English 
 industry until the closing decades of the nineteenth 
 century. Pressure of demand made it unnecessary for 
 the business man to devote his time to searching out 
 unformulated needs. Only in recent years, when the 
 development of production has potentially outstripped 
 the available market and shifted the emphasis to distri- 
 bution, has the manufacturer-merchant become a pio- 
 neer on the frontier of human desires and needs. 
 
 To-day the progressive business man makes careful, 
 intensive studies not merely of the consumer's recog- 
 nized wants but of his tastes, his habits, his tendencies 
 in all the common activities and relations of life. This 
 he does in order to track down unconscious needs, to 
 manufacture goods to satisfy them, to bring these 
 products to the attention of the consumer in the most 
 appealing ways, and finally to complete the cycle by 
 transporting the goods to him in response to an expressed 
 demand. His problem is chiefly one of adjustment. He 
 must bend the materials and forces of nature to the end 
 of human service. And, most diflScult task of all, he 
 must shape his making and selling policies alike to sat- 
 isfy contradictory conditions and methods and to em- 
 ploy without waste the divergent and overlapping agen- 
 cies through which present-day distribution is carried on. 
 In marketing identical products and appealing to 
 identical classes of consumers, for instance, no one 
 method of sale is recognized as the right or the most 
 profitable way. Sale by bulk, sale by sample, and sale by 
 description are often carried on for the same commod- 
 
 104
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 ities, at the same time, by the same organizations. Not 
 from choice, certainly, nor from lack of thought put 
 upon the question of simplifying the problem of creat- 
 ing demand and supplying it. But rather because, with 
 a few exceptions, the gathering of the necessary informa- 
 tion, the analyzing of the data secured, and the building 
 up of a standard method of marketing which would 
 satisfy all the conditions, is a task beyond the resources 
 of any but the greatest of our businesses. 
 
 Sale by bulk goes back to the first dim beginnings of 
 trade — the barter of a rude stone ax, perhaps, for an 
 ill-tanned deerskin beside the tribal campfire. It per- 
 sists in all the stages of advancement and occasional 
 retrogression through which commerce and civilization 
 have come together — when buyers and sellers met 
 personally at the market places or fairs, when the trav- 
 eling artisan went to the home of his customer and there 
 constructed the cart or chair or coat of which the latter 
 stood in need; when shops were set up in the towns and 
 the situation was reversed, the customer seeking the 
 seller. The significant factor in the process was the 
 same in all cases, however. There was personal contact 
 and the buyer saw the actual goods which he was asked 
 to purchase. 
 
 Later came sale by sample, where the customer ex- 
 amined and judged, not the actual goods to be re- 
 ceived, but a sample which the merchant guaranteed to 
 be similar in type and equal in quality and in every 
 other essential respect to the product which he would 
 deliver. The development of this method of sale was 
 dependent on progress in several apparently unrelated 
 directions. New manufacturing methods made possi- 
 ble a standardization of product as the market widened. 
 
 105
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 It was possible to produce substantially identical arti- 
 cles in large quantities. Grades and standards thus 
 could be established. At the same time there came into 
 being the higher code of commercial ethics necessary 
 to the extension of sale by sample. For the purchaser 
 must have confidence not only in the ability of the 
 producer to furnish goods identical with the sample, 
 but also in his intention to do so. 
 
 In sale by description, the purchaser does not see 
 even a sample of the goods. Instead, ideas about the 
 goods are communicated to him by the distributer 
 through the use of salesmen (his own or those of mid- 
 dlemen), advertising, catalogues, booklets, or letters. 
 Spoken, WTitten, or printed symbols take the place of 
 the sight of the goods themselves or a sample of them. 
 The use of the term " sj^mbols " rather than " words " 
 is necessitated hy the fact that photographs and draw- 
 ings to-day are important factors in sale by description. 
 For a picture of the commodity is frequently able to 
 convey to a prospective purchaser a more vivid im- 
 pression of the product than could pages of verbal 
 description. In the case of complicated machinery, in- 
 deed, the blueprint, with exact dimensions set down 
 and the exact relations between parts made plain, is 
 used to supplement and make clearer the conception 
 which the engineer purchaser acquires by examination 
 of the machine itself. 
 
 Sale by description demands a still higher plane of 
 business conduct than is required for sale by sample as 
 well as a higher level of general intelligence. The pur- 
 chaser must have enough of understanding and imag- 
 ination to grasp ideas conveyed either through spoken, 
 written, or printed symbols and to visualize the article 
 
 106
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 described and its projected effect on his own business or 
 pleasure. In the backward community, where educa- 
 tion is at a low level, sale in bulk and sale by sample 
 must continue as the chief method of distribution. In 
 a sense, therefore, sale by description, whatever its be- 
 ginnings, is in its modern application a by-product of 
 the public school and of the printing press. 
 
 At this stage of progress in selling methods, too, the 
 distinction between the demand-creating and the de- 
 mand-supplying functions of the distributor becomes 
 obvious. Barter is largely the matching of wits and 
 commodities by the two participants; demand creation 
 and actual supply are merged in the minds of both. 
 But in sale by sample and sale by description the line 
 of division is clear. The functions are differentiated 
 and can readily be distinguished. 
 
 The ideas to be conveyed to the prospective pur- 
 chaser in sale by description are such as will awaken 
 an effective desire for the product thus exploited. The 
 arousing of desire is the essential element in selling, 
 though the distributor has the further task of providing 
 machinery for the meeting of the resultant demand. 
 He must make his goods physically available to the 
 buyer. In sale by bulk, this activity merges with sell- 
 ing, since the goods are physically present when the 
 sale is made. In sale by sample and sale by description, 
 the physical distribution of the goods is a problem dis- 
 tinct from the creation of demand, though it is one 
 which must be considered at every step in any solution 
 arrived at. 
 
 All three methods of sale, however, are in use in 
 modern trading. The consumer still purchases a large 
 part of the commodities he uses under conditions which 
 
 107
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 make them sales in bulk. He views and inspects the 
 goods in a retail store before he buys them, though 
 thousands of farmers and small-town folk depend for 
 their supplies on mail-order concerns, which sell by de- 
 scription. The middleman, buying in larger quantities, 
 generally purchases from samples, often manufactured 
 singly or in small lots long before the season of sale to 
 consumers begins. This is the rule also with products 
 classed as specialties — vacuum cleaners, cash registers, 
 coJBFee percolators, and a thousand other products which 
 need a demonstration to clinch the prospective's desire 
 to possess them for the sake of utility value. To be 
 sure, this difficulty has been solved in part by the exten- 
 sion of the approval system. But sale by description 
 becomes each year of increasing importance in every 
 phase of distribution. Even where the purchaser ex- 
 amines a sample or the merchandise itself before the 
 sale is concluded, the earlier steps in the creation of in- 
 terest and desire probably have been by means of 
 description. 
 
 In marketing motor cars, for instance, sale by sample 
 is the method most generally used. Demonstration cars 
 are shown at the agencies and at the annual automobile 
 shows held throughout the country — modern equiva- 
 lents in a single line of the medieval fairs — and orders 
 are taken for future deliveries. Early in each successive 
 season orders are frequently booked, however, before 
 the new model has been completed, the customer buying 
 on the designer's specifications for both chassis and 
 body, backed up of course by the manufacturer's repu- 
 tation for turning out efficient and beautiful machines. 
 This is nothing more or less than sale by description, 
 since extensive magazine and newspaper advertising 
 
 108
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 campaigns (sale by description, again) contribute not 
 a little to the prestige of the leading cars and unques- 
 tionably stir the desire to possess a motor, which is the 
 genesis of every purchase made. Finally, when the 
 driving season is under way and the agency is stocked 
 up with machines manufactured and delivered during 
 the winter and early spring, hundreds of buyers give 
 their orders after examining the cars and participating 
 in a road test of the identical cars which are to be deliv- 
 ered to them. It may fairly be said, therefore, that the 
 automobile industry, barely twenty years old as yet, 
 uses all the methods of sale which have been developed 
 through many centuries in the marketing of staples. 
 
 This evolution in the mechanics of distribution is 
 graphically shown in Chart I (page 165). Though the 
 diagram summarizes the changes which have taken 
 place in marketing methods, it does not tell the whole 
 story. The forces behind the advances made are not 
 to be overlooked. The higher standard of business 
 ethics — what Veblen has called " mitigations of the 
 maxim. Caveat emptor " ^ — has meant a growth in 
 public confidence. The factory system and machine 
 processes have made possible a high degree of stand- 
 ardization, distinct differentiation of quality, and a 
 volume of production which has lowered factory costs 
 and thus brought innumerable articles within reach of 
 wider circles of consumers. 
 
 Most important, perhaps, have been the improve- 
 ments in the means of communication. The develop- 
 ment of the steamship, the railroad, the telegraph, the 
 telephone, and the post office have opened virtually 
 an unlimited market to the individual business man. 
 
 ^ Thorstein Veblen, " The Theory of Business Enterprise." 
 109
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 And this has meant further division of labor in market- 
 ing and further differentiation of functions. In fact, 
 speciahzed institutions have grown up, concerned with 
 finance, transportation, insurance, and other sales func- 
 tions originally assumed by the merchant producer 
 himself, but now surrendered to groups of what may be 
 termed functional middlemen. 
 
 Broadly, then, the problem of distribution is to bring 
 about an effective adjustment between demand crea- 
 tion and economical supply, to arouse the desired maxi- 
 mum of demand at a minimum of expense, and to supply 
 without leakage the largest possible percentage of this 
 demand. The second phase of the problem involves 
 the elements of time, convenience, and service. If the 
 demand which has been aroused among consumers is 
 to be fully utilized, it must be possible for them to ob- 
 tain the goods promptly and without undue effort at the 
 moment when the demand shows itself. In many cases 
 certain collateral services must be provided, such as 
 instruction in their use, subsequent repairs, and the like. 
 If the consumer's interest and desire to try a certain 
 food product have been stimulated through an adver- 
 tising campaign, but the product itself is not to be had 
 at a convenient grocery store when the buying impulse 
 is at its strongest, the resulting leakage of demand, if it 
 is at all general, is bound to defeat the manufacturer's 
 purpose. 
 
 Not a few costly failures in distribution campaigns 
 have been due to such a lack of coordination between 
 demand creation and physical supply, to the producer's 
 neglect either to provide stocks of his commodity at 
 points easy of access to the consumer or to enlist to the 
 same end the interest and the cooperation of the mid- 
 110
 
 DEMAND CREATION AND PHYSICAL SUPPLY 
 
 dlemen who are the usual sources of supply. Instead 
 of being a subsequent problem, this question of supply 
 must be met and answered before the work of demand 
 creation begins; otherwise the leakage may endanger 
 the result. Middlemen do not cover all fields efficiently. 
 The branch house or store is an expensive undertaking. 
 An exclusive sales force means a constant outlay. Mail 
 orders and direct shipments are effective only in distrib- 
 uting certain kinds of merchandise and in reaching 
 certain sections and classes of consumers. 
 
 The supply plan adopted may use only one of these 
 agencies; it may employ all. In any case, it must cor- 
 respond with and serve the districts, the social strata, or 
 the special classes of consumers at which the selling 
 appeals are directed. It follows, as a corollary, that the 
 demand-creation campaign must likewise be shaped to 
 fit the available facilities of demand supply — another 
 outcropping of the principles of interdependence and 
 of balance. Otherwise there is waste through over- 
 stimulation of demand where no adequate or economi- 
 cal means have been provided to satisfy it, and this 
 means a proportionate increase in the burden of dis- 
 tribution expense. 
 
 The consumer pays, of course. Yet it is neither sound 
 business nor ethical business that the price of any com- 
 modity or service should be loaded with the cost of 
 duplicate or unnecessary functions or of any other pre- 
 ventable leakage or waste. Stripped down to essentials, 
 a business succeeds only as it serves. Unless it adds to 
 the sum of human happiness or comfort or progress 
 something which no other activity or agent can supply 
 so cheaply or so well, its end is forecast; a more efficient 
 competitor is building to take its place. 
 
 Ill
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 On the other hand, each access of value, each price 
 reduction made feasible by the elimination of lost mo- 
 tions, widens the market of the article, brings it another 
 degree nearer the status of a necessity for daily use or 
 consumption by the multitude. Forty years ago — to 
 take one instance in a thousand offering — a woven 
 wire mattress ranked as a luxury, costing fifteen dollars 
 at wholesale, ten days' pay for the man who constructed 
 it. Now a better mattress, machine made, is sold to 
 dealers for eighty-five cents and no day laborer can 
 find a cheaper substitute. Similar achievements in 
 organization, in invention, in ceaseless pruning of ex- 
 pense and adaptation of materials, have driven soaps, 
 collars, cottons, kerosenes, books, furniture, and an 
 endless list of other things down from the plane of semi- 
 luxuries to facilitate and enrich our daily life. 
 
 112
 
 SECTION I 
 DEMAND CREATION
 
 CHAPTER IX 
 
 LOCATION, CONSTRUCTION, AND EQUIPMENT 
 OF THE PLANT 
 
 ALL the activities of distribution, we have seen, 
 have to do \sdth the two functions of demand 
 creation and physical supply. If we apply to each 
 group the same systematic approach used in our anal- 
 ysis of production, an interesting parallel is developed. 
 In both groups, we find, there are plant and operating 
 policies to be determined. Further, the plant policies 
 are governed, as in production, by the factors involved 
 in location, construction, and equipment; while the 
 operating policies are shaped by considerations bearing 
 on materials, agencies (grouped labor), and organiza- 
 tion. This is true broadly of both demand creation and 
 physical supply. The parallel may not extend to all the 
 factors involved in either case. Our knowledge of distri- 
 bution does not carry far enough to say definitely what 
 are all the factors. But at least the analysis will be 
 suflBciently complete to give direction to this discussion. 
 
 The problem of demand creation may be stated sim- 
 ply: How shall we transmit to the possible consumer 
 such ideas about the goods we have to sell as will arouse 
 a maximum demand for them at a minimum expense, 
 present and future sales both being considered .^^ 
 
 Starting with the plant activities, as in our analysis of 
 production, we find that we have to deal with the loca- 
 tion, construction, and equipment of the " plant " for 
 
 115
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 demand creation — in other words, the quarters for 
 the sales department and such other departments as 
 contribute directly or indirectly to the work of de- 
 mand creation. Like all other general business prob- 
 lems, that of location proves to be a complex of 
 several sub-problems. At least four simple solutions 
 are found to be common practice: the sales plant 
 may be at the factory, at the largest neighboring city, 
 at the largest trade market, at the chief city of the 
 country. Still another recognized plan combines a 
 central sales plant at one of these locations, with auxil- 
 iary district or branch organizations, each of which 
 concentrates on the peculiar distribution problems of 
 its own territory. Quite as successful campaigns have 
 been conducted from one location as from another; off- 
 hand choice of any one of them, therefore, without study 
 of the comparative advantages offered by the others, 
 would be shortsighted and possibly ineffectual. For 
 there are definite factors governing the placing of the 
 sales departments which should be recognized and 
 weighed by the management before a decision is made. 
 
 To reduce the problem to concrete form, let us con- 
 sider the location of the selling plant of a company mak- 
 ing furniture in a small town in southern Michigan. 
 Taking the first step in our systematic approach and 
 eliminating, or at least recognizing, the personal equa- 
 tion, it is obvious that the placing of the sales depart- 
 ment should be influenced principally by the marketing 
 needs of the business. 
 
 Personal conveniences should not dictate its location 
 at the factory; nor should prejudices or prepossessions 
 be allowed to weight the decision. The ambition of the 
 owner's wife or family to exchange life in a country 
 
 116
 
 LOCATION,CONSTRUCTION, AND EQUIPIVIENT 
 
 town for the greater social opportunities of New York 
 or Chicago should count for no more in fixing the loca- 
 tion at either point than should personal associations or 
 the traditions of the trade tip the scale in favor of Grand 
 Rapids. True, man does not live by business alone, and 
 personal considerations frequently supply valid reasons 
 for preferring one location to another. Even here, how- 
 ever, the manager who recognizes the personal element, 
 when it enters into his handling of a question, is in a 
 fair way to eliminate it or take precautions to compen- 
 sate its effects. If he yields to interests entirely foreign 
 to his business, at least he can measure the cost of his 
 yielding in definite terms. 
 
 Breaking up his problem into its constituent prob- 
 lems — the second step in the approach — he discovers 
 that in the placing of his sales " plant " he must con- 
 sider not only its relation to the activities of demand 
 creation, but also to the physical supplying of the con- 
 suming market and to the activities of production and 
 administration. There are innumerable sub-problems 
 arising out of the interplay of these various activities 
 and the necessity of establishing policies for their coor- 
 dination and direction. 
 
 Consider, in the first place, the influence of demand 
 upon the product, an influence which is felt by all spe- 
 cialties and by not a few established staples. Styles are 
 an essential factor in the furniture business; the sales 
 department which keeps in constant touch with middle- 
 men and consumers can sense the changing tendencies 
 of demand and transmit this information to the produc- 
 tion department. Under normal conditions, the more 
 frequent and intimate this contact, the smaller will be 
 the chance of mistaking the current drift and of manu- 
 
 117
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 facturing a stock which does not coincide with the pub- 
 lic's desire for chairs and tables and dressers in one or 
 another " period " style. From this viewpoint, a selling 
 plant at Chicago will have advantages over one in New 
 York; while location at Grand Rapids will further lessen 
 the danger of launching models in which retailers will put 
 none of the faith that is expressed in orders and which 
 housewives, therefore, will encounter only in post-season 
 clearance sales. For Grand Rapids is the seat of the 
 seasonal furniture expositions which virtually establish 
 the styles for the general trade. Would it not be pos- 
 sible, then, to place the permanent selling plant at the 
 factory and display samples and keep representatives at 
 Grand Rapids during the semi-annual exposition periods 
 in January and July, when nearly all the initial buy- 
 ing by the trade is done.f^ Few dealers come to market 
 at other times, except to conduct hotel men or other 
 large private buyers who have special contracts to 
 place. If a manufacturer's line and prices are likely to 
 appeal to such contract buyers it might be profitable 
 to keep a permanent display room at Grand Rapids. 
 
 From the viewpoint of administration, which must 
 also be considered, if the balance of cost, quality, and 
 service is to be maintained, the location should be tested 
 on the basis of expense, and of effect on organization. A 
 location at the plant would probably mean less outlay 
 than one at Grand Rapids, Chicago, or New York; it is 
 the part of the manager to determine whether the saving 
 more than compensates for the possible selling advan- 
 tages sacrificed. 
 
 Or considering the problem of organization, would it 
 be more effective to have the salesmen focus at the fac- 
 tory, at the principal market, or at one of the two chief 
 
 118
 
 LOCATION,CONSTRUCTION, AND EQUIPIVIENT 
 
 cities? If the selling force were handled by districts or 
 if the sales department were at Grand Rapids, Chicago, 
 or New York, could not semi-annual conventions be 
 held at the factory to familiarize the salesmen with all 
 the new numbers in the line, changes in style, mate- 
 rials, construction and finish, and similar new selling 
 arguments, thus also at the same time giving the de- 
 signers and foremen at the factory the benefit of the 
 salesmen's knowledge of what dealers and ultimate users 
 seem to favor or dislike? Information of this sort can 
 be gathered in other and more stable lines through sales- 
 men's daily reports and inspection trips by the sales- 
 manager or by special investigators. Many successful 
 manufacturers of specialties, like adding machines, cash 
 registers, and typewriters, supplement all their other 
 means of gathering information with factory conven- 
 tions of the sales force at which certain sessions are de- 
 voted entirely to discussion of existing weaknesses and 
 possible betterments in the machines and of the future 
 service needs of users. 
 
 The effect of location on the sales department and 
 its allied departments cannot safely be ignored. The 
 practical business man frequently must adapt his poli- 
 cies and methods to the temperaments and personal 
 needs of his valuable men. With the factory near 
 Grand Rapids, the center of the furniture district, men 
 work and live in a furniture environment. Furniture 
 is the absorbing interest, not merely of one community 
 but of several; competition is intense, and the result is 
 a stimulation of ideas and a constant interplay of opin- 
 ion which is clearly lacking in great general markets like 
 Chicago and New York. Such an atmosphere naturally 
 makes for enthusiasm and efficiency; the personnel of 
 
 119
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 the sales office feels it all the time; the members of the 
 field force feel it at each visit. 
 
 On the other hand, Chicago and New York are im- 
 portant centers of the trade. The competitive element 
 is not lacking. The standards set by the exclusive shops 
 and great retail stores which buy abroad as well as at 
 home are high. Advertising, perhaps, is on a plane of 
 greater efficiency. The interchange of information and 
 the discussion of sales problems in association circles is 
 more free and the experience of one line is more acces- 
 sible for adaptation by another. The cities, too, are 
 magnets for men of initiative and capacity; able sales- 
 men, or the raw materials of salesmen, are more easily 
 secured. Here, however, the angle of expense must not 
 be overlooked; higher salaries must be paid as a rule 
 to men whose homes are in the city. 
 
 A further factor to be considered is the effect of the 
 manufacturing organization's opinion and moral sup- 
 port on the temper and effort of both sales department 
 and selling force. With the sales plant at the factory to 
 maintain this contact without break, to represent the 
 field men there and in turn to interpret the production 
 departments to the salesmen, it is less difficult to culti- 
 vate a company spirit and solidarity which will react 
 on every man enployed and bring about an unusual 
 degree of cooperation and mutual interest. With cer- 
 tain of the larger industrial organizations, this con- 
 tact has become an important factor in both sales and 
 production. More than one agent has been carried 
 across the last barrier separating him from an order of 
 consequence by the knowledge that the factory flag 
 will be raised in his honor and that a thousand company 
 workmen will comment appreciatively on his exploit. 
 
 120
 
 LOCATION,CONSTRUCTION, AND EQUIPMENT 
 
 Workmen, too, feel the thrill of accomplishment when 
 new production records for the day or week or month are 
 celebrated in a special letter to the field force. Recently, 
 for instance, the executive officers and selling plant of a 
 big specialty house were moved from a mid-western 
 town to New York to accommodate the personal desire 
 of the president, who insisted on keeping in close touch 
 with all sales activities. A significant falling off in 
 orders, an increase in points of friction between the 
 making and sales departments, and an obvious slump 
 in the organization spirit during the ensuing year con- 
 vinced him that he had made a mistake in policy; and 
 the general offices and selling plant were hurried back 
 to their old quarters across the court from the as- 
 sembling departments. 
 
 Another vital sub-problem in location harks back to 
 the relations between the sales and production depart- 
 ments. If the selling policies are fixed at a distance 
 from the factory, there is the increased possibility of 
 tangential development on the part of both, which we 
 saw as a danger in earlier chapters. Separated, there is 
 a strong chance that the two forces will drift apart. 
 The factory is likely to plan a product which will 
 occasion the least trouble or expense in process. Pro- 
 duction considerations may govern, such as the length 
 and width of lumber in stock or offered by supply 
 houses, the gauge or quality of steel which is easiest or 
 cheapest to buy, the finish most readily applied, and so 
 on, without particular regard for demand as expressed 
 by the sales department. 
 
 The factory's emphasis is nearly always on cost; the 
 sales department is concerned with service and the im- 
 pression, first and last, which the product makes on the 
 
 121
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 customer. The requirements of the market, as regis- 
 tered by the sales department (within the limits set by 
 the principle of balance, of course) should be satisfied by 
 the factory. Otherwise the burden on the selling force 
 is too heavy and the cost of closing orders likely to be 
 out of all proportion. On the other hand, the sales de- 
 partment must not misrepresent the product; it must 
 sell what the factory is able to deliver. 
 
 With the sales plant at the works, this danger of tan- 
 gential development is minimized; only blind obsti- 
 nacy could prevent the sales and production managers 
 from getting together and reconciling any differences. 
 Where the factory is near the principal market, the situ- 
 ation is still more advantageous, since the plant for de- 
 mand creation can at the same time keep its contact 
 with the trade and with production. Where the factory 
 must of necessity be located at a considerable distance 
 from the market, as are the cotton mills of the South, 
 for instance, the coordinating function of the general 
 manager is called into play to compensate for the ob- 
 vious difficulty in maintaining contact. As a matter of 
 fact, the character of cotton cloth produced in the South 
 simplifies the question of demand creation. Moreover, 
 proximity to raw materials, and equable climate, and in 
 some cases lower wages, when translated into smaller 
 manufacturing costs, give a price margin which equal- 
 izes the marketing disadvantages. 
 
 Other constituent problems enter into this general 
 problem of where to place the plant for demand cre- 
 ation : the sales prestige of a big city location, its supe- 
 rior facilities for the preparation of advertising both 
 general and direct, its transportation advantages, and 
 the like. There are also local sub-problems which must 
 
 122
 
 LOCATION,CONSTRUCTION, AND EQUIPIVIENT 
 
 be settled: whether or not it is better, for example, to 
 maintain the sales department in the high-rent central 
 district, on a fashionable avenue, or on the Millinery 
 or Motor or Machinery " Row " where your trade natu- 
 rally congregates ; whether the saving in rent due to the 
 separation of your production and demand-creation 
 plants will cancel the loss in efficiency and mutual in- 
 spiration; whether ease of access for the customer is an 
 essential in sales departments (as in the case of millinery 
 houses, which seem to consider a central location abso- 
 lutely necessary), and so on through a series of detail 
 problems which will vary with the nature of the business 
 and with the physical factors of distance, transportation, 
 and quarters available. 
 
 It is interesting to note contradictory tendencies in 
 different businesses.^ There is, for instance, the typical 
 concentration of manufacturers of women's garments 
 on Fifth Avenue, New York, with the obvious end of 
 capitalizing the selling prestige such a location gives. 
 But, contrariwise, a great mail-order merchandise house 
 removed the sales department, general offices, and stock 
 rooms (the latter used to impress visitors and forming, 
 therefore, part of the plant for demand creation) from a 
 " show " building in the heart of a western city to a 
 dingy neighborhood much less accessible to the out-of- 
 town customers whose visits were desired. Here ad- 
 ministrative considerations commanded the change. 
 The company's building was being outgrown, trucking 
 costs were high, a prompt service of shipments was diffi- 
 cult to maintain, and the fixed investment, as measured 
 by the constantly increasing market value of the ground 
 
 ^ See the discussion in F. W. Taussig, " Principles of Economics," Vol, 
 II, pages 78-80. 
 
 128
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and building, was out of balance with the business done. 
 New warehouses, with direct railroad connections, were 
 indispensable; the entire establishment, therefore, was 
 transferred to the river front, where rail and water trans- 
 portation met, and the sales value of the central loca- 
 tion was sacrificed to the more vital considerations of 
 cost and service. Analysis of these sub-problems is 
 necessary, of course, in the placing of an actual selling 
 plant. The purpose of this chapter, however, is merely 
 to illustrate how a systematic approach can be applied 
 to a business problem rather than to pursue a typical 
 case through all its phases and conceivable variations. 
 
 Completing this analysis (in an actual case) , the next 
 process is to assemble all the relevant factors, listing 
 them if their number or complex character renders them 
 difl5cult to keep track of and giving each its approxi- 
 mate value in a scale of comparison. Many of these 
 were touched upon in the discussion of factory location 
 in the previous section, since the interdependence of the 
 activities of production and distribution is so close as to 
 make it almost impossible and, in fact, inadvisable to 
 fix a guiding policy in either field without first weighing 
 its probable effects in the other. 
 
 There follows, then, the final step, the adoption of a 
 fresh point of view. That the majority of furniture sales 
 departments are at the factory, for instance, is not con- 
 clusive evidence that this is the best policy. It may be 
 due to a tradition which has persisted without adequate 
 reason. A man with sufficient courage and initiative 
 to challenge the tradition and marshal the factors for 
 and against such a location may find that this trade 
 usage perpetuates the best balance between cost and 
 efficiency in the creation of demand; or under certain 
 
 124
 
 LOCATION,CONSTRUCTION, AND EQUIPMENT 
 
 circumstances and in certain lines he may discover that 
 the advantage lies with separation. 
 
 If it were the purpose of this book to indicate general 
 policies, broadly applicable, I might be tempted to write 
 that the weight of opinion and experience favors the 
 location of the selling plant at the factory. For special- 
 ties, at least, this would seem true, since a specialty is 
 the result of closer adaptation of a product to the needs 
 or the unformulated desires of the consumer, and the 
 more constant and intimate the relations between the 
 department which arouses or interprets the market's 
 demands and that which manufactures the commodities 
 to satisfy them, the more complete should be the co- 
 ordination of their efforts and the more successful the 
 final outcome. 
 
 Compared with the problem of the location of the 
 plant for demand creation, the problem of construc- 
 tion, the housing of the plant, is much less complex, 
 though still made up of several constituent problems. 
 When it comes to expressing his business in terms 
 of architecture, the feeling of the average successful 
 man is all for achieving as much of dignity, beauty, 
 and originality as utility and his financial resources will 
 permit. So strong is this feeling that it amounts to a 
 personal bias which frequently has handicapped the 
 growth of the business by diverting money needed for 
 more essential uses and given rise as often to the cyn- 
 ical prediction that the builder was erecting a monu- 
 ment for his undertaking. Approaching the question, 
 then, the prudent manager will recognize this very 
 human tendency and will try to eliminate the personal 
 factor, or at least to bring it into balance with factors 
 more vital to the main purpose, which is the erection of 
 
 125
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 the most effective building for housing the plant for de- 
 mand creation. 
 
 Analyzing his main problem, the manager finds him- 
 self, as in his approach to the construction of his fac- 
 tory, confronted first v/ith the necessity of determining 
 the size, type, character of construction, and cost of the 
 building to be erected. These general considerations 
 are for his decision ; the technical building problems he 
 must necessarily leave to an architect or engineer. 
 If the selling plant is to be located at the works, in style 
 and materials the building will match, or at least har- 
 monize with the general group. It may even occupy 
 part of one of the factory buildings, finished and fitted 
 up in a manner commensurate with the impression it is 
 desired to make on customers and prospective customers 
 visiting the works. 
 
 Broadly speaking, all these matters of size, type, char- 
 acter, and cost are governed by considerations similar 
 to those outlined in the study of factory construction; 
 just as the important administrative problem involved, 
 the financing of the selling plant, is also parallel. In the 
 construction of the factory, the accommodation of the 
 equipment was the first purpose, and efficiency of 
 operation the main end. In the housing of the plant for 
 demand creation, the same guiding motives appear. 
 The factor of advertising value, however, takes on 
 much greater importance. 
 
 The manager, in shaping his construction policies, 
 must consider the effect on customers, especially if his 
 industry happens to be in a line like furniture, silver- 
 ware, or motor cars, and customer visits to the factory 
 are encouraged. Even where advertising or the work 
 of traveling salesmen is the chief factor in demand 
 
 126
 
 LOCATION,CONSTRUCTION, AND EQUIPMENT 
 
 creation, the prestige obtained from imposing or un- 
 usual buildings is often important. Indeed, entire 
 groups of buildings occupied by houses like Butler 
 Brothers, Sears, Roebuck and Companj', the Strath- 
 more Paper Company, the National Cash Register 
 Company, and the leading automobile manufacturers 
 are brought into the scheme of demand creation and are 
 exploited in direct and general advertising to emphasize 
 the importance and the volume of business transacted 
 and to enhance the reputation of the product or the 
 merchandise distributed. 
 
 In assembling and striking a balance among all the 
 factors that enter into the solution of these sub-prob- 
 lems, an unbiased and individual point of view makes 
 for safety. What other men or other industries have 
 done is not always a safe guide. It may be necessary 
 to break absolutely new ground to arrive at the most 
 effective solution. Undue weight given to traditional 
 standards or trade usages is likely to prove as unfortu- 
 nate at this stage as would failure to recognize and make 
 allowances for the personal factor in the beginning. 
 
 Sometimes, indeed, the history of a single organiza- 
 tion supplies examples of absolutely opposed policies. 
 Fifteen years ago a large industrial concern provided 
 quarters for its sales department at the factory by 
 rearranging and refinishing one of its earlier mill-con- 
 struction buildings, choosing to add a new steel-and- 
 brick unit for manufacturing purposes rather than build 
 a special demand-creation plant. For ten years the 
 revamped structure served every purpose, including 
 that of impressing thousands of prospective custom- 
 ers who visited the works and tens of thousands who 
 viewed the factory group in advertisements. Then 
 
 127
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 came the idea of completing the factory group and pro- 
 viding a sort of monumental entrance to the works by 
 erecting a new home for the administrative and sales 
 departments at another point. What was virtually a 
 high-grade metropolitan office building was conceived 
 and executed on a scale so lavish as to provide space 
 for generous expansion for many years. 
 
 Officials of the company insist that the new structure 
 has been a wise investment, from the standpoint of 
 publicity as well as organization efficiency and inspira- 
 tion. This is probably true. It is entirely possible to 
 reconcile the earlier policy of investing the company's 
 resources in factory buildings with the later policy of 
 lavish outlay on an office structure. The decisive factor 
 in both cases may have been administrative, the neces- 
 sity of utilizing the first appropriation where it would 
 reduce manufacturing costs, and the subsequent growth 
 of a surplus which could be devoted to a selling plant. 
 As the advertising of this concern identffies the whole 
 factor with its sales and promotion activities, such a 
 building would adequately express the strength and 
 dignity of the business. It is obvious, however, that 
 the business man seeking facts on which to base his 
 own construction policies must exercise care and judg- 
 ment in his choice of examples to follow. 
 
 Equipment is the third main problem in this manage- 
 rial sequence for supplying the concrete needs of the 
 plan for demand creation. What the manager must 
 aim at is to furnish the sales and allied departments with 
 tools of a quality and effectiveness proportionate to the 
 results sought. His own experience with this or that 
 kind of lighting or ventilation, with this or that system 
 of filing or follow-up, with this make of typewriter or 
 
 128
 
 LOCATION,CONSTRUCTION, AND EQUIPMENT 
 
 that model of desk, may be of great value in determin- 
 ing his equipment standards. But personal preferences 
 alone should not inform his choice. Quality and quan- 
 tity of output are the two chief problems into which the 
 main question resolves itself. The parallel with the 
 results of our analysis of factory equipment is very 
 close. Even competent men are powerless to hold out- 
 put to a given standard unless their tools are of corre- 
 sponding capacities. It is the manager's task not 
 merely to put aside groundless prepossessions and preju- 
 dices, but to analyze and master the elements of each 
 sub-problem. Knowing these, he can assemble all the 
 factors bearing on the main issue and match and rec- 
 oncile them without regard for any solution which has 
 ever been offered before. 
 
 This fresh view^point, indeed, is vital in considering 
 equipment for the reason that in this activity of busi- 
 ness the principle of balance is rarely observed. Too 
 often the governing factor is cost rather than quality 
 or service. The average manager entertains no doubt 
 of the value of factory units whose production he can 
 measure in terms of lowered costs. He approves the 
 requisition even when it involves the exchange or scrap- 
 ping of costly machines. In buying equipment for 
 demand creation, however, he is frequently ultra- 
 conservative. 
 
 Salesmen for oflSce appliances say that while the pur- 
 chase of plant equipment is left more or less to subor- 
 dinate executives, the buying of ojQBce equipment is 
 usually under the direct control of the management. 
 The reason for such inconsistent policies is not hard to 
 find. An expenditure for machinery devoted to demand 
 creation is looked upon as " unproductive." The possi- 
 
 129
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 bility of saving time and money or of increasing output 
 as a means of increasing sales is not realized in its true 
 proportion. The only way to correct this shortsight- 
 edness is to put the fresh point of view into play. The 
 manager must coordinate his problem of finance with 
 that of equipping for effective demand creation. 
 
 So much for what may be called the plant policies of 
 demand creation. Lengthy as this chapter has become, 
 the principal factors have been no more than indicated. 
 The method of analysis worked out in our earlier ap- 
 proach to the plant policies of production has been 
 applied to one set of distribution problems — in this in- 
 stance concrete problems which allowed of a relatively 
 simple procedure. But in the following chapters we 
 shall consider the more intricate operating policies of 
 demand creation. Here again the method of approach 
 and analysis worked out in the consideration of produc- 
 tion activities will be applied to the policies governing 
 materials, agencies, and organization, the operating 
 problems of demand creation. 
 
 130
 
 CHAPTER X 
 
 MATERIALS 
 
 WHEN a factory starts to turn out goods, its static, 
 preparatory phase ceases and its dynamic stage 
 begins. Up to that moment the management has 
 been preoccupied with policies bearing on the plant 
 itself, where to place it, how and of what to build it, 
 how to equip it, though all these policies are shaped 
 and colored by the requirements of the operating activ- 
 ities which are to follow. Once the plant policies are 
 expressed in brick and mortar and machinery, their 
 broad outlines for the time at least remain fixed. The 
 guiding policies of operation, on the other hand, are 
 subject to constant change to accommodate varying 
 influences affecting materials, labor, and in lesser de- 
 gree organization. 
 
 In the activities of demand creation a close parallel to 
 these production conditions exists. We have seen that 
 the same method of approach, the same scheme of anal- 
 ysis can be applied to both; that the same language can 
 be carried over from one field to the other. In each, 
 plant policies have to do with location, construction, 
 and equipment, and operating policies with materials, 
 agencies for applying motion to materials, and organ- 
 ization. In both, plant policies practically crystallize 
 after they are established; while operating activities 
 present fresh problems for the manager every day. In 
 demand creation, these claims on his attention are mul- 
 
 131
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 tiplied and the difficulty of analysis and decision inten- 
 sified because of the general lack of standards and be- 
 cause the factors engaged are largely abstract quantities 
 not easily measurable, unstable human relations, or the 
 unexplored complexities of market psychology. 
 
 The materials of demand creation are ideas which, 
 when put into circulation by certain available agencies, 
 will cause consumers to want the goods. Like finished 
 products piled up in a factory warehouse, these ideas 
 are of no value until they are transferred to prospective 
 buyers. This, indeed, is the whole process of demand 
 creation — the transmission of ideas about a commodity 
 through the various agencies of distribution, middle- 
 men, salesmen, and general and direct advertising. 
 
 Any one or all of these agencies may be used. Prob- 
 lems of organization, are involved therefore, including 
 analysis of the market, determination of price policies, 
 the most effective combination of agencies, and coordi- 
 nation of the eflFort and money expended through each. 
 In the interests of simplicity, however, ideas about the 
 goods will be regarded, for the time being, simply as 
 materials for possible and not necessarily for specific 
 and actual use. For the same reason, in the chapters 
 immediately following, the agencies of demand creation 
 will be viewed merely as agencies, leaving the problems 
 of organization and coordination, of establishing the 
 right relations between the agencies and materials em- 
 ployed, for later discussion. 
 
 To consider ideas about the goods as the materials of 
 demand creation is, I think, a new conception. If it 
 holds in practical business, it should simplify all the 
 processes of our analysis, since ideas about the goods 
 are the selling points from which salesmen, middlemen, 
 
 132
 
 MATERIALS 
 
 and advertising men develop arguments and demon- 
 strations to arouse in the consumer's mind interest in a 
 product and a desire to possess and enjoy it. 
 
 Despite the fact that demand creation is largely a 
 problem in applied psychology, the analogy between 
 factory materials and selling points is close. Ideas about 
 the goods can be handled as definite, almost tangible 
 things. The results they produce can be measured with 
 a fair degree of accuracy, particularly in advertising, 
 where the personal influence of the salesman does not 
 enter tc) confuse the issue. This means that the effec- 
 tiveness of each idea can be tested, that different ar- 
 rangements of the same ideas can be compared and 
 measured by the results attained, and that the relative 
 eflSciency of the agencies employed for their transmis- 
 sion can be accurately determined. 
 
 The parallel may be continued still further; the writ- 
 ten or printed symbols which represent a given idea or 
 series of ideas can be classified, indexed, and filed as 
 readily and as definitely as the different materials used 
 in manufacturing are classified, inventoried, and put 
 away in bins until they are needed. 
 
 Here enters the value and significance of the concept 
 that ideas about the goods may be treated as the ma- 
 terials of demand creation. Considered as materials, 
 they can be subjected to an analysis similar to that 
 which was applied to the problem of materials in pro- 
 duction. There we found that the governing factors had 
 to do with the kind, the quality, the sources, and the con- 
 trol of materials, with their handling and utilization in 
 the factory, and finally with the determination of labo- 
 ratory standards to guide the activities involved and to 
 insure economy and efficiency. Utilizing the same ap- 
 
 133
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 proach, the manager finds that his problem of demand 
 creation is simplified, that it becomes to a great degree 
 one of definite measm'ement rather than of opinion 
 and personal judgment. 
 
 Considered in this way, the problem, so far as it is 
 concerned with the materials themselves rather than 
 with the means of transmitting them, breaks up, as 
 in production, into two groups of sub-problems hav- 
 ing to do (1) with the kind, quality, sources, control, 
 and utilization of the ideas in the sales department 
 (including their development, accumulation, arrange- 
 ment, and handling) and (2) with the adoption and 
 development of laboratory standards in the treatment 
 of these ideas. This last is by no means the least im- 
 portant step. It will be seen that something analogous 
 to the laboratory methods of analysis, employed to 
 determine what are the most effective and economical 
 factory materials, may be applied with corresponding 
 advantage to the less obvious field of market dis- 
 tribution. 
 
 First, then, is there a choice as to the kind of ma- 
 terials to be employed in demand creation, the kind 
 of ideas .5^ Clearly, j^es. The selling points of soap and 
 sewing machines, of rolled oats and window shades, are 
 certainly not interchangeable. Besides its general class 
 character, each product has special qualities or utilities 
 which should form the basis of the appeal to the con- 
 sumer and, through the hope of quick and profitable 
 sales, to the middleman as well. 
 
 It is quite as incumbent on the manager to discover 
 what are these special qualities or utilities as it was to 
 canvass the possible sources of factory materials and 
 determine whether wood or clay or alloy steel should 
 
 134
 
 MATERIALS 
 
 enter into the composition of the product. But not so 
 easy. In choosing his manufacturing materials, he 
 would have the advice and experience of his factory- 
 superintendent, perhaps of an industrial engineer or 
 chemist as well, to supplement his own inquiries and 
 conclusions. In many lines where the raw materials of 
 manufacture will be the finished products of more basic 
 industries, he can command the counsel of experts em- 
 ployed by these sources to analyze his needs and facil- 
 itate their supply. Above all, he has the recorded 
 knowledge and practice of his own and allied trades in 
 the solving of like problems and in the handling and 
 processing of the same or similar raw stock. 
 
 It is here, indeed, that the manager sees his two prob- 
 lems as to material diverging most emphatically. His 
 sales and advertising managers will give him help com- 
 parable to that of his superintendent and factory spe- 
 cialists; his advertising agency and the service bureaus 
 of various mediums of publicity will supply further 
 counsel. But in determining and measuring the value 
 of his materials of demand creation, neither he nor his 
 aids can find the experience of other manufacturers 
 recorded and available in anything like the same de- 
 gree as in production. Standards are so few that they 
 may almost be regarded as non-existent. 
 
 Yet the kind and the quality of the materials used 
 in demand creation are matters of utmost importance. 
 Salesmen frequently have it forced upon them that 
 some casual argument, some apparently minor selling 
 point, seems to engage the attention of prospects much 
 more readily than do those on which he and the sales 
 department behind him have placed their reliance.. In 
 other words, their conception of the consumer's wants 
 
 135
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and the motives likely to lead to purchase does not 
 coincide with the facts. Somewhere the analysis has 
 gone astray, and the most available materials for de- 
 mand creation have not been appraised at their true 
 value or assigned to their rightful tasks. 
 
 It is clear, therefore, that a test of quality is needed. 
 But is such a test practicable in dealing with quantities 
 so abstract and so variable as human motives and men- 
 tal reactions .f^ My own experience and knowledge of 
 what other business men have accomplished convince 
 me that it is practicable. The best ideas are those 
 which arouse the maximum of demand at a given ex- 
 pense or a required demand at a minimum of outlay, all 
 things considered. Definite methods can be applied to 
 the measurement of the response made by prospective 
 purchasers to any particular selling point or combina- 
 tion of selling points, especially when the vehicle of 
 communication is direct advertising and the conditions 
 under which the tests are made can be regulated or their 
 influence on the result can be traced. 
 
 At this point the considerations having to do with 
 quality merge with those governing the establishment 
 of standards in materials. The latter is the decisive 
 problem, indeed, in the manager's quest of the most 
 effective selling points. The standards thus developed 
 jQot only are criterions by which to guide judgments, 
 but represent actual, proved materials for demand crea- 
 tion — individual or associated ideas about the goods, 
 the reactions of which have been tested and demon- 
 strated, ideas which can be used over and over again 
 in diverse ways by the different agencies employed. 
 
 Before approaching this problem of standards, the 
 gathering, control, and utilization of the materials of 
 
 136
 
 MATERIALS 
 
 demand creation should be attended to. First of all, 
 sources of ideas about the goods and some dependable 
 method of bringing them together must be developed. 
 Reports from salesmen, letters from customers, the re- 
 sults of prize contests, transcripts of sales convention 
 speeches and demonstrations, and direct, intensive anal- 
 ysis of the product by the sales and advertising depart- 
 ments are the commonest sources of materials. It is 
 feasible, indeed, to work out as definite a routine for 
 assembling and classifying selling points and making 
 them accessible for use as the routine which the purchas- 
 ing agent employs in keeping his factory supplied with 
 the steel or lumber or lubricating oil which it consumes. 
 
 The parallel with production continues in the utiliza- 
 tion of ideas. When a manufacturer determines that 
 one particular material is the best the market affords 
 for his product (cost, quality, and service to the con- 
 sumer all considered) he adopts this as his standard 
 and keeps on using it just as long as results are satisfac- 
 tory and a better material is not discovered. Yet a 
 striking cause of waste in advertising and selling lies 
 in the fact that many distributors are using demand- 
 creation materials of secondary value because they con- 
 ceive that the appeal of their primary ideas has worn out. 
 
 They cast about, therefore, for new ideas about their 
 goods, forgetting that consumers buy again and again 
 to satisfy the same basic needs, and that long after you 
 are tired of your reiterated selling points, the majority 
 of the public is probably still unacquainted with them. 
 Nearly all the great and successful modern campaigns 
 have exploited persistently the dominant ideas about 
 their products. Just as the hundreds of novels published 
 each year have the common theme of love but achieve 
 
 137
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 interest and distinction by reason of new settings, new 
 situations, and new characters, so the most effective 
 advertisement or selling talk is that which is based on 
 the primary appeal of the goods to the consumer, no 
 matter how often that appeal has been used before. 
 So long as Ivory soap floats and is 99 j^ per cent pure, 
 these qualities will count most heavily in keeping up 
 its volume of sales. 
 
 When analysis of your product has uncovered all the 
 elements of service, advantage, or pleasure which it 
 holds for the consumer, there is relatively little profit 
 in having these selling points in stock unless each is 
 used in the right place, at the right time, and in the 
 right way to produce maximum demand with the 
 least possible expense and waste. This achievement 
 frequently involves the planning of a series of inten- 
 sive campaigns, each designed not only to employ 
 effectively some one of the agencies of demand creation, 
 but also to make the most of the available ideas about 
 the goods. To stimulate curiosity' and make the first sale 
 of a five-cent cigar or a ten-cent cereal is obviously an 
 easier undertaking than to perform the same function 
 for a high-priced motor boat or player piano. Belief in 
 the worth of such a product, even a strong desire for 
 that product, is not enough to make a man purchase it. 
 
 The average consumer will risk at any time a small 
 amount to try a new substitute for a personal or house- 
 hold necessity he is buying regularly. The lure of nov- 
 elty or a single outstanding selling point may suffice to 
 fix his attention and complete the sale. When he faces 
 a considerable outlay for some article lacking vital ap- 
 peal, however, or when to buy it means the expenditure 
 of money he had no intention of spending and feels no 
 
 138
 
 MATERIALS 
 
 necessity of spending, it may require a protected cam- 
 paign through magazine and newspaper advertisements, 
 through letters, catalogues, house organs, and salesman's 
 calls to educate him to the point of purchasing. Man- 
 ufacturers of cash registers, adding machines, type- 
 writers, and other business utilities sometimes pursue 
 a " live " but unresponsive prospect for several years 
 before the order is finally booked. 
 
 It follows that in any extended campaign of demand 
 creation the directing minds will have many alternatives 
 in selecting, combining, and arranging the selling points 
 suitable for use by each agency and in fixing the order 
 in which the various combinations shall be brought to 
 bear upon the prospect. Judgment and experience will 
 go a long way in guiding these decisions, in marking the 
 superfluous idea whose elimination will strengthen the 
 advertisement or sales talk, and in providing a logical 
 program for the presentation of the various arguments 
 by the coordinate agencies. Yet a high degree of ef- 
 ficiency in utilization' will not be attained unless the 
 methods as well as the materials of demand creation 
 (here the finished letter, advertisement, or sales talk) 
 are tested and reduced to standards — a not impossible 
 enterprize, as we shall see a few pages further on. 
 
 Making the materials of demand creation available 
 for use may be considered a process of utilization and 
 control closely akin to the handling and storing of 
 materials in the factory. The ideas, reduced to written 
 or printed symbols, must be classified and kept ready 
 for easy reference. In many cases the memory of the 
 salesman or advertising manager serves both as inven- 
 tory and storage place; in others, scrap books or filing 
 devices of various sorts preserve the materials and make 
 
 139
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 them accessible when needed. The most advanced sales 
 organizations maintain filing and index systems which 
 group similar ideas about the goods in the order of their 
 pulling power and provide means of automatically bring- 
 ing these to notice at intervals. 
 
 One sales manager of my acquaintance files every 
 scrap of information bearing on the functions and known 
 uses of his machines according to subject, supplements 
 this material with such clippings from magazines, news- 
 papers, and letters as contain suggestions or idea-germs, 
 and keeps a master index to all under the glass top of 
 his desk. To this store of ideas he adds from his corre- 
 spondence, his reading, his contact with men inside and 
 outside the organization; and he is able to bring the 
 whole to bear when planning an elaborate campaign or 
 dictating a single sales letter. A signally successful in- 
 dustrial concern improves on this practice by display- 
 ing all its live sales material on multi-leaved fixtures 
 which permit the bringing together of all the associ- 
 ated ideas, thus making the members of each group 
 visible at the same time and facilitating choice of the 
 best for the current purpose. Further, this company 
 has gathered into an indexed pocket volume several 
 hundred selling points for its various products, the most 
 effective answers to prospects' objections, and the ap- 
 proved methods of demonstrating its machines. This 
 is used for the refreshment of the memories of seasoned 
 salesmen as well as for the information of new recruits. 
 
 In both these instances the attitude toward the ma- 
 terials of demand creation is practically the same as that 
 of the average factory organization toward its raw and 
 partly processed stock. The guiding purposes are vir- 
 tually identical: to insure an adequate supply of ma- 
 
 140
 
 MATERIALS 
 
 terials, to classify these according to their character, 
 properties, and uses, and to keep them instantly acces- 
 sible through an inventory or master-index. 
 
 Quantity is a further factor in utilization. In manu- 
 facturing the chief element in determining quantity is 
 the cost of the materials consumed. The solution is not 
 reached until each unit of product contains the mini- 
 mum amount of raw stock consistent with the kind of 
 product desired. In demand creation, on the contrary, 
 quantity is a question of economizing the prospective 
 buyer's time and attention. Too many selling points 
 will tire him. Too few will certainly fail to persuade 
 him that he must purchase or else sacrifice a possible 
 advantage. 
 
 Every executive has had contact with the salesman 
 who " talks himself out of an order " as well as with the 
 young or mediocre man who does not know how to pre- 
 sent a convincing case for his merchandise. Both are 
 recognized types in business; both are victims of de- 
 fective analysis. In the first case, the process stops 
 with the gathering of a stock of ideas about the goods 
 and fails to discover the proper proportion in utilizing 
 them. In the second, there is probably no conscious 
 analysis either of the prospect's needs or of the qualities 
 in the merchandise which might engage his attention. 
 
 Analogous types of advertising are also common. 
 There is the letter, magazine page, or booklet which 
 crowds so many arguments upon the potential buyer 
 that he is repelled by the mass or is fatigued after he 
 begins to read; and at the other extreme is the adver- 
 tisement which puts forward so few or such general ideas 
 about the goods that no positive impression is made 
 upon the prospect's mind. In a word, the problem of 
 
 141
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 quantity is largely one of elimination and selection. 
 The necessity of establishing the right balance between 
 the materials used and the result desired is no less im- 
 perative than in manufacturing operations. 
 
 This brings us to the final problem confronting the 
 business man in his consideration of the materials of 
 demand creation, that of establishing standards com- 
 parable with the laboratory standards so common in 
 the industries. The possibility of formulating such 
 standards may be challenged by the average producer, 
 no matter how carefully he has standardized his fac- 
 tory materials. For here as elsewhere in the field of 
 distribution too many business men are guided by rule 
 of thumb. They guess at the most effective sales ideas 
 which analysis of their product discloses, guess at the 
 most forceful forms of expressing these, and guess at 
 the most efficient agencies and mediums for transmit- 
 ting them to prospective purchasers. They spend hun- 
 dreds of thousands of dollars on selling campaigns based 
 on a series of approximations or opinions arrived at in 
 this hit-or-miss fashion. 
 
 This means an enormous waste of money, effort, and 
 product, due simply to the general lack of standardiza- 
 tion. Nor is waste from this cause confined to the ma- 
 terials of demand creation. It is characteristic of our 
 distributive organization as a whole. In the aggregate, 
 it constitutes an overwhelming case for a thorough and 
 systematic study by government and cooperating pri- 
 vate agencies of all the activities of market distribution. 
 
 The abler business man, to be sure, makes an effort 
 to determine those properties of his goods which should 
 attract the attention of the possible buyer and awaken 
 the desire for possession Vvhich is the root of market 
 
 142
 
 MATERIALS 
 
 demand. He discovers by a general inquiry, if not by 
 intensive study, the points of superior quality, utility, 
 or service inherent in his products as distinguished 
 from competing goods. He turns for guidance as to the 
 form in which these ideas are to be conveyed to his 
 own experience, to the principles of English style, and 
 to the practice so far as he can ascertain it of other 
 men who are marketing commodities appealing to the 
 same classes of prospects, under conditions similar to 
 his own. 
 
 From the observations of successful sales and adver- 
 tising managers, psychologists, and students of busi- 
 ness he gathers the fundamental rule that the mental 
 energy of the prospective buyer must be conserved by 
 cutting down to a minimum the labor attendant on the 
 assimilation of new ideas and concepts. He learns to 
 avoid abstract statements and vague general claims. 
 He seeks short, familiar words which will convey in 
 homely phrase the exact meaning he wishes to transmit. 
 He employs apt illustrations and figurative statements 
 whenever they will help the prospect to visualize the 
 product and its effect on the comfort, the pleasure, the 
 health, or the safety of himself or those he holds dear. 
 And he uses the imperative mood judiciously when the 
 reaction desired is at length suggested. 
 
 These are only a few of the many precepts put for- 
 ward tentatively or more decisively by the psychologist 
 and by the specialist in sales. Their weakness lies in 
 their a priori basis. What the business man requires 
 is a practical test for certain ideas about his goods under 
 conditions approximating those of the physical or chem- 
 ical laboratory. He wants a workable method of trying 
 out selling points alone and in combination, in diverse 
 
 143
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 forms, and with varying degrees of emphasis, before 
 staking thousands of dollars on the effect they will 
 produce in an actual sales campaign. 
 
 In making these tests, the manufacturer can borrow 
 the methods by which statisticians in many fields of 
 social and business endeavor arrive at their results. 
 These latter are familiar with what are termed mass 
 phenomena and put their dependence in the law of aver- 
 ages. They know that they can approximate the aver- 
 age height of a country's population, for example, by 
 measuring a few thousands of representative folk drawn 
 from the larger body. If the smaller group is really typi- 
 cal of the mass, the statistician knows that its average 
 height will coincide roughly with the average height of 
 the total population. 
 
 The business man can apply this method to tests of 
 his demand-creation materials. In direct advertising, 
 the mailing of sales letters, circulars, catalogues, sam- 
 ples, and the like to prospective buyers, orders for the 
 goods are proof of an awakened demand which is ca- 
 pable of direct statistical measurement. Each piece of 
 advertising posted is a stimulus; the number of re- 
 sponses per thousand of communications can be deter- 
 mined as exactly as though the operations were con- 
 ducted in a physical or psychological laboratory. It is 
 a simple matter of so framing and keying the advertise- 
 ment that every response can be identified and only 
 those due to it shall be counted. 
 
 It is possible also to isolate a group from the general 
 mass (just as the sociologist isolates a group for study 
 of its stature) to test the average reaction induced by 
 any unit or series of advertisements, thus forecasting 
 the results that will be attained when the campaign is 
 
 144
 
 MATERIALS 
 
 extended to the mass. If the test is unsatisfactory, it 
 demonstrates either that the selling points put forward 
 lack persuasive or convincing power, or that the form 
 in which they are cast is defective. A third alternative 
 hinges on the chance that the group of individuals se- 
 lected for the test is not representative of the mass. 
 The trial list, however, can be revised again and again 
 until it is truly typical, until the average of responses 
 per thousand of letters or catalogues mailed does not 
 vary greatly between test and full campaign. 
 
 Once the business man secures a representative test 
 list or series of lists, he can try out at relatively small 
 expense all his materials of demand creation and deter- 
 mine which selling points and what forms of each are 
 most effective in stimulating the market for his prod- 
 ucts. This method of " sampling " has been successfully 
 employed so often that its value and accuracy are 
 beyond question. 
 
 Take the case of a producer of a food specialty plan- 
 ning a campaign to reach, not the consumers, but the 
 grocers of the country. The whole body of dealers in 
 this line falls little short of a quarter of a million. After 
 working out a series of selling points expressed in such 
 forms as he thinks likely to be effective, let the manu- 
 facturer test this material by mailing it to say one 
 thousand grocers. The group selected must be large 
 enough to give typical results and it must be so selected 
 as to be of the same general character as the whole 
 body of grocers. 
 
 Assuming these conditions, the number of orders or 
 inquiries received from the thousand grocers can be 
 tabulated, and the probable response per thousand can 
 be estimated when the same ideas in the same form are 
 
 145
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 conveyed to the two hundred and iEifty thousand. This 
 can be followed or paralleled by another direct mailing, 
 to another test list similarly constituted, of an alter- 
 nate campaign built on different selling arguments or 
 on the same arguments differently expressed. Half a 
 dozen test campaigns, indeed, might be profitably con- 
 ducted at the same time. The greater the number of 
 tests (always assuming that the trial lists are representa- 
 tive and of about the same character) the more certainly 
 the results will define the weightiest selling points and 
 the happiest ways of conveying them to the prospect. 
 
 Intensive study of the reactions produced by individ- 
 ual units in each test may even result in a composite 
 final campaign which will far exceed the returns from 
 the best of the trial campaigns. It is thus possible to 
 determine what ideas about the goods and what ar- 
 rangement and phrasing of those ideas are most effective 
 in arousing demand — in other words, to select your ma- 
 terials of demand creation with little more dependence 
 on guesswork than obtains in the choice of raw mate- 
 rials in a standardized factory. 
 
 That such tests are practical is indicated by the typ- 
 ical records presented in the table on the opposite page. 
 Here are shown the results of several " try-outs " and 
 complete mailings of various letters and circulars mak- 
 ing the same specific offer of certain commodities. The 
 tests and mailings covered five well-defined functional 
 groups of business men and two additional groups of 
 members of large local merchants' associations. In all 
 but two instances, B^ and D, it will be noted, the results 
 of the tests and full mailings were practically parallel. 
 In one of these, B^ mailed to master printers in June, 
 1915, the orders fell far short of what the test forecast; 
 
 146
 
 IVIATERIALS 
 
 yet the same letter and inclosure to mailed credit men 
 in May, 1915 (B^), brought orders corresponding very 
 closely with the results of the test mailing and also 
 
 TESTS AND MAILINGS TO RATED BUSINESS MEN 
 
 Minimum Standard, 16 Orders per 1,000 
 
 Test lists, each 500 names 
 
 Note: In the column headed "Material Mailed," it will be seen that the same letters 
 stand for identical material, the exponent numbers designating the different mailings. The 
 unit of sale was $5.00. 
 
 
 
 Tests 
 
 Mailings 
 
 Ma- 
 terial 
 Mailed 
 
 Classes 
 Addressed 
 
 Date 
 
 Orders 
 
 Re- 
 ceived 
 
 Orders 
 per 1000 
 
 Number 
 MaUed 
 
 Orders 
 
 Re- 
 ceived 
 
 Orders 
 per 1000 
 
 Ai 
 
 Printers 
 
 4/19/15 
 
 9 
 
 18 
 
 5,368 
 
 117 
 
 21.8 
 
 A2 
 
 Credit Men 
 
 9/14/14 
 
 13 
 
 26 
 
 
 
 
 A3 
 
 ti u 
 
 10/19/14 
 
 13 
 
 26 
 
 16,376 
 
 342 
 
 20.9 
 
 A* 
 
 Advertisers 
 
 10/22/14 
 
 9 
 
 18 
 
 13,890 
 
 295 
 
 21.2 
 
 A* 
 
 Architects 
 
 4/29/15 
 
 7 
 
 14 
 
 1,478 
 
 21 
 
 14.3 
 
 A" 
 
 Association 
 
 
 
 
 
 
 
 
 No. 1 
 
 4/29/15 
 
 16 
 
 32 
 
 2,770 
 
 75 
 
 27.1 
 
 A^ 
 
 Association 
 
 
 
 
 
 
 
 
 No. 2 
 
 
 Not 
 
 Tested 
 
 2,972 
 
 100 
 
 33.9 
 
 Bi 
 
 Printers 
 
 6/ 1/15 
 
 14 
 
 28 
 
 5,296 
 
 93 
 
 16.2 
 
 B2 
 
 Credit Men 
 
 4/15/15 
 
 9 
 
 18 
 
 15,412 
 
 271 
 
 17.6 
 
 B3 
 
 Advertisers 
 
 
 Not 
 
 Tested 
 
 13,848 
 
 197 
 
 14.3 
 
 B* 
 
 Manufacturers 
 
 6/1/15 
 
 7 
 
 14 
 
 8,492 
 
 148 
 
 17.3 
 
 C 
 
 Credit Men 
 
 8/20/14 
 
 2 
 
 8 
 
 
 
 
 D 
 
 (1 u 
 
 9/14/14 
 
 15 
 
 30 
 
 16,439 
 
 348 
 
 21.1 
 
 El 
 
 u a 
 
 1/25/15 
 
 4 
 
 8 
 
 
 
 
 E2 
 
 Advertisers 
 
 1/25/15 
 
 4 
 
 8 
 
 
 
 
 pi 
 
 Credit Men 
 
 3/25/15 
 
 6 
 
 12 
 
 
 
 
 F 
 
 Advertisers 
 
 3/25/15 
 
 2 
 
 4 
 
 
 
 
 Qi 
 
 Credit Men 
 
 7/22/15 
 
 4 
 
 8 
 
 
 
 
 G2 
 
 Advertisers 
 
 9/20/15 
 
 7 
 
 14 
 
 
 
 
 with the final results of B^ Mailed also to a selected 
 list of manufacturers in June (B^), the materials again 
 brought the standard result. In this last case the test 
 
 147
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 fell under the minimum, but because of the uniform 
 performance of B^ and B^, a chance was taken and the 
 letter sent out. In B^ mailed the same day to adver- 
 tisers, without a special list test, the returns fell under 
 the minimum standard allowed. 
 
 Study of this and hundreds of similar tests would 
 seem to indicate that lists classified along functional 
 lines require individual tests even for materials gener- 
 ally effective. In A^ for instance, the test result of 
 fourteen orders per thousand was disregarded because 
 the list was small and the letter and circular used had 
 been uniformly successful on several other lists. Yet 
 here the final result was almost identical with that of 
 the trial and netted only fifty-seven per cent of the 
 average from other lists. 
 
 As final evidence of the value of classifying lists ac- 
 cording to functions or dominant interests, observe 
 that A^ and A^ the same material sent to the members 
 of two large associations of business men, after a single 
 test, brought very satisfactory results from both lists. 
 Another important fact to note is that when the mini- 
 mum standard of orders is as low as sixteen and the 
 test group numbers only five hundred, there is danger 
 that the average will be disturbed where one individual 
 influences the giving of several orders. The larger the 
 trial group and the more clearly classified the various 
 lists used, the more exact an index will the test give on 
 the results from the complete mailing. Experience 
 with the varying factors involved teaches one to sense 
 the abnormal, whether in composition of mailing list or 
 ratio of responses. 
 
 This method of studying ideas and forms of expres- 
 sion in direct advertising would be important even 
 
 148
 
 MATERIALS 
 
 though its use did not extend to other agencies, since it 
 permits the business man to guide an extensive adver- 
 tising campaign by means of an investigation relatively 
 inexpensive. But the importance of the laboratory 
 method does not end with direct advertising. The root 
 idea is the same whatever the agency for selling em- 
 ployed. Selling is accomplished by communicating ideas 
 about the goods through middlemen, salesmen, gen- 
 eral advertising, or direct advertising. The ideas remain 
 virtually the same in all instances. The business man, 
 therefore, can determine in his direct selling laboratory 
 what ideas and what combinations of ideas constitute 
 the most effective selling material. He can then carry 
 over the results obtained to his selling by other agencies. 
 
 To illustrate: An extensive promotion campaign is 
 under consideration. The producer contemplates spend- 
 ing thousands of dollars upon advertising in certain 
 periodicals. What can the " distribution laboratory " 
 do to determine the ideas to be conveyed and the forms 
 of expression to be used to create the desired demand? 
 The circulation of a single periodical used may run 
 into the hundreds of thousands or even into the mil- 
 lions. It is the part of prudence, therefore, to test the 
 response that will result from the communication to this 
 enormous body of subscribers of ideas about his goods. 
 Accordingly he works out the most effective selling 
 points, the most effective arrangement, the most effec- 
 tive form of expression through direct trial mailings. He 
 can even test the final " copy " itself, just as it will ap- 
 pear in the periodical, by mailing it direct to relatively 
 small groups and noting the reaction which it produces. 
 
 Moreover, he can determine in this way the response 
 from differing economic strata. Ideas adapted to build 
 
 149
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 up a demand for a commodity on one economic or social 
 plane may fail utterly with another. The value of a 
 " try-out " for each lies in the fact that most periodicals 
 circulate within well-defined economic and social limits. 
 The ideas and forms of expression which created a de- 
 mand in direct advertising addressed to these strata is 
 likely to have a similar effect when used in periodicals 
 circulating among the same classes. 
 
 The application of laboratory tests to the selling ar- 
 guments used by salesmen is equally practicable. One 
 need only appreciate the fundamental identity of the 
 selling function, whatever the agency employed, to real- 
 ize that the results obtained in experiments in direct 
 advertising can be carried over into distribution by 
 salesmen. The whole structure of the selling talk can 
 be built up on the ideas, the order of arrangement, and 
 the forms of expression demonstrated as most eflScient 
 in creating demand through the medium of direct 
 advertising. 
 
 Many progressive organizations, indeed, insist on all 
 salesmen mastering a standard approach and selling 
 talk which marshals certain basic arguments in a settled 
 order and phrases them in a definite way. This is im- 
 pressed upon the new salesman as more likely to secure 
 the order than would any argument he could frame im- 
 promptu. It serves chiefly as a foundation for the argu- 
 ment he makes to possible buyers. The assumption 
 of the laboratory point of view by salesmen, the con- 
 scious testing of ideas and methods and the recording 
 of results, may in turn strengthen the entire sales 
 organization. 
 
 The general principles upon which the testing or samp- 
 ling of ideas depends apply also when the possibilities 
 
 150
 
 MATERIALS 
 
 of the whole market are studied by the intensive cultiva- 
 tion of one section. It is only necessary that social and 
 economic conditions be broadly identical. Other things 
 being equal, the result of a test in Kansas should hold 
 true for the states immediately north and south, though 
 it would not apply to Georgia or Alabama. On this 
 basis a localized selling campaign, narrow in extent, 
 will give relatively exact data from which the possibil- 
 ities of a general campaign of like character may be 
 forecast. 
 
 Again, the laboratory method here suggested lends 
 itself to a determination of just what elements of qual- 
 ity and service in a given product are deemed most es- 
 sential by the consumer. The effectiveness of the ideas 
 conveyed in building up a demand reflects the intensity 
 of human wants as to the elements of quality and service 
 described. The producer can sound the consumer and 
 can adapt his product to the latter's expressed demand. 
 In other words, the entire selling campaign can be 
 planned and directed by guarded tests comparable to 
 the laboratory investigations of the scientist. 
 
 It may be possible in time greatly to extend and sim- 
 plify this analysis of materials. As yet the professional 
 psychologist has addressed himself to few of the prob- 
 lems which engage the attention of business men. The 
 conclusions arrived at, too, are rarely of much practical 
 value, chiefly because the psychologist, in framing his 
 tests or phrasing his questions, is not always cognizant 
 of the thing the business man wants to know. In 
 some of these experiments a series of advertisements has 
 been submitted to a group of thirty, forty or fifty stu- 
 dents, themselves, perhaps, far removed in character, 
 tastes, and purchasing power from the average con- 
 
 151
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 sumer, and they have been asked to indicate which is 
 the most attractive or effective of the lot. 
 
 The question should have been more specific: " Which 
 advertisement would get your attention first? Which 
 one, if any, would hold your interest? Which one, if 
 any, would stir curiosity or make you want the product 
 enough to ask for it, by letter or at the nearest store? " 
 And so on, breaking the one question up into perhaps 
 a dozen specific inquiries which would give definite in- 
 formation as to the size, the shape, the quantity and 
 character of ideas, the kind of illustration, and the style 
 of argument which would secure and hold the reader's 
 attention and induce the desired action, whether this 
 be the filling out of a coupon or the inclosure of a signed 
 order and check. 
 
 Some encouraging attempts have been made to elim- 
 inate from such laboratory experiments the factors 
 which reduce the value of the results and to check up 
 these results by repeating the experiment under actual 
 business conditions. One of the most interesting of 
 these studies, conducted at the University of Michigan 
 in 1913, minimized the chances of error by submitting 
 the questions to five hundred and sixty persons, fifty 
 of whom were outside business men. The questions, 
 too, dealt with specific things, like the attention value 
 of various sizes of copy, from full pages to quarter pages, 
 and the efficiency of the coupon in securing answers. 
 The man who conducted the experiments was himself 
 a successful advertiser, and the copy which he submitted 
 to his subjects was tried out a little later in an advertis- 
 ing medium of known pulling power. ^ 
 
 It is significant that the results of both tests cor- 
 
 ^ See W. A. Shryer, " Analytical Advertising." 
 152
 
 MATERIALS 
 
 responded closely. The large number of subjects in 
 the test undoubtedly reduced the percentage of error. 
 Divided into groups of about twenty-five persons, it 
 was found that the opinions of a few of these groups 
 varied widely from the average, thus indicating the 
 danger of depending for a decisive test on a single small 
 body of subjects. At the same time, further illustra- 
 tion was given the principle that specific problems in 
 demand creation can be solved by laboratory tests con- 
 ducted under carefully guarded conditions on a suffi- 
 cient number of representative subjects. 
 
 153
 
 CHAPTER XI 
 
 AGENCIES OF DEMAND CREATION — THE 
 MIDDLEMAN 
 
 ■p^EMAND creation is a problem in the transmis- 
 -■— ' sion of ideas. Here is a business man seeking 
 a market for his product. It may be a standard 
 article somewhat improved, but at the usual price. It 
 may be the usual grade at a lower price, or a franldy 
 cheaper substitute depending on a marked price reduc- 
 tion for its appeal. Or he may offer a specialized com- 
 modity invented or adapted to satisfy a specific want 
 not yet fully realized by the consumer. The seller, as a 
 rule, believes that in one or more of the three essentials 
 — quality, price, or service — the product offers ad- 
 vantages over any other article of its kind. Else he 
 would not undertake the endless labor of stimulating 
 demand for an inferior commodity. It may sometimes 
 happen that an article not otherwise superior reflects 
 only the prestige or charm of the seller, or more ener- 
 getic selling power overcomes the attraction of greater 
 intrinsic merit in competing merchandise. But even 
 here faith in the goods is generally recognized as indis- 
 pensable. " First sell yourself " is an axiom in the 
 practice of salesmanship. 
 
 Now " selling one's self " is the rule-of -thumb equiv- 
 alent for the series of activities described in the last 
 chapter — the gathering, classifying, testing, and stand- 
 ardizing of the materials of demand creation. With 
 these proved ideas about the goods in hand, the mana- 
 
 154
 
 THE MIDDLEIVIAN 
 
 ger's task is to transmit them effectively to the minds of 
 possible consumers. When he thus leads consumers to 
 adopt his viewpoint, as to the utility and desirability 
 of his wares, he induces them to become purchasers. 
 For this work of transmission, he finds that three impor- 
 tant agencies have been developed, the middleman, the 
 exclusive sales force, and direct and general advertising. 
 
 His problem is to determine which one of the three 
 will best perform the function of demand creation with- 
 out undue interference with other necessary activities 
 of production and distribution. He may discover that 
 two or more can be combined advantageously, or that 
 different agencies will be most effective in different sec- 
 tions of his selling field. Following the analogy already 
 indicated, they may be considered as corresponding to 
 the labor factors in production. To these ideas or sell- 
 ing points, also, motion must be applied through human 
 effort in order to perfect and transmit them to the minds 
 of prospective users and thereby create desire for the 
 goods. 
 
 In any extended study of the agencies of demand crea- 
 tion it becomes evident that the sales record of a single 
 industry frequently includes all the significant func- 
 tions and operations which have marked the develop- 
 ment of distribution since exchange of commodities 
 began. The surplus product of a country blacksmith 
 shop in the early fifties — to quote a classic illustra- 
 tion in American business — consisted of several farm 
 wagons which the neighborhood had not been able or 
 willing to absorb. A local demand for a number of such 
 vehicles had been created with little effort. A simple 
 demonstration which convinced the prospective buyer 
 that wheels, axles, and other members were of sound 
 
 155
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 timber equal to the rough usage before them, was usu- 
 ally sufficient to complete the transaction, since the 
 product satisfied a known and imperative want of the 
 customer. When all the farmers in the district who 
 needed and could afford new wagons had been supplied, 
 however, the makers faced the alternative of finding a 
 market further afield or of suspending manufacturing 
 operations and reverting to the odd jobs with which the 
 average forge of the time occupied itself. 
 
 Had they been ordinary blacksmiths, they would have 
 taken the easier way, or probably would have failed to 
 realize that they had any choice in the matter. But they 
 possessed initiative, intelligence, and that pride in their 
 product which has been, I believe, an important motive 
 in the expansion of American business. One of the 
 brothers, therefore, hitched a team to a string of the 
 surplus wagons and drove forth in quest of buyers. He 
 found them on this and on subsequent excursions within 
 the country and without, in villages and towns, on farms 
 along the way, at crossroads stores, and even at smithies 
 whose owners had not the craft or capacity to build 
 wagons in numbers to supply the local demand. 
 
 Wherever conditions seemed to warrant and a com- 
 petent man could be interested, a representative was 
 appointed, at first to take orders on commission for the 
 wagons which the brothers learned to trade-mark and 
 guarantee, but later as a wholesale or retail dealer who 
 bought his stock and assumed all the functions and risks 
 of demand creation and supply in his territory. From 
 such homely beginnings the Studebaker business has 
 grown to international proportions, manufacturing a 
 long line of horse-drawn and motor vehicles and distrib- 
 uting them to every civilized corner of the earth. In 
 
 156
 
 THE MIDDLEMAN 
 
 developing this world market, virtually all the agencies 
 of demand creation have been employed : middlemen of 
 various sorts, wholesalers, retail dealers, and exclusive 
 agents; a direct sales force, including not only sales- 
 men, but branch houses, district sales organizations, and 
 even retail branches in the larger cities, as well as 
 direct and general advertising of every kind and class. 
 
 Whether the ideas about the goods are communicated 
 through spoken or written symbols, by middlemen, sales- 
 men, or advertising, the ultimate purpose is the same. 
 For the time being, however, we shall consider only 
 those problems which arise when the middleman is used 
 as the agency of demand creation. It cannot be over- 
 looked of course that this is by no means the only func- 
 tion which he has exercised in the past and continues 
 in many cases still to exercise. 
 
 The present tendency of many producers to go around 
 the middleman and eliminate him entirely or as fast as 
 possible from their schemes of distribution cannot 
 rightly be understood without review of what these his- 
 torical functions are. Nor can the current conditions 
 which enter as factors in the problems of demand crea- 
 tion be explained without analysis of these functions and 
 their influence on the producer's relations with the 
 middleman. These functions may be listed as follows: 
 
 1. Sharing the risk. 
 
 2. Transporting the goods. 
 
 3. Financing the operations. 
 
 4. Selling or demand creation. 
 
 5. Assembling, assorting, and reshipping the goods. 
 Each of these functions was at first divided among a 
 
 series of middlemen, the selling agent, the wholesaler, 
 and the retailer each assuming his part in turn. Each 
 
 157
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 middleman took the risk that the goods might lose 
 value or go stale on his hands, just as he bore the hazard 
 of destruction, damage, or theft of the goods while he 
 held title. Each took the chance of credit losses. Each 
 paid a proportion of the transportation charges from the 
 producer's stock room to the consumer's hands. Each 
 had a part in financing the entire operation. Each as- 
 sumed a share of the work of selling, of creating for 
 the goods he had purchased a demand in an outer circle 
 of middlemen or the final circle of consumers. And 
 each performed part of the task of assembling, assort- 
 ing, and reshipping the goods to make them physically 
 available for consumption. 
 
 In many instances the middleman exercises these 
 same five groups of functions to-day. If he overesti- 
 mates the probable demand for an article or line, he 
 pays the penalty for his bad judgment when he clears 
 his stock at the end of the season. If he encounters a 
 season of depression with heavy stocks, the loss is his 
 so long as he remains solvent. And unless he transfers 
 his liability by paying an insurance premium, he as- 
 sumes the risk of loss or damage in his warehouse or 
 while the goods are in transit. Frequently he delivers 
 the merchandise to his customers by team or motor 
 truck, especially if he is a wholesaler supplying a large 
 city and suburban trade, or a retailer in a city where 
 consumers are accustomed to such service. All along 
 the line, the middleman (except when he makes a virtue 
 of demanding cash payments by offering reduced prices) 
 lends his credit to his customers and thus finances the 
 selling operation. 
 
 His business is primarily to sell, but at times his man- 
 ner of performing this function presents one of the pro- 
 
 158
 
 THE MIDDLEMAN 
 
 ducer's gravest problems in distribution policy. The 
 attitude of many middlemen toward a new product or a 
 new line seems to be one of indifference or even latent 
 hostility, unless it offers for some standard commodity 
 a substitute which can be handled more easily or at a 
 greater profit. Middlemen of this type virtually say to 
 the manufacturer: " Create a demand for your goods 
 and I will handle them. I am here to give my custom- 
 ers what they ask for, not to make a market for some- 
 thing they know nothing about," 
 
 There is no denying that they have sound and prac- 
 tical reasons for this attitude. Hardly a day passes that 
 the wholesale merchant is not importuned to buy and 
 push some new branded product which can hardly be 
 distinguished from one or perhaps a dozen like it which 
 he already has in stock. If he yields, it means that he 
 puts an additional load, however small, on his financing, 
 on his stock-keeping, on his selling, and indeed on every 
 department of his business. One item, of course, 
 would not be much of a burden; but multiply that item 
 by five hundred or a thousand in a year (the latter is not 
 an impossible estimate in many lines) and the risk of 
 loss and spoilage becomes so great as to justify a defen- 
 sive policy. The same problem on a lesser scale con- 
 fronts the retailer. No wonder, therefore, that middle- 
 men of all classes look with disfavor on a new substitute 
 for established brands unless they can see in it an extra 
 profit or a ready market. 
 
 Because they exercise the final functions of assembling 
 and forwarding merchandise stocks, assorted according 
 to their customer's needs or expressed desires, the influ- 
 ence of middlemen on the choice of commodities is very 
 great. Their negative attitude toward demand creation 
 
 169
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 in many lines, coupled with this power of influencing 
 purchasers, has already affected the relations between 
 manufacturer and middleman and will undoubtedly 
 work further and radical changes in our system of dis- 
 tribution. 
 
 In certain lines the middleman frequently takes on the 
 constructive function of financing the producer directly 
 and indirectly. Many of the cotton mills of the South 
 were built with capital furnished by northern selling 
 agents and their associates. Often, too, the wholesaler 
 indorses the commercial paper of the manufacturer and 
 thus indirectly finances production. In the textile in- 
 dustry of New England, such " two-name " paper is 
 quite common, and it is frequently met with in other 
 trades where the design and manufacture of the goods 
 and their movement to the ultimate consumer occupy a 
 considerable cycle of time. Certain it is that in all cases 
 the middleman materially shortens the period of in- 
 vestment on the part of the producer and enables him 
 to make more rapid turnovers. 
 
 At a relatively early date in modern distribution be- 
 gan the taking over entire of certain single functions, 
 which each middleman had been performing in part. 
 The most notable functional middlemen to emerge thus 
 far are the insurance companies, the direct transporta- 
 tion companies, and the commercial banks. The insur- 
 ance company in a real sense performs a middleman's 
 function when it assumes the hazard of damage or loss 
 by fire, by flood, by theft, by wTcck, by non-delivery 
 on time, and by all the other mishaps to which goods are 
 exposed on their journey from factory to consumer. Fur- 
 ther, the insurance company is ready to assume practi- 
 cally the entire element of risk, to guarantee credits, to 
 
 160
 
 THE MIDDLEMAN 
 
 warrant the honesty of salesmen, agents, delivery men, 
 and other employees, to assume the liability for injury to 
 persons or property incident to operation, even to reim- 
 burse a commercial institution like a department store 
 or specialty shop for failure to earn profits because of 
 weather unfavorable to seasonal trade. 
 
 In much the same manner the great transportation 
 companies, railroads, ocean steamship lines, express 
 companies, and more recently the domestic and inter- 
 national parcel posts, undertake and perform an im- 
 portant oflSce originally shared with the producer by 
 successive fractional middlemen. 
 
 In financing the operations of distribution, too, the 
 bank has gone a long way toward relieving the middle- 
 man of his burden. By making advances on merchan- 
 dise stocks, on open accounts, and on commercial paper, 
 it has absorbed much of the function of finance in distri- 
 bution. There are exceptions, as noted above in the 
 textile and other long-cycle industries, but in the main 
 the bank, with associated agencies like commercial-paper 
 brokers, has been long recognized as an indispensable 
 functional middleman in the marketing process. The 
 establishment of the federal reserve bank system, with 
 its power to rediscount acceptable " two-name " paper 
 and bills of exchange, has led to further extension of 
 banking activities in this field. The application of the 
 corporate form to industrial organization has also con- 
 tributed to the producer's financial power by making it 
 possible for him to secure adequate working capital with- 
 out sharing or surrendering control of operations at any 
 point. 
 
 Development of these functional middlemen has re- 
 lieved the wholesaler and retailer of certain onerous 
 
 161
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 activities; in theory, at least, he is left free to concen- 
 trate on his remaining functions of demand creation and 
 physical supply of the merchandise which his customers 
 require. In a measure his field of demand creation has 
 been encroached upon, either as a consequence of his 
 indifference to particular commodities based on wide 
 possible choice among rival products or because of the 
 manufacturer's compelling interest in maximum sales. 
 Beyond question, producer advertising can be regarded 
 as a general or direct appeal, according to its character, 
 which tends to supplant the dealer's efforts in the stim- 
 ulating of a consumer market. 
 
 The advertising medium or agency is in this sense as 
 truly a functional middleman as either the insurance 
 company, the railroad, or the bank. But it is more re- 
 cent in origin and more undeveloped in operation. Ad- 
 vertising has as yet few dependable standards to direct 
 the use, for instance, of coincident newspaper, mag- 
 azine, letter and circular campaigns — especially when 
 they are conducted in cooperation with middlemen. 
 
 This inability to coordinate activities which can only 
 partly be controlled (though the entire expense must be 
 charged up against the product as delivered to the con- 
 sumer) and thus reduce duplication of effort, counts 
 heavily against the efficiency of advertising. Coupled 
 with the indifferent attitude of the middleman, it may 
 account for what seems to many keen students an ap- 
 parent tendency to eliminate the middleman in demand 
 creation. It is one phase, perhaps, of a general move- 
 ment looking to the integration of industries, to the con- 
 trol and coordination of all the activities involved in 
 production and distribution, from the supply of raw 
 materials for the factory, to the laying down of the mer- 
 
 162
 
 THE MIDDLEMAN 
 
 chandise at the consumer's receiving platform or the 
 back door of his residence. 
 
 This, indeed, is the producer's vital problem: Shall 
 I sell direct or through the trade? And if my goods 
 move to the consumer through the recognized channels, 
 shall I depend on the successive middlemen to do all 
 or any part of the work of demand creation? Or can I 
 take on the entire burden of advertising and selling 
 without prohibitive increase of expense and leave to the 
 wholesaler and retailer only the function of physical 
 supply, of stocking my products and handling them out 
 to the customer when they are called for? 
 
 Modern practice embraces the entire range of possible 
 procedure, from absolute dependence on the middleman 
 for demand creation through every degree of coopera- 
 tion in selling to complete assumption by the manufac- 
 turer of the task of developing and holding a market for 
 his goods. A great number of branded and nationally 
 advertised food products illustrate this last stage; 
 trade-marked and advertised lines of clothing, tools, 
 household appliances, and the like occupy the interme- 
 diate positions marked by cooperative selling; while 
 thousands of commodities prepared by concerns which 
 limit their activities to production alone represent the 
 first extreme of marketing dependence. 
 
 Price and conditions of supply being equal, the mid- 
 dleman will naturally prefer to handle those products 
 which require the minimum of effort to move along to 
 the next circle in distribution, which necessitate the ex- 
 ercise only of those functions of assembling, assorting, 
 and reshipping which fall under physical supply. And 
 here lies the crux of the producer's problem. The sys- 
 tem of discounts to dealers which has been more or less 
 
 163
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 traditional assumes that the middleman performs his 
 share of all the functions of distribution. The middle- 
 man may readily consent to delegate all the work of 
 demand creation to the producer, but in doing so he 
 only reluctantly surrenders any considerable part of 
 his customary recompense. 
 
 On certain branded and widely advertised products, 
 the dealer perforce accepts a reduced unit profit, partly 
 because this is compensated by the large number of 
 unit sales and partly because the consumer allows him 
 no choice. When a customer asks for Ivory soap or 
 Holeproof hosiery, Kellogg's corn flakes or Uneeda 
 biscuits, the retailer must either supply the requested 
 product, sell a substitute against a definite prejudice, 
 or risk the customer's departure and a black mark 
 against his store. The wise dealer stocks the popular 
 brands, therefore, even though the profit per sale is less 
 than for a non-advertised substitute. The difference is 
 his contribution to the manufacturer's campaign of de- 
 mand creation, though he does not always recognize it 
 as such. 
 
 The wholesaler in his turn feels the pressure of the con- 
 sumer's demand when the retailer orders ; he also makes 
 contribution to the demand-creation fund of the pro- 
 ducer. In the case of established, advertised brands, 
 middlemen as a rule accept the situation; to new trade- 
 marked products they are generally less hospitable. 
 With both, many of them put all their emphasis on the 
 promotion of competing articles which they themselves 
 control or which carry a higher margin of profit. Con- 
 sumers can often be influenced to buy these. 
 
 That the wholesaler and retailer should expect pay- 
 ment for a function which they have wholly or partially 
 
 164
 
 THE MIDDLEMAN 
 
 
 s . —J — 
 
 I I =1= 
 
 H< 
 
 
 
 
 
 
 
 
 J 
 
 
 1 
 
 1 
 
 
 5 
 
 
 £ 
 
 
 — I — 
 
 
 1! 
 
 
 3 
 
 
 — a — 
 
 
 
 
 g — 
 
 
 
 £ 
 
 
 « 
 
 
 =s- 
 
 1 
 
 
 1* 
 
 
 
 — a — 
 
 
 
 =Je 
 
 
 i2 2 
 o S 
 
 Sis 2 
 
 165
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ceased to perform is explicable on the ground of tradi- 
 tion. The middleman has had a long-established though 
 shifting position in the development of organized dis- 
 tribution. It is true that the history of his functions 
 has not been adequately studied and that only tentative 
 outlines of his evolution can be drawn. Pending the 
 fuller study which the subject should receive, even a cur- 
 sory and incomplete sketch may assist in giving the 
 manager a sense of direction. 
 
 The middleman, as we know him, is a product of a 
 complex industrial organization. Chart I attempts to 
 give in graphical form a rough notion of his evolution 
 from the remote time when producer dealt directly with 
 consumer to the rise of the factory system in England 
 and the development late in the eighteenth century and 
 early in the nineteenth of what may be called the ortho- 
 dox type of distribution. Under the primitive rule of 
 barter, the producer became a trader only at intervals 
 when he had accumulated a surplus of skins or cattle or 
 hunting gear and conceived these as of less value than 
 something a neighbor possessed. We find him as a 
 craftsman, traveling from house to house to make up 
 the consumer's own raw material into shoes or harness 
 or clothing, or attaching himself to the household of 
 some nobleman. Then as the crafts became specialized 
 and settled occupations and the town appeared, the 
 craftsman producer took on the functions of retailer, 
 making to order or on occasion selling directly from 
 stock. 
 
 As the market widened and the merchant appeared as 
 an organizer of trade, the producer concerned himself 
 solely with making things, in many cases becoming 
 practically an employee of the merchant who provided 
 
 166
 
 THE MIDDLEMAN 
 
 the stock and capital, took all the risks entailed, and 
 sold the jSnished goods in his own shop and at neigh- 
 boring town markets. Steadily this system of trading 
 widened until there was a metropolitan market for goods 
 of certain recognized types and qualities. While on one 
 side the merchant became more definitely a wholesaler, 
 disposing of the goods he took from the producer to dis- 
 tant retail merchants who in turn distributed them to 
 the consumers, on the other side the merchant-pro- 
 ducer gradually strengthened his financial position and 
 grew in importance. Assuming all the risks of pro- 
 duction, he distributed his product through various 
 wholesalers who sold in turn to retailers, the latter sup- 
 plying the consumers. As a world market appeared, 
 the producer disposed of a part of his product to the 
 exporting and importing houses which carried domestic 
 goods beyond the seas and brought back foreign luxuries 
 and necessities for home consumption.^ 
 
 ^ In Defoe's " Tour of Great Britain," -wTitten about 1725 (Volume II, 
 Letter II), also quoted in C. J. Bullock's "Selected Readings in Economics," 
 beginning on page 325 we get vivid and entertaining glimpses of the activi- 
 ties of both merchant wholesalers and retailers full two generations before 
 the industrial revolution gave its tremendous impetus to English trade. In 
 one chapter he reports the departiu-e of a merchant's representative for a 
 trip through the provinces, which might outlast a year, in such state and 
 splendor as no modern salesman could hope to emulate. In another passage 
 he describes the manner of retail and wholesale trade at the great Sturbridge 
 fair, one of the typical markets of the time, ^ath the same feeling for detail 
 which makes " Robinson Crusoe " seem a transcript from the experience of a 
 shipvsTecked sailor. Here is his picture of the retail section: 
 
 " It is impossible to describe all the Parts and Circumstances of this Fair 
 exactly; the Shops are placed in Rows like streets, whereof one is called 
 Cheapside; and here, as in several other Streets, were all Sorts of Traders, 
 who sell by retail, Brasiers, Turners, Milaners, Haberdashers, Hatters, 
 Mercers, Drapers, Peirt-terers, Chinawarehouses, and, in a word, all Trades 
 that can be found in London. I might proceed to speak of several other 
 Sorts of English Manufactures, which are brought hither to be sold; as all 
 Sorts of wTOught Iron, and Brass Ware from Birmingham, edged Tools, 
 Knives, etc. from Sheffield; glass wares, and Stockens, from Nottingham 
 
 167
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 After the advent of the factory system, producers 
 tended to lose their character as merchants and to de- 
 vote themselves wholly to the management of their 
 works and the invention and betterment of methods 
 and machines. The chief reason for this, as we saw in 
 an earlier chapter, was that the market constantly out- 
 ran production. The problem was not so much to create 
 a demand as to manufacture goods to supply it. The 
 pressure was all to increase the rate of output. With 
 the division of labor and the growing intricacy of oper- 
 ations, manufacturers found it both necessary and 
 profitable to concentrate their attention on production. 
 
 To relieve them of the alien task of merchandising, 
 the selling agent appeared as a new link in the chain of 
 distribution. He undertook to dispose of the total 
 product of the factory to wholesale merchants. These 
 in turn supplied retailers, and the retailers the consum- 
 ing public. So long as the demand exceeded output, 
 this arrangement satisfied all hands; but as mill capac- 
 ity outgrew the market, the producer discovered him- 
 
 and Leicester; and unaccountable Quantities of other things of similar 
 Value every Morning." 
 
 His account of the wholesale market has added interest because it sets 
 forth as an established trading process what must have been sale by sample, 
 if, indeed, it was not that most modern of merchandising expedients, sale by 
 description. And this was a century before the widespread use of power 
 machinery in the industries made uniformity of product easy of attainment. 
 
 " Besides," Defoe writes, " the prodigious Trade carried on here by 
 Wholesalesmen from London, and all Parts of England, who transact their 
 business wholly in their Pocket-books; and, meeting their Chapman from 
 all parts, make up their accounts, receive Money chiefly in Bills, and take 
 orders. These, they say, exceed by far the Sales of Goods actually brought 
 to the Fair, and delivered in Kind; it being frequent for the London Whole- 
 salesmen to carry back orders from their Dealers, for 10,000 Pounds-worth 
 of Goods a Man, who deal in heavy Goods, as Wholesale Grocers, Salters, 
 Brasiers, Iron merchants. Wine-merchants, and the like; but does not ex- 
 clude the dealers in Woolen manufacturers, who generally manage their 
 Business in this Manner." 
 
 168
 
 THE MIDDLEMAN 
 
 self at a serious disadvantage. The outlet for his goods 
 was controlled by middlemen; but while he had been 
 designing machinery, reducing costs, perfecting quality, 
 the middleman had intrenched himself with the trade as 
 the source of its supplies. The manufacturer, isolated 
 from the consumer and with fixed charges which com- 
 pelled him to operate continuously or lose money, 
 found the intervening middleman exerting pressure to 
 force him to accept a narrow margin, even at times to 
 operate without any profit at all. 
 
 The middleman, too, was the creator and definer of 
 demand. Any attempt of the manufacturer to differ- 
 entiate his product and thus lift it out of the ruck of 
 staples was subject to the approval of the middleman, 
 who might veto it because it introduced an untried ele- 
 ment into his selling plan. To escape this limitation, 
 the stronger producers in many lines have assumed the 
 neglected function of demand creation and have sought 
 ways of going around the selling agent, the wholesaler, 
 and even the retailer in order to establish a closer con- 
 tact with consumers of their goods. 
 
 The development of this tendency to reduce the num- 
 ber of factors intervening between the consumer and 
 the sources of the goods he needs is one of the dis- 
 tinguishing characteristics of modern business. Pro- 
 ducers assert that the movement has been an effort not 
 so much to get rid of the middleman as to eliminate 
 payment for functions which the middleman no longer 
 performs. The question narrows down, they declare, 
 to this : Shall there be double pay for a single perform- 
 ance of the function of demand creation .f^ Since the 
 product pays the charges, it is vital for them to dis- 
 cover a more economical way of creating and supplying 
 
 169
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 170
 
 THE MIDDLEMAN 
 
 demand than through the orthodox channels, if such a 
 way exists. It is not, as we shall see, a narrowly eco- 
 nomic problem, but one in which the human element 
 is frequently preponderant. 
 
 Salesmen and advertising have been the agencies 
 of the producer's gradual approach to the consumer. 
 Chart II illustrates the growth of the tendency to drop 
 out successive middlemen. The first step is to secure 
 a measure of independence from the selling agent 
 without actually parting company, by sending sales- 
 men to call upon wholesalers, to show samples, get 
 suggestions and book orders, and by advertising di- 
 rected at retailers (and sometimes at the jobber as 
 well) to exploit the goods before they are ready to be 
 shown and thus create a demand for them in anticipa- 
 tion of the market season. 
 
 In the next stage of the evolution the merchant- 
 producer takes on more and more of the functions of 
 demand creation, both primary and secondary, sup- 
 plementing the salesmen who call on wholesalers with 
 others who carry the work of education down to the 
 retailer and extending his dealer advertising to take 
 in the consumer. This is the intermediate stage to 
 which the great mass of progressive producers have 
 advanced at present, when the " specialty salesman " 
 is almost as common a visitor in the larger retail store 
 as are the representatives of the jobbers from whom 
 the dealer buys. " Specialty " in the dry goods, 
 grocery, hardware, and shoe trades usually means a 
 differentiated and trade-marked staple. No matter 
 who books the orders, they usually pass through the 
 regular jobber's hands. The latter usually acquiesces 
 in this plan because it relieves his road men of the 
 
 171
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 burden of selling and gives them more time to push 
 other lines without sacrificing profit. 
 
 Often this results in another step in the evolution of 
 the merchant-producer. Having direct and satisfactory 
 connections with the retailer, the producer decides to 
 eliminate the wholesaler and save a discount, especially 
 if his product has no great weight or bulk and reorders 
 can be forwarded quickly by express or parcel post. He 
 keeps in touch with his dealers through salesmen and 
 advertising, using the latter and frequently the former 
 to stimulate and renew the consumer's appreciation of 
 his goods. If he has a widely advertised and very de- 
 sirable line, he can even demand a contract with a sales 
 quota guaranteeing a certain volume of orders, and can 
 secure the adoption of those methods of display, pro- 
 motion, and selling which experience has proved best 
 for his products. 
 
 From this vantage ground the manufacturer can 
 make one further advance, to complete independence of 
 middlemen and absolute control of his trade. He can 
 establish his own retail stores — witness the success of 
 certain notable shoe and tobacco " chains." With a 
 smaller outlay of capital he can make exclusive con- 
 tracts with going concerns, as in the drug trade, which 
 insure his goods against competition in that store 
 through a profit-sharing clause. Again, he can take the 
 mail-order route, using general and direct advertising to 
 reach consumers and create his demand, and railroads, 
 express companies, and the parcel post to deliver his 
 merchandise. If he be a manufacturer of another class 
 of " specialties," such as machinery or labor-saving de- 
 vices for the oflBce, factory, store, farm, or household, 
 which require a high order of intelligence and skill to 
 
 172
 
 THE MIDDLEMAN 
 
 sell them to selected prospects, he will find it economi- 
 cal to employ both salesmen and advertising in his 
 work of demand creation. 
 
 This tendency to decrease the number of distinct 
 ownerships through which goods pass on the way to the 
 consumer will probably show further development in 
 the future unless new factors appear. The attempts 
 of wholesalers and retailers to check the growth of 
 direct selling or semi-direct selling through the mail- 
 order houses, while in many cases raising the efficiency 
 of the middlemen themselves, have not seriously in- 
 convenienced the rival agencies. Concerted action to 
 force the producer to dispose of his output through 
 regular trade channels has taken at times the form of a 
 virtual boycott. But even if this were of itself the 
 most effective method, the attitude of the federal and 
 state governments would probably discourage it. 
 
 Much more significant is the movement among job- 
 bers to establish their own house brands, thus reviv- 
 ing the conditions which obtained in production and 
 distribution before the manufacturer sought emancipa- 
 tion from the middleman. Neither this nor the re- 
 markable growth of the chain store nor the growth of 
 the cooperative trading associations should be left out 
 of the manager's calculations. Buying associations 
 have been organized by the " regular " retailers of many 
 cities and districts to meet the new condition. How 
 nearly each of these systems will balance the others and 
 what effect antagonistic factors will have on the final 
 adjustment of distribution are still matters of spec- 
 ulation. 
 
 For the individual manufacturer, however, they have 
 a practical aspect. In many localities the chain store 
 
 173
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and the retail buying association present immediate 
 problems. Is it sound policy, for instance, to ignore the 
 tremendous distributive power of the " chains " in a 
 city like Philadelphia, St. Louis, or New York because 
 of the prejudice the regular trade holds against them.? 
 The sales manager of one of the large cereal companies 
 not long since declared that the profits v.hich his com- 
 pany sacrificed by refusing to sell to chain stores in 
 Philadelphia alone amounted to tens of thousands of 
 dollars yearly. He put his problem and the problem of 
 every manufacturer whose product by its nature must 
 be handled through middlemen into a single query to 
 the trade. 
 
 " Will you," he asked in substance, " cooperate with 
 us and push our breakfast foods a little harder to make 
 up for a loss which is due to our policy of protecting your 
 interests in this respect? Or shall we throw down the 
 bars to chain stores and take all the orders which rightly 
 come to us.? " It happens that this cereal company has 
 been most scrupulous in avoiding sales policies which 
 were not frank and fair to middlemen. This extraor- 
 dinary appeal of the sales manager, therefore, would 
 seem to indicate that in the orthodox machinerj^ of dis- 
 tribution there is sometimes serious leakage of good 
 will and the spirit of cooperation. 
 
 This orthodox system is organized on a basis of 
 hastened profit. In normal seasons the selling agent, in 
 touch with the market and its current wants or fancies, 
 takes over the producer's output at a price which allows 
 him to dispose of it to wholesalers at a substantial ad- 
 vance, yet stUl leaves the jobber a satisfactory margin 
 when the goods are moved along to the retailer. Again, 
 the dealer's inducement to purchase is not primarily the 
 
 174
 
 THE MIDDLEMAN 
 
 quality or service the merchandise offers, but the oppor- 
 tunity to resell to the consumer at a profit. 
 
 Not until the storekeeper displays the commodity 
 before the customer or considers what shall be the ap- 
 peal of an advertisement does the stress fall on style, 
 utility, durability, or service as a buying argument. 
 In brief, the demand-creating ideas about the goods 
 vary at different stages in the process of distribution, by 
 reason of the different viewpoints of those who buy for 
 resale and those who buy for use or consumption. Price 
 and salability are the essential factors to the middle- 
 man; with the consumer, quality and service must 
 balance cost, or dissatisfaction ensues. 
 
 Analysis, not indictment, of current practice in the 
 distribution of unbranded commodities is the purpose 
 of these paragraphs. I do not mean to convey the im- 
 pression that wholesalers and retailers generally con- 
 sult no other standards than price and salability when 
 they lay in their stocks. Within certain limits price 
 is here also a question of comparative values; and the 
 merchant who does not employ some measure of value 
 in his buying will soon have the mistake called to his 
 attention by the complaints of his customers or their 
 neglect of his offerings. He may capitalize for one or 
 two seasons an earned reputation for fair prices and 
 foist on his trade goods of inferior quality ; or if he be a 
 retailer with a store in a large city, he may continue this 
 policy indefinitely because he has a mass of fresh pros- 
 pects to draw upon. As a matter of fact, the most suc- 
 cessful wholesale houses and retail stores in the country 
 are those which apply rigid tests for quality and service 
 to the commodities they purchase, both for their own 
 protection and for the benefit of their customers. 
 
 175
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Before buying goods not trade-marked they go so 
 far as to inspect minutely the factories where they are 
 made, the raw materials, the processes. They count 
 the number of threads to the square inch in textiles, 
 the number of knots in carpets. They use chemical tests 
 to determine the proportion of cotton in mixed silk and 
 woolen goods. They expose colored fabrics to the action 
 of sun, wind, and weather to determine the permanency 
 of the dye and their wearing qualities. They sample 
 canned goods and packaged foods of all kinds to make 
 sure of their character and freshness. They even employ 
 production engineers to analyze the items in their prin- 
 cipal lines and estimate the factory cost of each as 
 checks on the prices quoted by the manufacturers. 
 
 Because of training and superior intelligence, too, 
 their buyers usually are keener judges of value than the 
 factory inspectors who spend their lives passing on the 
 same kinds of goods; and this knowledge and judgment 
 is exercised every time a new sample comes under their 
 eye or a new consignment is received in the stock rooms. 
 The big department stores recognize three grades of 
 quality in their major lines, a best, a medium, and a low- 
 priced line, but no effort is spared to make each grade 
 represent the utmost of value which can be bought 
 within its price range. The average wholesaler or re- 
 tailer is content to supply full lines in two price ranges, 
 grading upward or downward from the medium ac- 
 cording to the character of his trade. The measure of 
 value in many instances is based on personal judgment, 
 experience, and rule-of-thumb, but it is applied with 
 painstaking care. 
 
 Thus equipped, the large retailer, knowing that his 
 customers look to him to make good any defective mer- 
 
 176
 
 THE MIDDLEIMAN 
 
 chandise, even when it carries the maker's name, fre- 
 quently takes the final step of assuming all responsi- 
 bility and finds an anonymous manufacturer to turn out 
 his clothing, hardware, toilet articles, silverware, shoes, 
 or what not, according to specifications and at a lower 
 price usually than that which the trade-marked product 
 commands. 
 
 Summed up, this struggle for integration — the inte- 
 gration of industry on the one hand, and opposing it the 
 integration of trade — hinges on the division of profit. 
 The manufacturer is sometimes forced, as he views it, 
 to take over the function of demand creation because 
 he sees the work neglected or ineffectively performed. 
 Naturally he is averse to paying the middleman for 
 something the latter no longer is required to do for his 
 product, though he recognizes that jobbers and dealers 
 are effective instruments of physical supply and is 
 quite willing to pay them for the performance of this 
 function. On the other hand, so long as they continue 
 to be paid for creating a demand for his product he 
 naturally desires to see them actually and adequately 
 doing so. 
 
 177
 
 CHAPTER XII 
 
 AGENCIES OF DEMAND CREATION — DIRECT 
 SALESMEN 
 
 WHEN a manufacturer considers the problem of 
 building permanent demand for his product, he 
 looks first to those elements of quality and service 
 which will render his goods more desirable to the con- 
 sumer and more likely to be bought repeatedly as the 
 latter reenters the market. Without neglecting the fac- 
 tors which appeal to the middleman, since the latter 
 must be reckoned with in innumerable lines as the es- 
 tablished agency of physical supply, he adds as much 
 of style, finish, or special utility as his cost and produc- 
 tion conditions will permit and takes steps to insure that 
 his entire output shall conform to this standard. 
 
 If he trade-marks the resulting product and its supe- 
 riority to competing commodities is evident, he can 
 count on a certain proportion of repeat orders. Con- 
 sumers who have bought it will recall its excellence and 
 will try to secure it when they have need of such goods 
 again. The middleman also can be trusted to make an 
 effort to stock it in order to satisfy the consumer's 
 desire to buy. 
 
 He finds, however, that at first the chances of his spe- 
 cialized product suffer in about the same degree as its 
 adaptation to the specific needs of the final user (often 
 latent or unrecognized needs) differentiates it from the 
 bulk of similar products. A soap for washing woolens 
 without shrinking them is likely to prove more difficult 
 
 178
 
 DIRECT SALESMEN 
 
 to market through the regular channels, for instance, 
 than would a laundry bar of the conventional size, shape, 
 and color. First, because its specific claims are likely to 
 inhibit its purchase and use for any but this particular 
 purpose. And, second because the " how " and " why " 
 of its non-shrinking properties, to be effective in the 
 creation of demand, must be conveyed accurately and 
 vividly to the minds of prospective users. 
 
 Orders from wholesalers and retailers, therefore, will 
 not entirely bridge the gap that separates the article 
 and the consumer. They and their salesman, as a rule, 
 do not have sufficient time either to master its essential 
 selling points or adequately to present them to their 
 customers. The qualities which appeal actively to mid- 
 dlemen, who buy for resale, are often quite distinct from 
 those which attract the consumer, who is directly in- 
 terested in value and service as well as appearance and 
 price. The ideas that the retailer must communicate 
 to the consumer, therefore, in order to create in him a 
 desire for the product are not necessarily the ideas with 
 which the wholesaler interested the retailer. Not until 
 the latter addresses himself to the consumer is emphasis 
 laid exclusively upon the special features which the 
 manufacturer is counting upon to make the final sale. 
 
 This does not mean that wholesalers or retailers are 
 indifferent to conspicuous quality in a product when it 
 does not bear an established trade-mark or a quality 
 price ticket. Quite the contrary. Such " good buys " 
 strengthen any merchandise line or department stock 
 and add luster to house and personal reputations. Fre- 
 quently, too, they provide opportunities for better-than- 
 regular mark-ups and corresponding immediate profits. 
 
 But the great hindrance to maximum sales for a spe- 
 
 179
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 cialty through orthodox channels is that it is difficult 
 by this system to secure the individual handling at every 
 stage of demand creation which a differentiated product 
 requires. The jobber's salesman has hundreds of items 
 to display to each dealer in the course of a fev>^ hours. 
 The retail clerk's value is usually measured by his sales 
 totals and his skill in serving a maximum number of 
 customers in a limited time. The transmission to the 
 consumer, therefore, of the ideas necessary to sell any 
 given specialty call for the intelligence, efficiency, and 
 good will of at least four persons, the jobber and his 
 salesman, the retailer and his clerk. If any one of the 
 four is careless or indifferent or incompetent, the chain 
 of transmission snaps and the product is left to make 
 a dumb appeal to the consideration of the consumer. 
 
 For an increasing variety of non-staple articles, then, 
 the special salesman is recognized as an essential agent 
 in demand creation. If the product is simply a bet- 
 tered staple, retaining its familiar characteristics and 
 depending on refinement of design or finish to recom- 
 mend it, no particular urging is necessary to stimulate 
 the middleman's interest and secure his order. But any 
 advance in price or decrease in quantity operates against 
 this easy acceptance, even when its specialization gives 
 the product an added utility. 
 
 Where the character, quality, or use of the article is 
 not evident but requires demonstration or the develop- 
 ment of ideas unfamiliar to the trade and the consumer, 
 the work of transmitting its selling points to the final 
 user through two or more middlemen becomes difficult 
 and uncertain. And finally, the minimum of sales effort 
 is all that can be expected from the middleman, when 
 a trade-marked product is in direct competition with 
 
 180
 
 DIRECT SALESMEN 
 
 his " liouse brands " or with accustomed articles which 
 with a larger percentage of profit satisfy his customers. 
 
 The producer's problem, therefore, is to determine 
 whether a direct sales force is necessary to create max- 
 imum demand for his goods with a minimum outlay of 
 money and effort. Enough has been said to indicate 
 the sub-problems which he must address and the factors 
 to be considered in their solution. Too much stress 
 cannot be laid on the importance of a fresh point of 
 view. The great initiatives in distribution during the 
 last thirty years have all been breaks with merchandis- 
 ing traditions. The Butler system of wholesaling by 
 catalogue, for instance, was a new departure. So also 
 was the Larkin method of selling soap and similar house- 
 hold and personal requisites through consumers' clubs. 
 Another notable case was the long-range retailing of 
 staples and specialties by mail-order houses. These 
 revolutions in marketing custom have substituted ad- 
 vertising for personal solicitation of orders, the written 
 for the spoken word; yet at the same time the exclusive 
 salesman has become a most important agency of de- 
 mand creation not only for invented specialties but also 
 for innumerable lines developed from the staples which 
 were formerly distributed entirely through middlemen. 
 
 Whether advertising and the producer's sales force 
 have usurped the middleman's function of demand 
 creation or were evolved to perform a neglected task is 
 of no immediate moment. All three agencies exist and 
 are likely to continue active as long as individualism 
 holds its place in business. This chapter is concerned 
 primarily with the direct salesman and with the prob- 
 lems which his employment in demand creation set up 
 for the management. His relations with the other agen- 
 
 181
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 cies or activities of distribution, production, and ad- 
 ministration, because of their number and complexity, 
 can only be touched upon here. The broader question 
 of when and how the different agencies of demand cre- 
 ation can be used independently or in combination must 
 be reserved for discussion later. 
 
 Nowhere do the policies of production and distribution 
 seem to approach so closely as when the common human 
 agent in operation is considered. In an earlier chapter 
 dealing with the policies of production we saw that the 
 manager's labor problems have to do with four chief 
 contacts with his men: (1) hiring, (2) training, (3) pay- 
 ing, and (4) directing. The common purpose guiding 
 him in all these functions is to secure for each dollar ex- 
 pended the largest regular day-by-day return in pro- 
 ductive effort. But the workman is interested mainly 
 in the amount of money in his pay envelope; the success 
 of the manager, therefore, is measured by his skill in 
 utilizing this dominant motive to further his own pur- 
 pose. His method of fixing and paying wages may add 
 to their pocket value by making clear the definite rela- 
 tion between the sum paid and the amount and quality 
 of the effort received. Pride, ambition, loyalty may be 
 awakened; latent capacities may be developed. But 
 the range of motives to which the employer of factory 
 labor can appeal remains narrow; the wage system is 
 the pivot on which most of these must be swung. 
 
 The employer of salesmen has a larger margin of 
 psychological values (if the phrase is permissible) with 
 which to work. As a group or class, eflficient salesmen 
 have marked characteristics. The well-worn proverb, "A 
 salesman is born, not made," had as its kernel of truth 
 the fact that conspicuous success in selling, before 
 
 18^
 
 DIRECT SALESMEN 
 
 analysis was applied to salesmanship and training 
 courses were developed, was usually attendant on cer- 
 tain personal or temperamental qualities. 
 
 The typical salesman was ambitious, intuitive, quick- 
 witted, capable of enthusiasms, and sympathetic, or at 
 least was able to see his product or service from the 
 other fellow's viewpoint and in terms of the latter's 
 needs. He was willing to take a chance, not only on 
 the closing of an immediate or future order, but also 
 on the possibilities of a new connection for himself. 
 His function was to sell, to persuade prospects to buy his 
 merchandise in maximum quantities; and frequently 
 the lack of balance which comes of over-emphasis on 
 one function led him to oversell his customers. 
 
 Now all these qualities, tempered by experience and 
 training, and by a saner conception of customer service 
 survive in the salesman of to-day. Twenty years ago 
 they were the natural qualities which marked a man 
 as a potential salesman until some discerning employer 
 supplied him with a sample case, a price list, and a set 
 of mileage books. To-day they have been cultivated as 
 part of their selling equipment by thousands of men and 
 developed in other thousands by astute employers. 
 Whether native or acquired, they may almost be con- 
 sidered stock characteristics on which managers can 
 build plans to secure sustained and maximum efforts 
 from their field forces. 
 
 Not that the economic motive can be overlooked in 
 dealing with salesmen, any more than in handling fac- 
 tory workers or customers. But the economic element 
 is nearly always relatively constant; everyone must 
 pay the market rate of wages and be satisfied with the 
 approximate market price of his commodity. The ef- 
 
 183
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ficient business man, then, is likely to give his attention 
 to getting extra production or extra sales volume by 
 appealing to motives outside and beyond the money 
 involved. 
 
 It is a commonplace of management, indeed, that 
 added pay alone, unless it be a considerable addition 
 to the market rate, will not secure this desired extra 
 effort continuously. Mechanics and " handy men " will 
 work at top speed over long periods in plants like the 
 Ford motor works, not merely because the pay is good, 
 but largely because their jobs are coveted by so many 
 other men outside the organization. This outside pres- 
 sure reminds them constantly that their pay and con- 
 dition are exceptional and that exceptional service is 
 demanded of them. But only in extraordinary situa- 
 tions can the average man be prevailed upon for higher 
 day wages or a better salary to put forth all the produc- 
 tive energy he possesses. To tap his reserves, something 
 more than a money reward is required. 
 
 This is especially true of men who sell goods. The 
 typical salesman is keen about money. He insists on 
 receiving as much from his present employment as he 
 can command elsewhere. It is a point of pride with him 
 that he is paid as much or more than the average man 
 in his line and grade. He is more interested as a rule in 
 his gross income than in the net, a fact that must be 
 weighed in deciding his method of payment. If he re- 
 ceives a salary of $5,000 a year and an allowance of 
 $2,000 for expenses, he is likely to consider liimself 
 underpaid as compared with a friend who gets $7,000 
 a year salary but pays his own expenses. Straight com- 
 mission or a lump sum for salary and expenses, there- 
 fore, is likely to prove the more effective method of 
 
 184
 
 DIRECT SALESMEN 
 
 compensation, not only because the results bulk larger 
 in the salesman's eyes, but also because it supplies a curb 
 on his class extravagance, his liking for the best hotels, 
 the fastest trains, taxicabs, parlor cars, and the like. 
 
 But a fancy salary will not always obtain a salesman's 
 best efforts unless the connection between these efforts 
 and the consequent reward is visualized for him. That 
 is one reason why the high-grade specialty man nearly 
 always works on commission. The plan suits his tem- 
 perament. He is willing to take the risks involved in 
 return for the opportunity of independent action. And 
 the experience of manufacturers would seem to show 
 that no other method of payment gives such dependable 
 results and allows such varied appeals to the motives 
 which stimulate sustained effort. 
 
 The plan, it is true, has its drawbacks. Salesmen 
 working on commission are more difBcult to control, are 
 inclined to resent interference with their personal pro- 
 grams of work and play. Their time, they consider, is 
 their own. If they "let dowTi " in their pursuit of orders, 
 they know they pay for the lapse in lessened income. 
 That the house also loses in volume of sales is an inci- 
 dental thing. So vital is this matter of control that 
 many concerns prefer to pay salaries, depending on 
 regular advances backed up by daily reports and other 
 checks to insure a fair level of industry and initiative. 
 Bonus systems similar in principle to the Taylor method 
 of paying factory workers have also been developed for 
 the encouragement of salaried salesmen. Whatever his 
 method of payment, however, no manager can afford to 
 neglect the psychological appeals open to him for the 
 incitement of his men. 
 
 Ambition is primary among the motives which can 
 
 185
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 be enlisted. The salesman on commission, with a guar- 
 anteed territory, is virtually in business for himself. 
 Of every order closed he knows he will retain a fixed per- 
 centage for himself. The wise manager makes the most 
 of this fact in urging him to continued activity after he 
 has reached the monthly or seasonal volume which as 
 a rule he sets for himself. It is those plus orders, of 
 course, that make the difference between ordinary 
 profits and extra dividends. And it is usually the sales- 
 man's tendency to self-indulgence or some similar men- 
 tal handicap which stands between the management and 
 this " velvet." 
 
 Salesmen, as a rule, have individual conceptions of 
 the amount of money they should earn monthly or 
 yearly or during a trade season. One is a $2,500 man. 
 Another puts himself in the $5,000 class. A third fixes 
 his earning power at $10,000. And when they have 
 reached the self-appointed mark or its monthly frac- 
 tion, the tendency to self-indulgence is ordinarily so 
 strong that they slacken their efforts, begin to go to ball 
 games, quit their territories on Friday nights, and neg- 
 lect out-of-the-way or doubtful prospects. 
 
 These slumps in activity ordinarily come at the end 
 of the month or near the close of the selling period, when 
 the desired goal is within easy reach. But many sales- 
 men of the lower grade unconsciously adopt daily sched- 
 ules that include a certain amount of loafing because 
 they have learned that so many hours of " hustling " 
 will in the average give them the $25 or $30 or $40 a 
 week which they decide on as their standard of compen- 
 sation, or, if they are on salary, the volume of orders 
 which will satisfy their employers. 
 
 This indeed is the chief problem the manager of sales- 
 
 186
 
 DIRECT SALESMEN 
 
 men faces — how to keep each salesman in his territory 
 and at work after he has secured the volume of orders 
 which he has set for himself. 
 
 A manufacturer of my acquaintance has tried again 
 and again to get added endeavor by added payment. 
 Individuals have here and there responded, but as a 
 group the effort failed. Originally he paid straight 
 commissions. Reducing these commissions in certain 
 fertile territories, he found that his salesmen continued 
 to make about the same income as before. Apparently 
 their margin of leisure was large enough to allow con- 
 siderable increase in activity without making undue 
 demands on their time and energy. 
 
 But high levels in sales are not reached and main- 
 tained by cutting commissions, any more than maxi- 
 mum production is attained in the long run by lowering 
 the factory piece rates. So the manufacturer went at 
 the thing from the other angle. To hold his men and 
 get the best they had he would enable them to make 
 more money. So he increased commissions; but the 
 increase did not bring the results he expected. Instead, 
 the number and volume of orders went down — an 
 outcome that is difficult to understand or account for 
 unless you consider how strongly the inclination to 
 self-indulgence operates when the accustomed standard 
 of life has been attained. 
 
 Nor was this experience out of the common; it can 
 be matched by almost any business which pays its 
 salesmen on a percentage basis. For years the " in- 
 side " slogan of the organization which probably has 
 developed more high-grade salesmen than any other 
 company in America has been: '* Time not spent in the 
 presence of prospective purchasers is time lost." And 
 
 187
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 I once heard a great sales manager, now many times a 
 millionaire, put all the emphasis of an important sales 
 convention address on this same point. 
 
 *' When I was an agent myself," he said, " I started 
 each month with the thought that I was in debt until I 
 had earned commissions enough to meet my office, home, 
 and traveling expenses. I couldn't rest until I was even 
 with the board. Sometimes it took me ten working days, 
 sometimes three weeks or more to get square. But when 
 I did get square, I simply could not quit because I 
 realized that I was just beginning to work for myself. 
 In every dollar I made beyond that point one hundred 
 cents belonged to me. And what a fool I would have 
 been to waste one hour loafing or amusing myself, when 
 I had slaved the best part of the month to pay for this 
 opportunity to work for myself! " 
 
 That in substance was the message of a $50,000 sales 
 manager to two hundred agents, no one of whom could 
 retain his territory and earn less than five or six thou- 
 sand dollars a year. I quote it simply to show that sales- 
 men of all grades have " the defects of their qualities " 
 and that the unending problem in managing them is to 
 find and utilize the motives which will stimulate them 
 to do their best. There is even the danger, as suggested 
 above, that their compensation can be made so liberal 
 that it will defeat its purpose. 
 
 Personal pride and social emulation are motives gen- 
 erally appealed to in this vitalizing of sales effort. By 
 substituting for the individual salesman's conception of 
 what his territory should produce a sales quota propor- 
 tionate to its population or the number of live prospects 
 it contains, a sound basis is established for comparing 
 the personal records of all members of the field force. 
 
 188
 
 DIRECT SALESMEN 
 
 Then by means of prize contests open to all on these 
 equalized terms, the instinct of competition and the 
 desire of leadership are aroused in the interest of indi- 
 vidual efficiency. The prize itself is the least of the 
 incentives. In many cases it is no more than the print- 
 ing of the winners' portraits and the announcement of 
 their records in the house organ or in a circular letter. 
 Even when it takes the form of money or some article 
 of personal equipment, its chief value is as a symbol of 
 the distinction attained. 
 
 The sales contest is a thing so familiar that no analy- 
 sis of its technique or effects is needed here. The most 
 successful competitions have been those in which the 
 struggle was visualized as a game or sport, like a hun- 
 dred-mile motor race, a baseball or football game, a 
 flying machine race, or any other form of contest that 
 the season or the current news suggests as certain to 
 enlist the interest of the field force. 
 
 Frequently this individual competition is varied by 
 appeals to organization spirit. District is marshaled 
 against district, city territories against country terri- 
 tories, the East against the West, and so on. In one con- 
 cern, at least, the rivalry has been carried across inter- 
 national boundary lines. Before the present European 
 conflict began the big biennial competition was a three- 
 months' " war " between the American and the foreign 
 sales forces. Even here individual achievement was 
 the burden of the "news from the front"; the skill and 
 .courage of this or that agent or salesman in landing 
 large orders was acclaimed. 
 
 Coincident with these monthly and quarterly team 
 contests an individual competition was carried on 
 against a special yearly quota averaging forty per 
 
 189
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 cent above the regular quota. This contest was primar- 
 ily against a monthly " bogy " of one hundred points; as 
 soon as a man secures twelve of these, or orders total- 
 ing twelve hundred points, he takes his place among the 
 stars of the " Hundred Point Club." The instinct of 
 personal emulation was further enlisted by awarding the 
 presidency and the lesser offices in the club to the first 
 winners of the coveted title of " Hundred Pointer." 
 The material " Hundred Point " prize was a trip, free of 
 all expense, to the annual sales convention at the factory. 
 
 The average manager has to determine whether con- 
 ditions in his organization make it sound policy to carry 
 on such sales contests. It has been argued that sales- 
 men soon tire of them and refuse to respond to them. 
 As a matter of fact, they have been conducted month 
 after month and year after year by certain sales organi- 
 zations; success seems to depend on the character of the 
 contest and the vitality and tact of the management 
 rather than on novelty to the sales force. It must be 
 remembered, however, that not a few houses of high 
 standing object to the principle of pitting employees 
 against one another; while others find it difficult to 
 standardize conditions so that fair quotas can be es- 
 tablished. 
 
 In fixing quotas it must be borne in mind that some 
 salesmen are more militant than others, and that while 
 you are directing them in groups you must also take 
 their individual characteristics into consideration. The 
 quota which will stir the fighting qualities of one man 
 to supreme effort will merely discourage another; and 
 the figure which will be quite easy to secure in March 
 or October may be entirely beyond reach in July or 
 January. The most effective quota, therefore, is likely 
 
 190
 
 DIRECT SALESMEN 
 
 to be a compromise, a little higher than the average for 
 normal months, with a margin of increase in "rush" 
 periods and particularly favorable seasons. 
 
 \Miether it should be a standard based on population, 
 or the number of rated prospects, or a more flexible 
 quantity based perhaps on past performances of each 
 salesman, would depend on the number of men involved, 
 their relations with the house, and the advantages of 
 individual arrangements and individual treatment. Of 
 the wisdom of a carefully adjusted quota there can 
 hardly be a question, no matter what method of pay- 
 ment is in use. For one thing, it supplies a standard 
 of industry and ability for each salesman, it fixes a def- 
 inite volume of sales for his territory instead of de- 
 pending on his hazy personal notion of what he ought 
 to earn. And again, it gives a more accurate statistical 
 basis for the fixing of salaries and commissions and for 
 the comparison of individual records. 
 
 Mention has been made of a bonus system of pay- 
 ment for salesmen adapted from the Taylor and Gantt 
 systems described in Chapter VI of this book. These 
 have taken varying forms, from the primitive salary- 
 and-commission method to more scientific plans. The 
 fundamental principle in all is recognition of the excess 
 profit in sales made over and above the ordinary volume 
 of the industry. On his quota of this ordinary volume 
 the salesman receives the usual commission or salary; 
 on all sales above this quota he is paid at a much 
 higher rate. 
 
 Certain companies have gone further, working out for 
 their salesmen a close parallel of the Taylor differential 
 piece rate, with marked increases in commissions as the 
 bonus standard is reached and passed and corresponding 
 
 191
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 lowering of the rate when the salesman falls short. Here 
 daily quotas and bonuses keep the advantages of appli- 
 cation constantly before the individual. But such an 
 intensive use of the motives of utility and money gain 
 is practicable only in lines where the salesman has a 
 number of regular customers or definite prospects to 
 see every day and where the demand for the product is 
 continuing and susceptible of increase. 
 
 Fearing over-emphasis upon these group methods of 
 control, payment, and stimulation, there is another 
 school of management which puts its faith in individual 
 analysis of the salesman's character and abilities first, 
 followed by individual training and placing in the ter- 
 ritory to which each man seems best adapted, and in- 
 dividual managing when the latter has taken hold. 
 Broadly speaking, this is the method of the successful 
 old-school manager, who frequently is an uncommon 
 judge of men and has sufficient force and personality 
 to command the loyalty and enthusiasm of his sub- 
 ordinates. 
 
 In such cases, however, the securing and holding of 
 salesmen is largely a process of " trial by error "; if the 
 man fails to " make good," he is dropped and another 
 candidate takes his place. In a small organization, this 
 is a practicable method, particularly when the direction 
 of sales is in the hands of the man who built the busi- 
 ness. In a large sales force, it might easily lead to dis- 
 organization, loss of sales, heavy expense, and sacrifice 
 of prestige and good will, unless it were backed up by an 
 efficient system for taking care of records and all the 
 routine activities of selling. Supplemented in this 
 fashion, individual hiring, training, and managing of 
 salesmen, without attempting any group incentive, is a 
 
 192
 
 DIRECT SALESMEN 
 
 practical ideal which has been realized with rather large 
 forces by sales directors who concentrate on this one 
 function and leave routine to assistants. In such an 
 organization records and results are considered only as 
 they demonstrate individual selling efficiency. 
 
 Whether he shall stress group or individual manage- 
 ment of his salesmen, therefore, is a question which 
 each executive must settle for himself. He will first 
 eliminate the personal equation so far as is possible, 
 then he will break the main problem up into its con- 
 stituent problems, analyzing these and assembling the 
 factors which influence them, and he will adopt a point 
 of view independent of tradition, trade customs, and 
 even personal experience. The decisive factor may be 
 the character of the salesmen available or actually at 
 work, the trade requirements which must be satisfied 
 in marketing his product, the quality, price, simplicity, 
 or complexity of that product, or any one of a score of 
 conditions or combinations of conditions which it is 
 imperative to consider. 
 
 Whatever his choice, he will face certain stock prob- 
 lems which every sales manager faces. These problems 
 arise from the ambition of salesmen to make records, 
 from their readiness in particular cases to rely on their 
 own judgment rather than to observe the house policy, 
 from the lack of balance which often goes with the 
 valuable quality of enthusiasm, from the tendency to 
 oversell or unconsciously misrepresent the product to 
 customers, from the narrow view of sales possibilities 
 which holds to accustomed paths in the search for 
 prospects, from the itch for change which makes the 
 investment value of a new salesman an uncertain quan- 
 tity, from the lack of self-discipline which crops out in 
 
 193
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 careless expense accounts and in the slighting of tasks 
 which have no immediate influence on sales or com- 
 pensation. 
 
 It is problems like these that put a premium on train- 
 ing courses and carefully organized systems of super- 
 vision, which leave nothing to the salesman's initiative 
 except the actual face-to-face dealing with customers. 
 So much has been written about the schooling of sales- 
 men, and the methods employed are so familiar, that 
 they are referred to here only to emphasize the need of 
 making this training as comprehensive as it can be and 
 of continuing it as long as the salesman represents the 
 company. 
 
 To acquaint him with all your ideas about your goods 
 is naturally the first purpose, since it is his function to 
 transmit these ideas to consumers or to the middlemen 
 who will aid in distributing your product. This might 
 be a mere memory exercise for him, and for you the 
 gathering and compiling of a manual of selling points. 
 Only this is not enough. To reduce a mass of informa- 
 tion to an orderly store of living knowledge, each idea 
 instantly accessible when occasion arises for its use, 
 requires on your part careful analysis and logical ar- 
 rangement of the material, and intelligent study and 
 practice on his. 
 
 Goods can be sold and are sold by the million dollars' 
 worth without either. It is merely a question of your 
 taking the pains to analyze your product and to prepare 
 beforehand an approach and a series of selling points 
 which are convincing, or else allowing your salesman to 
 make the same analysis in the presence of a dozen or a 
 hundred or perhaps a thousand prospects before he 
 learns all the possible contacts and reactions and is able 
 
 194
 
 DIRECT SALESMEN 
 
 to perfect his exposition. If you leave it to his initia- 
 tive, he may be years discovering and putting to use 
 the selling points which a month or two of schooling at 
 the factory or in selected training ground would have 
 given him. And every business man who has bought 
 merchandise or equipment for any length of time has 
 met salesman after salesman whose goods had to sell 
 themselves in spite of an awkward approach and blun- 
 dering demonstration. 
 
 I am not advocating the crude, ready-made canvass 
 and selling talk which is as ineffective as it is common. 
 But experience both as seller and buyer and some ac- 
 quaintance with the best current sales practice have 
 persuaded me that every step in the selling process can 
 be standardized, from analysis of the product and its 
 market to the closing argument which clinches the 
 order. 
 
 All that was said in a previous chapter about the de- 
 velopment of laboratory standards as applied to the 
 materials of demand creation holds true for face-to-face 
 selling as well as for advertising. Not only can the 
 most effective arguments be determined, but also the 
 most effective order in which they can be brought to 
 bear upon the prospective customer. There may be 
 forty cogent reasons, for example, why the detail sales 
 strip in a cash register is a record of value to various 
 merchants, yet only four of these may be of direct in- 
 terest to a certain storekeeper. The salesman's task is 
 to discern the special conditions which govern in the 
 case and to put forward the selected arguments of 
 greatest weight. 
 
 It is obvious that a standard selling talk could not be 
 framed to fit equally well all the store situations and 
 
 195
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 personal idiosyncrasies which an equipment salesman 
 will encounter. But if he has a logical method of ap- 
 proaching his prospect and uncovering the latter's busi- 
 ness needs and mental attitude, if he can draw from a 
 store of tested selling points the illustrations and argu- 
 ments which fit the case most exactly, and if he knows 
 the proper sequence in which these should be intro- 
 duced to bring his prospect to the point of buying, 
 his chances of landing the order are decidedly and 
 measurably increased. 
 
 Between such a skilled approach and demonstration, 
 with its definite plan and its wide latitude in the use of 
 selling points, and the parrot-like recital of standard 
 sales talk there is all the difference in the world. Part 
 of this difference lies in the faulty analysis and scanty 
 materials on which the latter is built; a large part, how- 
 ever, lies in the difference in the training given the sales- 
 men — a careful and persistent instruction, with the 
 development of mental poise and initiative on the one 
 hand and the hasty cramming of a sales formula on 
 the other. 
 
 Not every business needs or is able to develop train- 
 ing courses for salesmen like those which the large spe- 
 cialty companies maintain. Products may be so simple 
 or the items in the line manufacture so few that an 
 elaborate analysis would cost more than it would make. 
 Customers and prospects may be confined to a single 
 trade and be so nearly of the same type that a salesman 
 of moderate resource can deal with them satisfactorily. 
 As the product becomes complex or the line longer or an 
 expanding market takes in prospects engaged in many 
 different kinds of business, the value of analysis, classi- 
 fication, and organization of your materials of demand 
 
 196
 
 DIRECT SALESMEN 
 
 creation increases rapidly and the necessity of a defi- 
 nite method of instruction grows. 
 
 Even where a formal school for salesmen is out of the 
 question, other means are successfully employed for 
 their training. The commonest and the most effective 
 in small organizations is personal attention on the part 
 of the sales manager, both in the " breaking in " of new 
 salesmen and in the aid and supervision given to sea- 
 soned members of the force. The house organ and the 
 sales convention are also widely used. The former 
 ranges from a weekly or monthly mimeographed record 
 of individual sales, with a " ginger talk " and new selling 
 points added, to admirably edited and printed weekly 
 or monthly' magazines, full of new facts about the prod- 
 ucts, personal mention of salesmen and their achieve- 
 ments, and inspirational messages and editorials. When 
 contests of the spectacular sort are on, the house organ 
 frequently becomes a daily paper, with every appeal to 
 the social emulation, ambition, and sportmanship of 
 the individual salesman intensified. 
 
 Every possible change, too, is rung on the sales con- 
 vention. There may be daily and weekly meetings of 
 city salesmen, monthly gatherings of district forces, 
 annual visits to the factory for three days or a week of 
 instruction, with interchange of ideas, good fellowship, 
 and judicious cultivation of organization spirit and per- 
 sonal enthusiasm. 
 
 To do more than indicate the broad problems and 
 policies involved in the hiring, training, and manage- 
 ment of salesmen would be impossible in a single chap- 
 ter. If I have seemed to slight the problems of hiring, 
 it is because there is less of agreement among business 
 men and psychologists as to many of the specific qual- 
 
 197
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ities which make a salesman and as to the tests which 
 reveal these qualities than in almost any other depart- 
 ment of business. 
 
 Certain qualities which are recognized as fundamental 
 have been suggested earlier in this chapter. But even 
 here dependable scientific tests to determine whether 
 or not a salesman possesses them have not been devel- 
 oped. Promising experiments and investigations are 
 under way by the psychological departments of several 
 universities in cooperation with industrial concerns. 
 The results thus far are meager, however, and the aver- 
 age business man in hiring a new salesman must still 
 depend on the old process of a personal " sizing up," 
 checked up by the record of past performances and ref- 
 erences as to the subtler qualities of character which 
 limit or enlarge the salesman's usefulness in demand 
 creation. 
 
 No methods of hiring, training, paying, and direct- 
 ing, however, can result in satisfactory relations with 
 salesmen unless backed by a certain human friendli- 
 ness and understanding. If this quality is present, the 
 salesman inevitably senses it; nor can any amount of 
 simulation long conceal its absence. Analyzed, it may 
 be found largely a keen perception of the salesman's 
 pleasure in effective functioning, and of the house's in- 
 terest in supplying him the most favorable conditions. 
 But it must include a generous measure of simple good 
 will. Without this, true loyalty and cooperation are 
 not to be expected. 
 
 198
 
 CHAPTER XIII 
 AGENCIES OF DEMAND CREATION — ADVERTISING 
 
 A DVERTISING is the communication of ideas about 
 -^*- the goods to possible purchasers by means of writ- 
 ten or printed symbols. As in the transmission of like 
 ideas through middlemen and through direct salesmen, 
 its purpose is to create a demand for a product or to 
 divert a demand already existing. It has been devel- 
 oped not solely to take the place of the middleman or 
 salesman in demand creation but as a means of doing 
 quickly, cheaply, and efifectively much of the work 
 these other agencies have done in the past and much 
 which neither could profitably undertake at present. 
 
 Like them, it has its limitations and its spheres of 
 special utility where it satisfies all selling requirements. 
 Apart from these specific fields — mail-order selling, for 
 example — the manager's problem is to determine what 
 part of his work of demand creation can be done more 
 cheaply or more efficiently by advertising than by his 
 sales force and what kind or class of advertising is best 
 adapted to perform each of these tasks. There is the 
 further problem of coordinating the different elements 
 in his selling plan, in order to eliminate the duplication 
 of effort which pyramids distribution costs to-day. But 
 that is matter for a succeeding chapter on organization. 
 Two broad classifications of advertising are recog- 
 nized — general and direct. The first aims to find new 
 prospects either addressing the broader public through 
 
 199
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 magazines, newspapers, painted or electric signs, posters 
 or street-car cards, or appealing to a special social 
 stratum or a professional or business group through 
 the class periodical or trade journal. 
 
 The second seeks to turn known prospects into pur- 
 chasers and to cultivate good will and stimulate in- 
 creased buying among regular and occasional customers. 
 Various mediums are used, either independently or 
 in combination — personal letters, booklets, circulars, 
 and mailing cards, house organs, catalogs, premiums, 
 and novelties. Direct advertising is intensive advertis- 
 ing in that it is aimed at individuals selected from the 
 mass, whose needs, tastes, and inclination to buy have 
 been fairly well established and each of whom has a local 
 habitation and a place on the mailing list. 
 
 For all practical purposes, both direct and general 
 advertising can be treated as one phase of sale by de- 
 scription. All advertising, indeed, is a logical outgrowth 
 of sale by description. So long as prevailing commercial 
 ethics made sale in bulk the only practical method of 
 distribution, the middleman was indispensable. As 
 business morals bettered and manufacturing methods 
 improved so that standardization of products was pos- 
 sible, sale by sample appeared. The producer found that 
 he could send his own salesman to the prospective pur- 
 chaser instead of depending solely upon the selling 
 efforts of a middleman to obtain an outlet. 
 
 When sale by description appeared, with a still higher 
 code of conduct and a higher level of general intelli- 
 gence, a third selling agency took form in advertising. 
 Its development as an important tool of business falls 
 well within the last fifty years, with its significance 
 greatly increased toward the end of the period. In ad- 
 
 200
 
 ADVERTISING 
 
 vertising, as in selling through salesmen, it is the com- 
 munication of ideas about the goods to the prospective 
 purchaser that creates demand. When the purchaser 
 insisted on seeing and testing the actual goods before 
 purchasing, sale by advertising was impracticable. 
 This remained true, on the whole, even after sale by 
 sample became common. 
 
 Increasing general intelligence, however, has made it 
 possible to picture and describe merchandise so clearly 
 that the prospective purchaser is able to buy what he 
 wants and to know what he will receive without having 
 a sample for examination. Confidence, of course, is 
 the fundamental thing in such buying. For purely self- 
 ish reasons, therefore, to say nothing of the current 
 code of business ethics, the merchant can usually be 
 depended upon to supply the goods described. Under 
 these circumstances, advertising becomes in many 
 businesses the most economical agency of demand crea- 
 tion. Even where the actual sale is made by a salesman 
 from a sample, advertising is used beforehand as an 
 agency to stimulate the desire which the salesman by 
 his selling talk and demonstration turns into expressed 
 demand. Sale on approval makes it possible also to 
 create effective demand through advertising even when 
 the prospect insists on seeing the goods before conclud- 
 ing his purchase. 
 
 It follows that advertising may be employed either as 
 a substitute for middlemen and salesmen or as an aux- 
 iliary force to aid them in their exercise of the selling 
 function. It tends to displace these other agencies, in 
 whole or in part, whenever it is a less expensive or more 
 direct means of communicating ideas to the consumer. 
 It cannot be dismissed as mere " puffing," because its 
 
 201
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 substantial usefulness in our present scheme of distribu- 
 tion has been demonstrated. 
 
 That there are wastes and abuses in its employment 
 may be frankly admitted. It is a new economic force, 
 as yet only partly understood, which has brought change 
 and readjustment in all our machinery of distribution 
 and is itself undergoing constant modification and ad- 
 justment. It has been used extravagantly in not a few 
 instances to exploit commodities having a wide margin 
 of profit. In others it has been expected to prove a 
 panacea for weaknesses in departments remotely related 
 to selling. And in a great many cases the failure of in- 
 dividual campaigns has been charged up against adver- 
 tising as a marketing force when the blame was due to 
 ignorance or neglect of some important factor or ele- 
 ment on the part of the advertiser. 
 
 Against some advertising campaigns the indictment 
 has been brought that they were simply "weapons of 
 destructive competition," ^ serving no useful purpose. 
 Professor Taussig, observing that "mere effrontery in 
 puffing your wares is an important factor in modern 
 trade," offers this illustration of such action: 
 
 "Among articles equally good, that which is syste- 
 matically paraded is likely to be most readily sold. 
 People are led to buy Smith's wares rather than Jones's. 
 One might suppose that if Smith's wares were equally 
 good, and were sold at a lower price (made possible by 
 eliminating the advertising expense) he would hold his 
 own in spite of Jones's preposterous puffing. But in 
 fact, Jones's wares are preferred; some vague impression 
 of superiority is produced by the incessant boasting. 
 
 ^ F. W. Taussig, " Principles of Economics," Vol. II, p. 428. 
 202
 
 ADVERTISING 
 
 Plentiful cash is the sine qua non of an effective adver- 
 tising campaign." 
 
 But from the standpoint of the business man, Jones's 
 advertising, if it be typical of most campaigns, does not 
 accomplish its full and proper work if it simply diverts 
 demand which would otherwise go to Smith's product 
 or be divided with the latter. If the advertising has 
 convincing selling quality, it will, with the great major- 
 ity of products, create demand which did not exist be- 
 fore and will thus widen the market for all makes of 
 the article it describes. Numerous instances could be 
 quoted in support of this statement, covering a wide 
 range of human wants and appetites, from breakfast 
 foods to farming implements. 
 
 I feel sure, too, that something more than "a vague 
 impression of superiority " is produced by any adver- 
 tising campaign worthy of the name. With many 
 low-priced articles, frequently bought and quickly con- 
 sumed, repeated statement of their merits may be 
 enough to influence the first and succeeding purchases. 
 In any case involving greater expenditure, however, the 
 impression built up by the advertising would have to be 
 positive and convincing to overcome the prospects' 
 natural inertia and disinclination to change his buying 
 habits. With utility values equal, too, there is the 
 question of psychic value. Because advertising has 
 given Jones's product a vogue, there is to the purchaser 
 a distinct element of satisfaction in the certainty that 
 it will be recognized as a standard product needing no 
 apology or explanation. 
 
 From the viewpoint of economy alone advertising 
 has made possible the marketing of thousands of com- 
 modities on a national scale, with the consequent sav- 
 
 203
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ings that always come from large-scale production and 
 distribution efficiently carried out and that are certain 
 to be reflected in the long run in lower prices to the 
 consumer. It has set new and higher standards of 
 quality, utility, and value, not only in the things ad- 
 vertised, but in nearly everything that the ordinary 
 man or woman uses or consumes in everyday life. By 
 the same process it has constantly brought these new 
 facilities and comforts to the attention of those who 
 needed them. And it has accomplished all this by the 
 diversion of a far smaller sum of human energy than 
 any other knowai method of demand creation would 
 have required — granting that any other method would 
 have been able at all to perform the same service. 
 
 It is an accepted truth among advertisers that the cre- 
 ation of permanent demand for any commodity or 
 specialty is impossible, no matter how great the ex- 
 penditure, unless it is at least equal in value to any 
 competing product. This value may be partly psychic, 
 in the sense that better design or higher finish or an 
 attractive or dirt-proof package is the basis of the spe- 
 cial appeal to the consumer. But there must also be 
 solid value to satisfy the consumer's need, or the cam- 
 paign exploiting it is bound ultimately to fail. 
 
 The marvelous expansion of the clothing industry in 
 the United States suggests how advertising reacts on 
 the product and effects standardization and increase in 
 values as well as reduction in marketing expense. To 
 begin with, we have the testimony of the largest makers 
 and distributors of trade-marked clothing, who spend 
 in some cases upwards of half a million dollars a year on 
 publicity, that their selling cost per suit is less now than 
 before they began advertising. Because they are able 
 
 204
 
 ADVERTISING 
 
 to manufacture and market suits by the hundred thou- 
 sand, they have been able not only to reduce their 
 advertising outlay per unit sale, but also to effect re- 
 markable economies in buying, manufacturing, selling, 
 and the handling of reserve stocks. 
 
 Differentiation and the development of selling points 
 are fundamentals of successful advertising and hence 
 influence the character of the product sold. If the 
 first analysis of the product does not bring out a con- 
 vincing array of selling points, the pressure on designers 
 and production men to supply them becomes imperative. 
 Merely general claims may at times close orders when 
 backed up by the salesman's skilled presentation and 
 his personality. But cold type and pictures demand 
 the presentation of specific advantages to the buyer. 
 They may take the shape of reduced factory costs, re- 
 flected in lower prices, of added utility, beauty, conven- 
 ience, or durability, of more careful handling, packing, 
 and delivery, of more intelligent adaptation to the in- 
 dividual consumer's needs or increased service. What- 
 ever the line of betterment pursued, almost invariably 
 the consumer profits. Instead of creating a "vague 
 impression of superiority," any successful advertising 
 campaign must be based on the creation of substantial 
 and demonstrable points of superiority to differentiate 
 the advertised product from the mass of competing 
 articles. 
 
 Granting that these added selling features are some- 
 times of little actual and permanent value, there is no 
 question in my mind that the influence of advertising, 
 particularly in recent years, has been increasingly on 
 the side of betterment and heightened utility in the 
 products advertised. Anyone whose memory is long 
 
 205
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 enough to recall the ready-made medium and high- 
 grade clothing of twenty years ago and compare its 
 style, fit, comfort, fabrics, and prices with the same 
 elements in the clothing retailed to-day has a good meas- 
 ure of the social service which advertising has contrib- 
 uted in this and many other fields. 
 
 Consumers and middlemen have been educated to 
 discrimination in their buying; non-advertising manu- 
 facturers have had their attention concentrated on the 
 qualities advertised and on the internal conditions in 
 their own businesses which might be standing between 
 themselves and like advantages. Certainly the quality 
 of non-advertised lines, sold under the dealer's label, 
 has been greatly affected by the general advertising of 
 the great houses and by the demand of consumers for 
 style, fit, and materials of a grade appropriate to the 
 prices asked. 
 
 The remarkable development of the American motor 
 car is another case in print. Without advertising to call 
 the attention of prospective buyers to the improvements 
 and new features of their cars, thus stimulating demand 
 for the new models and creating desire for possession in 
 non-owners, current standards in design, construction, 
 and price would not have been reached for many years 
 to come. European experience demonstrates this and 
 European prices prove again that advertising, the key 
 to sale volume, instead of increasing the cost to the con- 
 sumer, almost always reduces the prices and increases 
 the quality or utility of the article exploited. Without 
 exception, it reacts on manufacture, since the pressure 
 is to produce an article which can be more effectively 
 advertised, whether it be quality or price that is the 
 chosen appeal to the consumer. 
 
 206
 
 ADVERTISING 
 
 When waste does occur in advertising, it may gener- 
 ally be attributed to one of five things: positive lack 
 in the product of those elements of quality or service 
 which appeal to the consuming public's need or desire; 
 ignorance of the true function of advertising as an agent 
 of demand creation for the particular product in hand; 
 blundering application of recognized principles; failure 
 to develop laboratory standards for the testing of sell- 
 ing materials and mediums; or neglect to utilize and 
 keep in operating balance the other essential agencies 
 of distribution. 
 
 It need hardly be said that if the goods advertised are 
 not adapted to satisfy a real want, the advertising can- 
 not produce results ; attempting to sell a thing for which 
 no one has actual or potential use is wasted effort. Even 
 with a desirable product, the medium used for the trans- 
 mission of ideas about it may not be the one reaching 
 an economic or social group in which are many individ- 
 uals having a latent need for the commodity. 
 
 The most serious cause of inefficiency usually lies in 
 the fact that the ideas about the goods or the form in 
 which they are communicated are not adapted to secure 
 the reaction desired. Enough has been said, however, 
 about methods of measuring the value of advertising to 
 indicate the policies whose observance will correct either 
 of these conditions. The final important cause of waste 
 in advertising, neglect to provide for adequate physical 
 distribution and thus realize maximum results from 
 aroused demand, will be discussed in another place. 
 Here, as in all the other activities of distribution, the 
 manager must preserve a balance between the time, 
 area, and volume factors both in advertising and physi- 
 cal distribution, or the leakage of demand will destroy 
 
 207
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 all the gain which should come from the most effective 
 campaign. 
 
 I believe that the social disadvantages attributed to 
 advertising do not in the main exist, but that consid- 
 ered as one agency of selling and utilized in its proper 
 place, advertising is a modern social force of high value. 
 To the producer the advantages possessed by advertis- 
 ing over other agencies in demand creation fall under 
 three heads. 
 
 I. Efficiency. In advertising there is virtually no 
 limit to the number of prospective buyers who can be 
 addressed simultaneously either through one medium 
 or through many. Intensive cultivation of one or more 
 selected districts is equally practicable. In both cases 
 quick action makes immediate results possible, whether 
 the object is sales volume or the defining of the market 
 in its broad lines. By choosing the right mediums and 
 adapting the appeal to the class addressed, every prom- 
 ising social and economic level can be explored and those 
 who will buy your product for consumption or resale 
 can be discovered. 
 
 Besides this selective method of appealing to classes 
 and individuals, a balanced campaign has cumulative 
 force, influencing the public mind as well as individuals. 
 The psychological effect of getting everybody talking 
 about the goods is too well known to require elabora- 
 tion. Another important by-product of advertising is 
 the stimulus exerted on those to whom it is not directly 
 addressed. The surest road to the interest of the mid- 
 dleman, for instance, is a convincing campaign aimed 
 at prospective consumers. 
 
 Where the ideas about the goods are difficult to com- 
 municate because they are new and different, or where 
 
 208
 
 ADVERTISING 
 
 for any other reason *' the trade " cannot be depended 
 upon for their transmission, advertising offers the only 
 available means of rapid and accurate transmission. 
 A further special utility is the stabilizing of the market 
 for a product by inducing the ultimate user to insist on 
 having it and thus limiting the dealer's power to supply 
 a substitute. 
 
 II. Economy. For the same outlay advertising will 
 establish a contact with twenty or a hundred prospec- 
 tive purchasers where a salesman or a middleman can 
 call on only one. Granting that the latter contact is 
 usually more likely to show immediate results, it re- 
 mains true that except where the total of possible pur- 
 chasers is small or where they are concentrated in a 
 small area the work of finding them and acquainting 
 them with the product can be done much more cheaply 
 by advertising. 
 
 In preparing the ground for the salesman's call, in 
 giving variety and interest to the follow-up which con- 
 tinues the education of the consumer, and in supple- 
 menting the salesman's efforts in an intensive campaign, 
 advertising has developed an exclusive sphere of action 
 by securing results at minimum cost. In the mail- 
 order field, of course, advertising has demonstrated its 
 ability to market the most diversified lines both eco- 
 nomically and efficiently. 
 
 III. Controls. The materials of demand creation can 
 be presented in the exact order and in the particular 
 form which tests have proved to be the most effective. 
 The producer's ideas about his goods thus become 
 fixed quantities for the consumer and the middleman. 
 There is no chance of a salesman's failing to transmit 
 them fully and convincingly, no danger of his misrepre- 
 
 S09
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 senting the product and the policies behind it, or of 
 substituting for sound and honest salesmanship the 
 specious pull of personality, with its bad after effects. 
 Advertising writes a sales platform for the house to 
 which not only the selling force but also the production 
 and service departments must conform. 
 
 This permanence of the written or printed word and 
 symbol gives the reader confidence in the claims put 
 forward for the goods, particularly when the medium 
 of transmission is one to which he looks for necessary or 
 valued information. All his training, from school days 
 on, has accustomed him to receive important ideas 
 through the eye. The impression made by advertising, 
 therefore, is clearer and more lasting; any complex 
 statement, put in written or printed form, is more easily 
 analyzed and understood. Psychologists are agreed on 
 this point; and salesmen by the thousand apply the 
 principle every day when they use the advertising is- 
 sued by their concerns to concentrate the attention of 
 customers on some vital point or to reinforce word-of- 
 mouth statements about their products and policies. 
 
 The character of the demand created is an important 
 factor in determining advertising policy. This may be 
 either one of three general kinds: (1) expressed de- 
 mand, (2) unexpressed conscious demand, and (3) sub- 
 conscious demand. 
 
 To illustrate the distinction between them, suppose 
 that a specialized product distributed through retailers 
 is advertised in one or several periodicals of large cir- 
 culation. In response to this publicity, 30,000 persons 
 go to convenient stores and ask for it, 60,000 make a 
 mental note of its name and qualities and decide to buy 
 it when next they need such an article, and 100,000 get 
 
 210
 
 ADVERTISING 
 
 a favorable impression which makes them receptive to 
 further exciting forces, Hke recognition of the product 
 in a store, pkis a clerk's effort to sell it. Here the 
 30,000 who want to buy the article represent the ex- 
 pressed demand, the 60,000 the unexpressed conscious 
 demand, and the 100,000 the subconscious demand. 
 
 This, of course, is the simplest statement of a situa- 
 tion that grows more involved the deeper you penetrate 
 into mail-order selling and other forms of direct market- 
 ing. In the case of high-priced office or factory equip- 
 ment, which ordinarily requires demonstration by the 
 maker's salesman, the briefest request for further in- 
 formation or a salesman's visit would be taken as ex- 
 pressed demand, though the actual sale might be months 
 or years away. 
 
 Speaking broadly, however, expressed demand stands 
 for immediate sales and unexpressed conscious demand 
 for future sales if no unfavorable motive intervenes, 
 while subconscious demand means that the field has 
 been fertilized but that additional selling impulses are 
 needed to produce orders. It is quite true that unex- 
 pressed conscious demand and subconscious demand 
 are hard to measure or appraise. Yet both must be 
 taken into account in determining the advertising poli- 
 cies of a business. To ignore either or to neglect the 
 available means of turning them into positive demand 
 might spell the difference between success and failure. 
 Here again the principles of balance, of interdepend- 
 ence, and of cumulative differentials come in to govern 
 the emphasis laid on each element of the campaign. 
 Intelligent testing of these elements in preliminary 
 trials will supply a basis for coordination and will fur- 
 nish standards for anticipating the values of the three 
 
 211
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 kinds of demand created, in terms of actual and poten- 
 tial sales. 
 
 Breaking his advertising problem up into its con- 
 stituent problems, the manager finds that the first of 
 these has to do with the materials of demand creation. 
 Does the product belong to a class that can be effec- 
 tively advertised.'^ Can it be sufficiently differentiated 
 from others of like use to give it an individuality or a 
 special utility which will commend it to the prospective 
 consumer and induce him to buy it either direct or at 
 his usual source of supply .^^ Is the margin of profit 
 great enough to justify the outlay necessary for pro- 
 motion.? Would increased volume allow such reduction 
 in unit factory and distributing costs as to make up the 
 difference per sale.'' Are potential consumers or possible 
 new uses numerous enough to render this increase in 
 volume probable.'^ 
 
 Here enter the basic elements of price and utility. 
 Take up any problem in any phase of demand creation, 
 indeed, and you will find that it cannot be solved satis- 
 factorily until all the other factors in distribution have 
 been given their rightful weight and value in the general 
 scheme. For instance, a mistaken price policy, if per- 
 sisted in, may neutralize the effect of a brilliant adver- 
 tising and selling campaign. A balance between price 
 and utility must be established in fact and in the minds 
 of possible consumers. And utility may be taken to in- 
 clude both practical and aesthetic returns to the buyer, 
 the degree of adaptation in the product to his needs and 
 tastes, its quality, durability, perfection of design and 
 finish, and the service which is the sum of these con- 
 stituents. Where the middleman takes part in its dis- 
 tribution, its sales utility is also closely related to price. 
 
 212
 
 ADVERTISING 
 
 More of this, however, in a later chapter on price 
 policies. 
 
 Such are some of the questions which should be set- 
 tled affirmatively before an advertising campaign is 
 undertaken, though the necessary adjustments often 
 are made unconsciously. Analysis and accommoda- 
 tion of factors, too, must be carried much further. There 
 is the character of the demand to be created, for example. 
 Is the product one for which only a temporary market 
 can be made, like the fads in women's dress acces- 
 sories which dominate but rarely outlive the season? 
 Or has it solid, enduring qualities which insure perma- 
 nent demand and repeat orders at intervals? 
 
 In the first case the margin of profit must be greater 
 and the appeal of the advertising more urgent. Since 
 immediate sales are the only sales possible, the effort 
 must be to close the maximum number before the de- 
 mand fades or competitors enter to dispute the market. 
 In the second instance, all the profit from the first sale 
 may be absorbed by the cost of making the connection, 
 because the initial purchase is depended upon to influ- 
 ence future purchasers and the object is continuous 
 profits over a long period. 
 
 The remarkably low price put on some office and 
 store appliances is a development of the latter policy. 
 The machine itself is sold at cost or less than the actual 
 delivered cost, the maker reaping his profits from the 
 subsequent trade in supplies. Manufacturers of food 
 and toilet specialties frequently go much further to in- 
 troduce a new product. A full-sized unit is either given 
 away or sold at a special price, though the retailer is 
 paid his full profit on every unit distributed. Such ap- 
 parently extravagant practices are justffied when the 
 
 213
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 margin of profit is great and repeat orders can be looked 
 to for long-time returns. 
 
 The novelty or familiarity of the product to the pro- 
 spective buyer raises another problem in advertising. 
 If it be a better or cheaper substitute for something he 
 is already using, the task is relatively simple. The 
 points of superiority must be demonstrated in a con- 
 vincing way and their effect on the user made plain, the 
 stress on the arguments increasing as the difference in 
 price ranges up from zero. Recent campaigns to exploit 
 men's trade-marked underwear illustrate this. 
 
 When a lower price is the basis of the appeal, it must 
 be shown that the essential utility of the article has not 
 been sacrificed. Apart from the design and finish of 
 the cabinet and other parts, the fifteen-dollar talking 
 machine is practically the same, to any but the culti- 
 vated ear, as that which retails for twice or thrice as 
 much. But the customer remains skeptical until he 
 has been convinced by advertising, perhaps with the 
 aid of a demonstration for which the advertising has 
 caused him to ask. 
 
 The current tendency in marketing goods through 
 advertising is to emphasize not price, but the differen- 
 tiation from staple types and the closer adaptation to 
 the user's needs. It is only in certain progressively com- 
 petitive fields, where a national market and a tremen- 
 dous number of possible buyers hold out opportunities 
 of economy through large-scale operations, that price re- 
 duction is made one of the important talking points. The 
 exploiting of safety razors, vacuum cleaners, and dollar 
 watches is an example of this policy. The motor car 
 supplies another striking instance. But here at least 
 the effectiveness of the advertising is increased by the 
 
 214
 
 ADVERTISING 
 
 combination of the two basic appeals already touched 
 on — extraordinary improvements in the safety, econ- 
 omy, comfort, convenience, beauty, and reliability of 
 the cars themselves (in a word, their utility) and equally 
 remarkable reductions in selling prices. And to these 
 factors should be added the fundamental conscious 
 need and subconscious demand for the transportation 
 eflSciency offered, both constantly stimulated by a flood 
 of advertising competitive in purpose but actually 
 cooperative in its heightening of the prospect's desire 
 to possess an automobile. 
 
 Except for bettered staples, however, few specialized 
 products encounter a developed need and a waiting 
 market. The history of modern business is a record of 
 imagination and intelligence applied to the searching 
 out of valid but unrecognized needs and the invention 
 of new foods or furniture, apparel or machines, to sat- 
 isfy them. The constructive specialty manufacturer is 
 a pioneer always a day's march ahead of the general 
 public and usually under the necessity of educating the 
 public to perceive a particular need it has lying latent 
 and to apply his product to that need. 
 
 For many years this education was carried on through 
 salesmen. Within the last two decades, however, adver- 
 tising has been developed into a more effective agency 
 for the transmission of ideas about the goods both to 
 consumers and to the intervening middlemen. Some 
 may prefer to date this development from the second 
 half of the seventeenth century, when manufacturers 
 and retail shopkeepers solicited custom through ele- 
 gant printed dodgers and the weekly newspaper offered 
 a medium for exploiting books, patent medicines, and 
 merchandise of various sorts ; not a few might insist that 
 
 215
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 intelligent use of advertising is a thing of very recent 
 years. Whichever opinion the reader holds, there is no 
 ignoring the contrast between the long process of intro- 
 ducing the pioneer models of typewriters, adding ma- 
 chines, and other modern tools of business through 
 salesmen and the rapid expansion of these industries 
 in recent years when advertising has been developed to 
 the point where it can take over its proper share of the 
 work of demand creation. 
 
 What constitutes that share in the case of his own 
 product is one of the manager's primary problems. He 
 has certain ideas about his product, which must be 
 communicated to consumers or potential consumers in 
 order to stimulate buying. If he has a going business 
 with an established scheme of distribution, his effort 
 will be to discover where and how he can increase the 
 efficiency of his selling efforts through advertising or 
 where and how he can substitute advertising of one kind 
 or another for a more costly form of demand creation 
 currently employed. 
 
 Fundamentally he has the same need for a general 
 knowledge of the advertising machinery and mediums 
 at his disposal as the man just launching a new trading 
 venture. Otherwise his choice and use of mediums are 
 likely to be biased by prejudice or insufficient informa- 
 tion. Such a survey, except in outline, is beyond the 
 purpose of this volume, which attempts to deal with the 
 problems of business from the owner's or manager's 
 viewpoint and assumes that he has the aid of depart- 
 ment executives in guiding the activities of his business. 
 In this case, a competent advertising manager or an 
 outside service agency would supply in a qualified meas- 
 ure the sort of counsel which a superintendent or works 
 
 216
 
 ADVERTISING 
 
 engineer would bring to the settlement of a question of 
 factory construction or equipment. 
 
 The distinction between general and direct adver- 
 tising, as we have already seen, rests on the scope of the 
 appeal made. General advertising is addressed to the 
 public at large or to a considerable section of it. Direct 
 advertising is aimed at a specific individual or a group 
 of individuals who have been sifted out of the mass by 
 one process or another and classified as prospective 
 buyers. It may be said to occupy halfway ground be- 
 tween the impersonal appeal of general publicity and 
 the individual contact of the salesman. 
 
 It is in manner of approach rather than purpose or 
 function that general and direct advertising diverge. 
 A perfectly balanced campaign might include the use 
 of every class of general and direct mediums, with a 
 distinct function allotted to each class and a different 
 appeal framed for each medium. And another equally 
 intelligent campaign might concentrate on the use of a 
 single class of either direct or general mediums. Be- 
 tween these two extremes, any one of a great number of 
 combinations of mediums and appeals might be the one 
 to prove the most effective. 
 
 The practical plan, indeed, depends on so many 
 factors that only a few can be suggested here. The 
 character and price of the product, for example : is it a 
 necessity, a utility, or a relative luxury .^^ An article of 
 business, personal, or household use or one of pleasure? 
 A thing of daily consumption or of long service? Is its 
 price large or small as compared with the "consumer's 
 surplus "? Can its purchase and use be extended by 
 judicious reduction of its cost and quality? Is it trade- 
 marked or otherwise differentiated so that the demand 
 
 217
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 created for it cannot be diverted to similar competing 
 products? And so on through a long list of things to 
 be considered, including the possibility of doing all the 
 work of demand creation through advertising or the 
 necessity of employing middlemen, the agency adopted 
 for physical supply, the existence of mediums reaching 
 the classes who are prospective purchasers, the degree 
 of intelligence and the reading and buying habits of 
 these prospects. 
 
 No small amount of the waste in money and effort 
 which has attended advertising in the past was due to 
 the failure of managers to analyze the problem, to break 
 it up into its constituent problems, to list the factors of 
 importance in solving each of these constituent prob- 
 lems, to put aside personal preferences or prejudices in 
 valuing these factors, to refuse to accept precedents as 
 rules, to insist on laboratory standards in both mate- 
 rials and mediums, and to realize that advertising is 
 a distinct and separate activity needing a fresh view- 
 point and a new angle of approach to be grasped and 
 effectively performed. 
 
 218
 
 CHAPTER XIV 
 
 ORGANIZATION OF DEMAND CREATION 
 ANALYSIS OF THE MARKET 
 
 WHEN the manager of a business has assembled his 
 materials of demand creation and has considered 
 the agencies at his disposal for conveying them to pos- 
 sible consumers, he faces the further task of putting 
 the work of idea-transmission on a sound and efficient 
 basis. Breaking his problem up into its constituent 
 problems, he finds these are three in number: first, to 
 discover how many persons want his goods or can be 
 induced to buy and use them, who these persons are, 
 where they are located, and how often they are likely 
 to come into the market; second, to learn how much 
 these prospective consumers are willing to pay for the 
 goods; and third, to determine what agency or group 
 of agencies will be the most effective in creating an 
 adequate demand. 
 
 His problem, in a word, is one of organization; his 
 three sub-problems are analysis of the market, deter- 
 mination of price, and combination and coordination of 
 the agencies to be employed in stimulating demand in 
 the chosen market at the predetermined price. No one 
 of the three problems takes particular precedence over 
 the others, for like all the factors and activities of busi- 
 ness, they are interdependent. The extent of the mar- 
 ket is determined largely by the price, and the price 
 cannot be fixed without considering its effect on the 
 broadening or limiting of demand. Again the size and 
 
 219
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 character of the market and the price of the commodity 
 (here also the margin available for distribution expense 
 and profit) will go a long way toward deciding what 
 agencies can be used for demand creation; the agencies 
 chosen will likewise have an important bearing on both 
 price and the width of the market. 
 
 Nor are these policies strictly departmental in their 
 scope or application. They must conform to the general 
 policies of the management; they cannot be formulated 
 without considering, not alone their effect on one an- 
 other, but also their influence on all the other activities 
 of the business, on production, distribution, and ad- 
 ministration. The principle of balance must be ad- 
 hered to, the organization of the activities of demand 
 creation must be governed by the same ratio of cost, 
 quality, and service which is observed in production 
 and administration, as well as in that other function of 
 distribution still to be discussed, physical supply. 
 
 The market for a commodity is made up of the two 
 classes of customers and prospects, the latter com- 
 prising all those who have an unexpressed conscious 
 desire or subconscious need for the goods in question. 
 Demand creation must have for its objective either the 
 development of unrecognized wants or new uses for the 
 product in the minds of customers already on the books 
 or the discovery and transformation of prospects into 
 new buyers. These prospects may be in territory al- 
 ready covered by the selling plan or in territories further 
 afield. They may be buying substitutes or competing 
 goods, or they may not be buying anything resembling 
 the product because an unconscious need or subcon- 
 scious desire has never been developed into positive 
 demand. 
 
 220
 
 ANALYSIS OF THE MARKET 
 
 In the first approach to the problem, individuals do 
 not count except as representatives of the groups to 
 which they belong. It is with the group and its reaction 
 to his product that the manager must deal, since his 
 initial concern is to estimate possible sales volume, the 
 number of product units for which he can reasonably 
 expect to find buyers when his machinery of demand 
 creation has gathered headway. 
 
 This first survey is vital. It must uncover a sufficient 
 number of prospective purchasers in the chosen field 
 to insure a fixed minimum of sales; otherwise the busi- 
 ness remains merely a project or settles down for "a 
 long pull " while demand grows up to it. The domestic 
 market for aeroplanes early in 1914, for instance, was 
 so restricted and there was such competition for the 
 attention of the few rich sportsmen who were possible 
 buyers that a new aeroplane business would have been 
 a doubtful undertaking. Since the outbreak of the Eu- 
 ropean war, however, the military demand has brought 
 about such improvement in design and production 
 methods and the safety of flying has been so thoroughly 
 demonstrated that much larger American sales of ma- 
 chines for sport, military, and business purposes may 
 be anticipated when the war is over. This may be 
 the more confidently expected since increased factory 
 facilities, reduced production costs, and the necessity 
 of keeping enlarged establishments busy are sure to be 
 reflected in lower prices — another evidence that the 
 principle of interdependence never ceases to apply in 
 all the activities of business. 
 
 In analyzing his market, the business man faces an 
 indefinite body of possible purchasers, widely distributed 
 geographically and exhibiting various extremes of pur- 
 
 221
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 chasing power, intelligence, and conscious and unrecog- 
 nized needs. The effective demand of the individual 
 depends not alone on his ability to pay for the product 
 offered, but also upon his wants and tastes as cultivated 
 or repressed by his character, education, habits, occu- 
 pation, and economic, religious, and social environ- 
 ments. The market, therefore, splits up into economic 
 and social strata as well as into geographic divisions. 
 
 Turning first to the territorial distribution of the con- 
 suming public, he finds any number of factors influenc- 
 ing the probable demand for his goods. If his product 
 be designed for a special and limited use, like an im- 
 proved breaking plow or cream separator, his market is 
 narrowed immediately by the exclusion of the millions 
 who live in cities and in industrial districts. Though 
 even here it will be well for him to consider the thou- 
 sands of business men and investors who own farms and 
 are interested in the latest labor-saving devices for farm 
 w^ork. Means exist for estimating pretty closely the 
 extent of this marginal market and the mediums of 
 demand creation for reaching it. 
 
 But towTi and country are only the first broad geo- 
 graphic divisions of the market. Continuing with agri- 
 cultural implements, for the sake of simplicity, the New 
 England farmer as a prospect for improved machinery 
 is in a different class from the man in the "corn belt " 
 of Illinois or Iowa, just as the southern planter has very 
 little in common with the dairyman of Ohio or Wiscon- 
 sin. It follows that the farmers of the United States 
 cannot be successfully approached as a single group, 
 but must be classified as a number of sectional groups, 
 the members of each group having a general likeness in 
 their responsiveness to new^ ideas, in resources, and in 
 
 11%
 
 ANALYSIS OF THE MARKET 
 
 standards of working and living, but breaking up into 
 many sub-groups according to their dominant crop or 
 occupational interests. 
 
 All, it is true, have similar fundamental needs for 
 wagons, harness, plows, axes, pumps, and so on. Once 
 away from these necessities, however, great diversity 
 is encountered in felt needs and the mental attitude 
 toward unfamiliar things. Even in buying pumps, 
 where the New England hill farmer would be content 
 with a low-priced wooden or iron pump, the Indiana 
 man would be likely to demand a better grade and then 
 hitch a windmill to it, while the Illinois or Iowa stock 
 raiser would add a gasoline engine to insure him a de- 
 pendapable water supply at all times. 
 
 Climate and soil conditions are responsible for further 
 sub-groupings, while the racial origins of the people and 
 the school conditions are further factors in limiting the 
 demands of local sub-groups. The surface character of 
 the land under cultivation and the size of farms must 
 also be considered in determining the market for the 
 larger tools of agriculture like oil tractors, self-binders, 
 gang plows, and motor trucks. 
 
 Equally important in such an analj^sis is a realiza- 
 tion of what may be termed the market contour; the 
 market is never a level plain. It is composed of dif- 
 fering economic and social strata, though the distinc- 
 tion between the various levels is not always apparent 
 at first glance. Noting the average size of farms and 
 the land values in the prosperous sections of the middle 
 west, for example, the manufacturer of a moderate- 
 priced heating system for country houses might con- 
 clude that here were many prospects for the thing he had 
 to sell. By carrying his analysis a little further, how- 
 
 223
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 ever, easily accessible statistics would show him that a 
 large proportion of the better farms are in the hands of 
 tenants, who might be persuaded to put their surpluses 
 into motor cars or talking machines, but hardly into 
 permanent heating systems. The non-resident owners, 
 of course, would be less apt to make an investment 
 seemingly so unproductive. 
 
 Even in the case of farm necessities, like plows or 
 harvesters, tenant farming would be an important fac- 
 tor in the analysis, since many tenants are obliged to 
 operate on credit, and the credit terms usual in the ter- 
 ritory might make too severe a drain on the producer's 
 financial resources. At the least, they would influence 
 prices and would present on the whole a more serious 
 problem than the credit situation in a more recently 
 settled district where customers, though lacking ready 
 money, have the landowner's advantage in making 
 loans. 
 
 This element of market contour takes on increasing 
 significance when the product is designed to appeal to 
 the general public, without regard for geographic or oc- 
 cupational lines. The distributor of a three-dollar trade- 
 marked hat for men must obviously direct his appeal to 
 different economic and social strata, must consider dif- 
 ferent buying motives, and must adopt different selling 
 policies, as compared with the distributors of two or 
 five-dollar hats. He has this advantage over the latter, 
 that his prospects include dwellers in every village of the 
 country large enough to boast a general store and that 
 a city like New York or Chicago offers him twenty or 
 perhaps fifty neighborhood centers of distribution as 
 against two or three open to the five-dollar hat. But 
 he must keep in mind the fact that he will encounter 
 
 224
 
 ANALYSIS OF THE MARKET 
 
 forceful competition from established hats of the same 
 or lower prices and that absolutely local conditions will 
 frequently determine the volume of his local sales. 
 
 Statistics of the population will afford him only initial 
 help, therefore, in arriving at the possible demand for 
 his products in any given district or community. He 
 will have to investigate a number of average neighbor- 
 hoods and communities, the retailers supplying them, 
 the competing goods bought by them, and the methods 
 used in exploiting those goods. Half a dozen surveys 
 of typical outlying centers of trade would give a fair 
 average of conditions, perhaps, for the whole of Chi- 
 cago. The fact that New York has been the scene of a 
 greater number of like investigations and "try-out" 
 campaigns would have to be taken into consideration 
 in deciding whether the results of such a "sampling" 
 there would be less dependable. Worcester, Massa- 
 chusetts, or Dayton, Ohio, should supply data on which 
 to base an estimate of what industrial communities, 
 east and west, would offer in the way of sales. 
 
 And so, up and down the population scale, tjq^ical 
 cities could be tested at moderate expense to determine 
 how other towns of their size and class would receive 
 a new trade-marked hat and what selling efforts must 
 be put forth to secure the attention of dealers and ulti- 
 mate users. I need hardly add that an analysis of the 
 market on a national scale would be useless expenditure 
 except for a business contemplating national distribu- 
 tion, or that an analysis of the national market for a 
 general utility must recognize certain broad divisions 
 of the country with distinctive trade and social usages 
 certain to affect demand as well as the processes of de- 
 mand creation. For the average smaller business the 
 
 225
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 only safe and economical way would be to confine the 
 analysis to its strategic territory and concentrate selling 
 effort on this region, expanding as the manufacturing 
 and sales organization proved themselves capable of 
 taking on more work. 
 
 With a product appealing to only a single element of 
 the population, the analysis becomes more complex 
 and the geographical and economic factors are more 
 difl5cult to align. Take the case of a publisher who is 
 mapping out a selling campaign for a Catholic magazine 
 or encyclopedia. It is essential that he take into ac- 
 count not only the geographic distribution of the Cath- 
 olic population in the United States, the regions where 
 it is relatively dense, and the regions where it constitutes 
 only a small part in the population, but also the con- 
 stitution of that population through the economic 
 strata. 
 
 A method of distribution successful in New Orleans, 
 where the denser Catholic population includes those 
 of all degrees of purchasing power, might well fail if 
 applied in Maine, where the Catholic population is rela- 
 tively sparse and is composed largely of French-Cana- 
 dian mill hands. The irregularity of distribution of the 
 Catholic population, however, would be compensated 
 by the definite information available about its location 
 and its average buying power, the latter influenced in 
 no small degree by its sympathy and responsivecfess to 
 church appeals. 
 
 Density of population is an element which rarely can 
 be neglected, either in deciding whether a profitable 
 market exists for the goods or in determining how a 
 possible market can be developed and supplied. It is 
 here that analysis of the market, price policies, and com- 
 
 226
 
 ANALYSIS OF THE IVIARKET 
 
 bination of demand-creation agencies are most closely 
 interwoven. Where population is dense the means 
 of creating demand multiply. Intensive cultivation 
 by direct salesmen, for instance, becomes possible. 
 Where people are widely scattered, prospects may be 
 so few that no practicable market exists unless a com- 
 bination of agencies can be worked out which will over- 
 come the handicaps of distance, or unless a price can be 
 secured which will take care of the extra selling cost. 
 
 If the manager "lumps" his costs of demand creation 
 through the salesmen or any group of agencies which he 
 may be using and strikes a balance for the whole mar- 
 ket, he may be "playing safe" and insuring himself a 
 profit. But it is certain that if he ignores or fails to 
 detect the fact that the salesman is an unprofitable agent 
 in one or several sparsely settled territories, he is mak- 
 ing a double sacrifice of profit — that which he must 
 substract from his net returns in densely populated dis- 
 tricts to make up these individual deficits and that 
 which he probably would earn in the unprofitable terri- 
 tories if through analysis and tests he were to discover 
 the right agency or combination of agencies to cope with 
 local conditions. Again, if the manager bases his esti- 
 mate of the average cost of selling for the whole market 
 on his experience in a few densely populated or easily 
 accessible territories, he may easily go wrong. Such 
 a test cannot fairly represent the possibilities of the 
 larger and differently constituted general field. 
 
 The typical business man seldom appreciates the im- 
 portance of market contours in their relation to the 
 distribution of his product. His method is the familiar 
 "trial by error." He sends his salesmen or his direct 
 advertising to dealer-prospects or consumer-prospects 
 
 227
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 until he gets a positive or a decidedly negative result. 
 With a large margin of profit, this method, though 
 costly, frequently proves effective for a time. But for 
 the product in a competitive field, analysis of the market 
 is indispensable. 
 
 The motives to which a product appeals may differ 
 widely in the various social strata of the market. The 
 materials of demand creation, the ideas about the prod- 
 uct which will arouse a desire for its possession in one 
 level or section of a city's population, are not always 
 effective when used to reach another section. 
 
 Low price and wearing qualities rather than fineness 
 of finish or exclusiveness of design will appeal to con- 
 sumers having small incomes, and thus naturally to the 
 retailers who serve them. These arguments would not 
 have the same force on consumers in the higher economic 
 levels to whom quality of materials and finish and 
 beauty or convenience in design would probably be of 
 prime importance. Many articles of daily use, it may 
 be added, achieve the marketing ideal of general appeal 
 to all classes, either because their enduring quality or 
 essential utility overshadows the price objection or be- 
 cause sentiment or social emulation becomes a factor 
 in the purchase. The presence in our cities and indus- 
 trial towns of large populations of foreign birth, with 
 traditions, tastes, buying habits, and standards of value 
 all their own, and with varying degrees of intelligence 
 and acquaintance with the English language, further 
 complicates this matter of market location and puts a 
 premium on painstaking analysis. 
 
 Innumerable businesses recognize the existence of 
 market contours by putting out their product in two 
 or more grades, retaining as much of the essential util- 
 
 228
 
 ANALYSIS OF THE MARKET 
 
 ity or style as is compatible with the reductions in price 
 which will bring the various grades within reach of 
 prospects on various economic planes. This is accepted 
 practice among manufacturers of clothing for both men 
 and women, watches, talking machines, cameras, hand 
 tools, and a long list of other products having general 
 appeal. 
 
 It is likewise a basic policy in the more advanced 
 retail stores. Some of these not only maintain base- 
 ment departments frankly offering medium and low- 
 priced substitutes for the more substantial or costly 
 lines displayed on the main selling floors, but also break 
 up their important departments, such as women's cloth- 
 ing, shoes, millinery, and furniture, into sections divided 
 on lines of price and thus advertised. The woman who 
 is able to pay only eighteen dollars for a suit is recog- 
 nized as a class quite as well worth catering to, because 
 of the number of individuals involved and therefore 
 the sales volume and rate of turnover that are possible, 
 as the women whose standards of dress make thirty- 
 five, fifty, or one hundred dollars the minimum outlay. 
 
 Nor is it forgotten that despite the emphasis put on 
 price, the real standard is one of values and that buyers 
 will cross their customary' lines, paying more than usual 
 or making extra purchases when the inducement is 
 sufficient. Accordingly, great technical skill and finan- 
 cial resources are often brought to bear on these base- 
 ment stores and department sections in providing 
 maximum values within fixed price ranges. Thej' are 
 comparatively recent innovations in retailing, trace- 
 able, I think, to the realization by store managers that 
 the local market is made up of several economic levels, 
 and that the problem of sales volume and stability is 
 
 229
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 one of general and particular appeals to all these 
 classes. 
 
 There are stores, of course, which concentrate on one 
 or two of the classes at the top of the scale just as there 
 are manufacturers of exclusive pianos and motor cars, 
 jewelry and silverware, who ignore all but the few who 
 can put quality and individuality of product above 
 price. In fact, the specialty shop and the producer of 
 exclusive merchandise, both of them outgrowths of more 
 intensive analysis of the market, have so multiplied in 
 recent years that they have kept pace with what might 
 be called the democratization for profit of the large 
 retail store. 
 
 One instance will illustrate the transformation which 
 an intelligent market analysis can effect in an established 
 business. Taking over the conduct of an important St. 
 Louis store, a few years ago, a new manager was amazed 
 at a forty per cent slump in sales during the summer 
 vacation. Railroad statistics indicated that about 
 55,000 persons went away for a fortnight or longer dur- 
 ing July and August. That the absence of seven per cent 
 of the population for a part of the time should cause a 
 forty per cent reduction in sales suggested to the man- 
 ager that the merchandising plan of the store was badly 
 out of balance. 
 
 Checking July deliveries on a route map of the city 
 visualized this lack of balance. Sixty per cent of the 
 deliveries had been made on seven routes in the better 
 residence districts. On five of the seven the July de- 
 liveries numbered only half those made in June; on 
 two others the decrease was much smaller. On four 
 other routes which covered two thirds of the total area 
 of the city, the falling off was negligible; but on the 
 
 230
 
 ANALYSIS OF THE IVIARKET 
 
 basis of the number of parcels handled, these four dis- 
 tricts accounted for only thirty-two per cent of the 
 July business. 
 
 The outstanding fact from this analysis was that the 
 store was neglecting the great market stratum made up 
 of wage earners and thrifty salaried folk, who did not 
 take long vacations but did buy dependable medium- 
 priced merchandise all the year round. The remedy 
 was obvious; it was to stock the lines and grades that 
 would appeal to these neglected prospects and to let 
 them know about the goods by means of an intensive 
 campaign of direct and general advertising. The prob- 
 lem as to what kinds and classes of merchandise should 
 be stocked was a further serious exercise in market 
 analysis. The correctness of the first broad analysis 
 was proved by steadily mounting store sales and by a 
 vacation business the following year that wiped out 
 the former deficits. 
 
 For a new product in an occupied field, analysis of the 
 market involves much more than the classification of 
 economic strata in your selling field, the testing of their 
 reactions to your materials of demand creation, and a 
 survey of the available agencies of demand creation 
 with their relation to your production and sales plan. 
 The following list of factors is suggested by MacMartin 
 as an essential preliminary to a study of national dis- 
 tribution. 
 
 (a) The present annual consumption of similar products or 
 of those for which your product would prove a substitute. 
 
 (6) The per capita consumption among adult men and 
 women or of the particular classes, by age, to which your 
 product will appeal. 
 
 (c) The territorial variations in the consumption, 
 
 231
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 (d) The classes of dealers now handling a similar article. 
 
 (e) The classes of trade sold by these dealers. 
 
 (/) The sales and advertising policies and plans of com- 
 petitors. 
 
 (g) The volume of business secured at the present time 
 by various competitors, with regard especially to the kind, 
 quality, finish, and price of the product each is marketing. 
 
 (h) The percentage of repeat sales competitors seem to 
 be able to secure. 
 
 Consideration of all these factors certainly should sup- 
 plement analysis of the market from the viewpoint of its 
 economic and social levels. How far it would be pos- 
 sible to obtain dependable information covering all the 
 points listed would depend on the nature of the busi- 
 ness and the character of the products. Statistics on 
 consumption are, in general, far from accurate; except 
 in highly competitive fields, a special investigation, by 
 the sampling method at least, would be necessary to 
 secure the specific information on which sound policies 
 are based. 
 
 It might be well to carry the inquiry beyond mere 
 market factors. Business is a competition of individuals 
 as well as of products. Your analysis of the market, 
 then, is not complete unless you know all that can be 
 learned about the major influences that affect that 
 market. What, then, of the organizations and the men 
 behind the products your goods must meet.^^ What are 
 their characters.'^ Their attitude toward business? 
 Their experiences^ The capacities of their organizations ? 
 Their capital and resources.'^ 
 
 It is of importance to know whether your competitors 
 are dreamers and fighters, willing to risk all they have for 
 the sake of the business and their vision of its future, or 
 whether they are conservatives insisting on their annual 
 
 23^
 
 ANALYSIS OF THE MARKET 
 
 profits and grudging expenditures that are not imme- 
 diately productive. You ask whether they are new- 
 comers in the field or whether they have the background 
 of emergency knowledge acquired only by those who 
 have grown up in a trade. Are their making and selling 
 forces competent and efficient and the routine of their 
 production and distribution activities well established.'' 
 Is their selling plan positive yet flexible enough to adjust 
 itself to unusual situations.'* And finally, have they re- 
 sources more or less ample than your own.'' 
 
 Abundant capital will enable them to perfect their 
 products, to secure good men for all important positions, 
 to install the best equipment and standardize all their 
 factory and sales operations, to finance, build up, and 
 train a capable sales force, to test their materials of 
 demand creation, to buy materials in the lowest cash 
 market, to make easier terms to dealers and consumers, 
 to offer free trials of the product, to carry on extensive 
 advertising campaigns to exploit their goods, their 
 terms, and their service. Capital alone will not do all 
 these things; it only renders them possible. But the 
 possibilities as well as the actualities of competition are 
 elements which the manager of a new undertaking or of 
 an old business expanding must consider in the analysis 
 of his market. That market, when all prior claims have 
 been subtracted, may prove to be no market at all. 
 
 233
 
 CHAPTER XV 
 
 ORGANIZATION OF DEMAND CREATION 
 PRICE POLICIES 
 
 SINCE gain is the immediate aim of business, it seems 
 obvious that the price which a manufacturer obtains 
 for his product must be based on factory cost, with 
 sales and deHvery expense and an allowance for profit 
 added. Taking a line of trade or all articles of the same 
 class as a whole, this is true; income must exceed outgo 
 or a business soon exhausts its resources. But the in- 
 dividual manager's price problem seldom takes rise in 
 the factory, nor is it solved by simple addition of the 
 elements of delivered cost. 
 
 Usually the process is reversed. The established mar- 
 ket price for the article or for substitutes having the 
 same general function or utility is the starting point as 
 well as the imperative factor in the solution of the 
 problem. In fixing this market price, over short periods 
 or single selling seasons, even a basic element like the 
 cost of raw materials is a consideration secondary to the 
 economic relations of supply and demand for the fin- 
 ished goods. 1 
 
 ^ A study of the relations between quotations for raw cotton and the 
 market price of three well-known lines of bleached muslins, Lonsdales, 
 Hopes, and Lonsdale cambrics, from 1879 to 1914, discloses an interesting 
 example of an economic principle at work. Despite the fact that standardi- 
 zation of products and processes had been carried further, perhaps, in cot- 
 ton textiles than in any other important industry, a comparison of the cost 
 of raw cotton and the prices for finished goods show frequent and sharp 
 divergences where, because of manufactiu-ing conditions, close parallels 
 might be expected. Again and again advances attended reductions in costs 
 
 £34
 
 PRICE POLICIES 
 
 There are practical grounds for this approach, though 
 the rule of thumb enters too much into price-fixing, 
 and too often producers launching a new commodity go 
 on the assumption that the prevailing market price for 
 any given grade is right. They trust to their own re- 
 sourcefulness or the ingenuity of their works executives 
 to turn out the same thing at a margin of profit. Ex- 
 perience has taught them, too, that the first factory 
 output of a new article is usually at a cost higher than 
 the level easily maintained when the operations have 
 become familiar, when lost motion has been eliminated, 
 and continuous production has cut down incidental 
 wastes. 
 
 The decisive reason for considering the market quo- 
 tations for competing goods as a basis in fixing prices is 
 the necessity of making sales. But unless in the long 
 run the price is such that the product can be moved 
 from the warehouse floor and placed in the hands of the 
 consumer at a profit, the whole process of its production, 
 
 of materials, and reductions were made while the market for raw cotton 
 was rising. 
 
 Commenting on the figures, which it secured and pubhshed, " The Journal 
 of Commerce " had this to say about how prices are controlled by the law 
 of supply and demand: 
 
 " Many traders have come to believe that trading values are fixed by the 
 value of the materials and production cost represented in them. Time and 
 again this belief has been shown to be an economic fallacy. The shrewd 
 trader in merchandise is the one who gauges his selling price by the possi- 
 ble demand for what he has to sell, and the supply of such merchandise is 
 the factor that measures the success of distribution. 
 
 " These figures show that time and again cloth values have risen and been 
 maintained without regard to the cost of the raw material values represented 
 in the merchandise. The manufacturer naturally looks upon values from his 
 own viewpoint, which is the one of cost of production. But the merchant 
 must consider other factors. The potential supply in the markets, the power 
 of the buyer to purchase and distribute, and the inherent worth of the goods 
 in competition are powerful factors, and they are the ones that must govern 
 the judgment of the trader." 
 
 235
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 distribution, and administration is a useless thing. The 
 price, therefore, must be adjusted to the conditions that 
 govern in both factory and market. It need not be the 
 highest that can be obtained from a Hmited group of 
 buyers nor the lowest that will yield a fair gain. The 
 practical ideal is the price which will satisfy the present 
 needs of the business yet conserve all the opportunities 
 of profit which the future holds. 
 
 Take the manufacturer of an article competing with 
 commodities substantially identical in kind or func- 
 tion; how shall he determine what the price shall be.'' 
 Analyzing his problem, he finds first of all that he 
 has choice of three general price policies: (1) selling at 
 the market minus, (2) selling at the market par, and (3) 
 selling at the market plus. The economist would find 
 slightly different terms for these, describing them in 
 turn as selling below the normal price, selHng at the 
 normal price, and selling above the normal price. For 
 our purpose, however, the question of terminology is not 
 significant. It is enough that there are these three dis- 
 tinct policies available, and that the business man may 
 adopt any one of these to the exclusion of the others or 
 may use two or more of them in combination to effect 
 his purpose. 
 
 Selling at the market minus aims to increase the vol- 
 ume of sales by reducing the price. It is founded on the 
 trading axiom that small profits on individual transac- 
 tion mean a large turnover. It is the business man's 
 recognition of the truth expressed in the generalization 
 of economic theory that lower prices will increase the 
 amount demanded in the market. Theoretically, at 
 least, the man who sells his product at a price below the 
 figure established by competitors attracts the customers 
 
 236
 
 PRICE POLICIES 
 
 of other distributors. His lower price also means that 
 they can buy more units of his commodity with the same 
 total expenditure, thus insuring him increased volume 
 and corresponding increase in his units of gain. 
 
 But other buyers besides those already in the market 
 are now able to enter. Those whose unexpressed de- 
 mand was not effective before, because they placed a 
 higher value on the money in their possession or upon 
 other commodities which could be purchased with it, 
 now come into the market. The result is that a much 
 larger amount of the commodity is distributed. This 
 is true, in theory, at least, though in practice lower 
 prices do not always over short periods attract cus- 
 tomers from their habitual trading places. Estab- 
 lished credit arrangements, congenial personal relations, 
 all the forces we include under "habit" and the factors 
 classified under "good will" work in opposition to the 
 purely economic motive. It is only in the long run that 
 these deterrent influences are overcome and competi- 
 tion works out its uniform result. 
 
 Selling at the market minus does not ordinarily in- 
 volve a differentiation of the product from the stock 
 product of like nature, though there are many excep- 
 tions to the rule. Nor are trade-marks, brands, or 
 trade names always used. The producer depends upon 
 the lower price to attract purchasers in numbers and 
 looks to a larger turnover to give a reduced propor- 
 tion of overhead expense per unit sold. He also de- 
 pends upon lower factory costs, made possible by the 
 economies of larger scale production, to add further 
 savings. These two forces together extend his area of 
 profit, and he operates at increasing returns. Yet from 
 the point of view of distributing policy the difference in 
 
 237
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 price forms the basis of the appeal to the consumer. 
 And the successful use of the policy in a competitive 
 market, therefore, depends upon continual ability to un- 
 dersell distributors of substantially identical products. 
 
 This policy finds illustration in the sales plans of a 
 great many department stores. It is the basis of bar- 
 gain-counter selling; and in the bargain department 
 stores, selling at the market minus is the dominant 
 policy. Their businesses are organized and operated 
 with an eye single to selling under the market. They 
 ransack the country in their search for bankrupt stocks 
 and the excess stocks of sound wholesale and retail con- 
 cerns which have overbought in certain lines or under- 
 sold because of local weather or industrial conditions. 
 They advertise these bankrupt stocks, mill ends, and 
 wholesale clearances as the basis of their ability to offer 
 standard goods at cut prices. Many of them operate 
 on a cash basis and are thus able to take advantage of 
 the opportunity for close, buying offered by the tem- 
 porary difficulties of other merchants. 
 
 Many others adopt the policy of dealing in "seconds " 
 or in sub-standard goods which resemble standard 
 products but have been cheapened by the dropping of 
 say ten threads to the square inch in a silk or cotton 
 fabric or the slighting of quality or finish in wood or 
 metal product. The enormous trade in sub-standard 
 merchandise may be regarded as an outgrowth of the 
 policy of selling at the market minus, though logically 
 it is an effort to differentiate commodities from the 
 price viewpoint and adjust them to the needs and re- 
 sources of social strata less exacting than those for 
 whom the originals were created. 
 
 But nearly all department stores at times reduce 
 
 2SS
 
 PRICE POLICIES 
 
 prices upon staple commodities either to attract cus- 
 tomers to a current sale or as a permanent store policy. 
 Here the increased volumes, arising from trade drawn 
 from competitors and from new consumers brought into 
 the market, decrease the proportion of overhead ex- 
 pense each unit must bear and allow purchases in larger 
 quantities. These larger quantities put the manager 
 into a position to force the producer to share with him 
 the economies of large-scale production. He demands 
 and secures the jobbers' prices. Often, indeed, he is 
 able to take over the entire output of certain factories, 
 and in some cases he deems it advantageous to operate 
 factories of his own, a procedure which he has had to 
 share recently with the "chain stores " which are 
 making advances in every part of the country and with 
 many defensive buying associations of small retailers. 
 In the department store, moreover, the customer at- 
 tracted by the offer of a staple commodity at less than 
 the prevailing price is likely also to purchase other com- 
 modities yielding a wider margin of profit. 
 
 This, indeed, is the motive at the back of almost every 
 bargain sale advertised by a store selling standard mer- 
 chandise; it aims to bring the customer into the store 
 and within range of its countless selling influences, 
 even at the sacrifice of all profit on the " leaders " adver- 
 tised. The "leaders" offered by the catalogue houses 
 are in exactly the same category. I am not criticising 
 such bargain advertising, but only recognizing its pur- 
 pose, as every business man must in considering his own 
 price policies. The majority of producers and merchants, 
 indeed, make use of "leaders" to exploit their regular 
 lines, nor should such bargain events be confused with 
 the clearing sales of overstocks of seasonal goods which 
 
 239
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 merchants, wholesale and retail, find advisable at the 
 end of each selling period. These clearing sales are not 
 to be classed with selling under the market, because 
 the psychic value of the merchandise to the consumer 
 has fallen since the season was young. 
 
 To the manager putting out a new product in a com- 
 petitive field, selling at the market minus is the easy 
 method of securing distribution. If middlemen are 
 essential to his plan, they will expect a price concession 
 as an inducement to stock and push the article. The 
 larger the concession the greater will be their interest in 
 diverting demand from established lines giving them less 
 return. It may not be necessary to cut the retail price; 
 so great is the influence of the average retailer on the 
 buying of his customers, outside the large cities, that he 
 will have little trouble in substituting an honestly made 
 product on his own assurance that it is good. 
 
 To this substitution nationally advertised brands 
 offer serious impediment, as do also distinctive trade- 
 marked specialties long in the market and usually asked 
 for by name. Even against advertised articles, the 
 dealer with an established following can make the first 
 sale of the new product by suggesting or demonstrating 
 its chief selling points and urging its purchase. Repeat 
 orders then will depend on the impression the product 
 makes on the user. Should its value seem less than that 
 of established competing goods, the retailer's favor will 
 not keep it moving; and the cost to the consumer of the 
 new commodity will have to be lowered to the point 
 where the balance favors price on the one hand as 
 against well-known quality and service on the other. 
 
 The bulk of non-trade-marked specialties and of goods 
 which carry dealers' brands are distributed below the 
 
 240
 
 PRICE POLICIES 
 
 normal market price. The theory of such distribution 
 is sound enough. The middleman by undertaking all 
 the work of demand creation relieves the producer of 
 this effort and expense and allows him to give all his 
 time to manufacturing operations. The margin between 
 the open market price and the price actually charged the 
 middleman is the latter's pay for the function he per- 
 forms in demand creation. The disadvantage, from the 
 manufacturer's standpoint, is that the origin of his 
 product usually is not impressed on consumers in any 
 way and he thus loses the repeat orders which its excel- 
 lences should bring him. Either the wholesaler or re- 
 tailer is at liberty to switch this demand to a substitute 
 at a moment's notice. 
 
 A great many makers of trade-marked products also 
 deliberately adopt the policy of selling under the mar- 
 ket because of the powerful buying appeal resident in a 
 lower price. A new concern is almost forced to offer 
 some such inducement to purchase. It is the recognized 
 way of entering the market, except when the product is 
 obviously superior to the thing it would displace or 
 when its less distinctive quality or utility is exploited 
 by direct salesmen or consumer advertising. Like the 
 producer himself, the middleman is in business to make 
 all the money he can; and a new product affords oppor- 
 tunity for a better bargain in exchange for the outlet the 
 manufacturer must have. 
 
 A new brand of parlor matches, for instance, may be 
 in every way equal to the matches he has been distribut- 
 ing either for resale or for use; but unless there is an evi- 
 dent waiting market for it, there is no reason why he 
 should invest capital, devote space and time to it, and 
 take the risks involved unless it gives him greater re- 
 
 241
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 turns. If he can buy the new match for ten or fifteen 
 cents a case cheaper than competing brands, he has a 
 motive for handhng it and for trying to create demand 
 for it. 
 
 The same market situation faces the manufacturer 
 of any one of a thousand staples and semi-specialties 
 distributed through trade channels. In a lesser degree 
 it applies to highly differentiated specialties of many 
 sorts, including those marketed through direct sales 
 forces. Given approximately equal quality, utility, 
 and prestige, ability to "shade the price" is often the 
 deciding factor in a sale. Such ability, of course, may 
 be the result of low production costs, of economies in 
 distribution or administration, or of unusually effective 
 organization of all the activities of the business. It il- 
 lustrates again the law of interdependence which threads 
 through all these operations. But unless the delivered 
 cost of the goods, as compared with competing lines, 
 is low enough to permit the necessary reductions, adop- 
 tion of the policy of selling at the market minus is no 
 • more than a deferred invitation to the sheriff to take 
 charge. 
 
 Another important consideration here is the influence 
 of the lowered price in bringing new consumers into the 
 market. Chart III, on the opposite page, illustrates 
 the action of this policy. 
 
 The curve of the line L-M, of course, is hypothetical. 
 The diagram is incomplete also, because it does not indi- 
 cate the important factor that other producers are selling 
 at a higher level and that customers are attracted from 
 them, or that old customers are likely to increase the 
 amount of their purchases as the price decreases. How 
 far this is true depends upon the contour of the curve. 
 
 £42
 
 PRICE POLICIES 
 
 If L-M is nearly horizontal, that is, if the demand is 
 elastic, a small decrease of price will bring a large num- 
 ber of purchasers into the market. On the other hand, 
 as L-M turns toward the vertical, fewer purchasers will 
 
 Cii.uii- III 
 
 This chart attempts to show graphically the operation on the demand side of the maricet 
 of the price policy termed "selling at the market minus." On the ordinate ox is laid off a 
 scale of prices for the commodity. On the abscissa oy are laid off the number of purchasers. 
 The arc LM shows the number of purchasers at a given price, growing fewer as the price in- 
 creases and greater as the price decreases. 
 
 Now if oa represents the prevailing market price for the commodity, and oc the number of 
 purchasers at that price, it is apparent that if the price is reduced from oa to oa', new con- 
 sumers will be brought into the market and the number of purchasers at the price oa' will be 
 oc', a number greater than oc. 
 
 enter the market at any given price. Here the demand 
 is said to be inelastic, as is the case with the necessaries 
 of life. 
 
 A certain amount of salt will be bought, no matter 
 what the price. A high price will check consumption to 
 some extent, and a low price will encourage liberal, even 
 careless use. But once the supply indispensable to the 
 support of life is secured, the demand falls off rapidly. 
 The business man adopting the policy of selling at the 
 market minus, therefore, must determine whether the 
 demand for his product is elastic or inelastic. Is he 
 
 243
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 selling a necessity or a luxury? The success or failure 
 of his campaign may depend on the care with which he 
 makes his analysis. 
 
 Selling at the market par was the distribution policy 
 characteristic of the period during which the stress was 
 on production. It is still a common policy in the mar- 
 keting of staple goods. Briefly this policy consists in 
 the acceptance of the market price existing for the com- 
 modity. The producer does not seek to attract pur- 
 chasers by maintaining a price somewhat lower than 
 that at which his competitors sell, nor does he attempt 
 to establish his product, as a distinct commodity, upon 
 a new and higher price basis. He recognizes the market 
 price as something beyond his control and he sells his 
 commodity at this established level. 
 
 With this price policy, the merchant-producer has two 
 general methods of increasing his area of profit. He 
 may devote himself to reducing his factory expense by 
 a better organization of his plant, or he may seek to in- 
 crease his sales, in order to secure the economies of large- 
 scale production and a reduced proportion of overhead 
 expense on each unit of product turned out. 
 
 A field in which the policy of selling at the market is 
 general is the steel business in normal times. The small 
 independent manufacturer accepts the market price of 
 a given steel product as a fixed condition, sells his 
 "share" of the total output, and depends upon holding 
 down his plant costs to maintain his profits. 
 
 In general, if the merchant producer adopts the policy 
 of selling at the market, he must differentiate his prod- 
 uct from that of his competitors and then build up a 
 particular demand for it. To do this he must employ 
 the same means he would use to establish the product as 
 
 244
 
 PRICE POLICIES 
 
 a distinct commodity upon a higher price level. Trade- 
 marks and trade names, coupled with added utiHty, 
 niceties of finish, evenness of quality, or more conven- 
 ient packages, serve at the basis for an increased de- 
 mand for the commodity upon the same price level as 
 substantially identical products. And in selling at the 
 market, superior promptness in delivery may become 
 a factor of great importance in increasing sales. 
 
 Recent developments in the textile industry illus- 
 trate the adoption of the policy of selling at the market, 
 combined with an attempt to increase sales at the 
 market price by a differentiation of the product. Ap- 
 parently the textile manufacturers who are branding 
 their goods do not seek to establish a new price level for 
 their product. Instead they aim to increase their sales 
 by building up a demand for their commodity in com- 
 petition with the product of other manufacturers at the 
 prevailing price level. 
 
 Chart IV is a graphic representation of one phase of 
 this policy. It brings out the idea that new consumers 
 may be drawn into the market at an existing price level 
 by giving the differentiated commodity an added im- 
 portance in the eyes of the consumer. To revert to the 
 economist's terminology again, a higher subjective val- 
 uation is placed upon the commodity by the consumer. 
 Its importance to him increases. Though at prevailing 
 prices his subjective ratio of exchange is too low to per- 
 mit purchasing the stock commodity, he may purchase 
 the differentiated commodity at the same price because 
 of a greater subjective valuation which he places upon 
 it. And this greater subjective valuation may be due 
 to the prestige of a well-known trade-mark or to the 
 greater convenience or service of an improved pack- 
 
 245
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 age, such as the air-tight cartons in which cereal prod- 
 ucts are distributed. 
 
 SelHng at the market plus is perhaps the most char- 
 acteristic price policy of modern distribution. In recent 
 
 X 
 
 a 
 
 j/ Selling at the Market 
 
 I ! ■ 'M 
 
 1 1 
 1 1 
 ■ 1 
 
 O 
 
 c C Y 
 
 Chart IV 
 
 This is an attempt to show graphically the effect of a stimulation of increased demand for 
 a commodity without any increase in the price at which it is marketed. 
 
 The ordinate, ox, is a scale of increasing price. The abscissa, oy, shows the number of pur- 
 chasers. The arc LM indicates the number of purchasers at any given price, growing less as 
 the price is increased and greater as the price is decreased. 
 
 If the established market price is represented by oa, the number of purchasers at that price 
 will be represented by oc. If then by stimulating an increased demand for his product, the 
 merchant-producer is able to increase proportionally the number of purchasers at each price 
 level, the demand curve LM will be replaced by L'M', and at the price, oa, a greater number 
 of purchasers, oc', will purchase. 
 
 This chart does not, of course, show how customers already in the market are drawn from 
 other merchant-producers to the purchase of a differentiated product for which a demand is 
 stimulated at the same price level as the products of the other merchant-producers. 
 
 years the effort among progressive manufacturers has 
 been not alone to anticipate unformulated human wants 
 and create new products to satisfy them, but to improve 
 staple commodities and develop new psychic values and 
 utilities in them. The limit set by a common market 
 price for commodities of the kind and class which they 
 manufacture has been rejected. They find opportuni- 
 ties of demand creation and profit in the difference be- 
 tween this established market price and the varying 
 
 ^6
 
 PRICE POLICIES 
 
 subjective valuation placed upon their commodities by 
 consumers of different purchasing power and different 
 social positions and personal habits. 
 
 The economists tell us the "consumer's surplus" is 
 the difference between the market value of a commodity 
 and the subjective value of the commodity to the user.^ 
 Each individual sets up for himself a ratio of exchange 
 between commodities which finds expression in the price 
 he would be willing to pay for any one of them rather 
 than go without it. These subjective valuations con- 
 stitute the demand side of the market. 
 
 The interplay of supply and demand gives rise in a 
 competitive market to a market price at which the con- 
 sumer can obtain the commodity. If this market price 
 is above that fixed by the subjective ratio of exchange 
 of the consumer, he drops out of the market, utilizing 
 his purchasing power to secure other commodities. 
 But if the market price is below that which the con- 
 sumer would be willing to pay to obtain the commodity, 
 he purchases; and the difference between his subjec- 
 tive ratio of exchange and the objective market ratio of 
 exchange constitutes his "consumer's surplus." The 
 man of means, for example, who buys his morning paper 
 for a cent, would still purchase it if the price were fixed 
 at five cents, at ten cents, or possibly more. Somewhere 
 in the ascending scale a point would be reached at which 
 even the man of means would drop out of the market. 
 But long before that point was reached, less well-to-do 
 readers would have ceased to purchase the paper. And 
 the difference between the price at which the well-to-do 
 man would drop out of the market and the market price 
 
 1 See F. W. Taussig, " Principles of Economics," Vol. I, page 128 and 
 following. 
 
 !W7
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 of one cent which he actually pays represents his per- 
 sonal "consumer's surplus." 
 
 The more able distributors turn, though usually un- 
 consciously, to the existence of this margin as the basis 
 of a demand for what is to all intents and purposes a new 
 commodity. That is, they differentiate a product from 
 a staple commodity for which a market price has been 
 established and establish an effective demand for the 
 modified product upon a new price level, higher than 
 that established for the commodity of which it is a 
 modification. 
 
 The means used for differentiation are more numerous 
 and more effective than those employed when the price 
 policy adopted is selling at the market par, since the 
 margin available for the differentiation is larger and the 
 potential gain greater. Sometimes slight modifications 
 are enough to render it better adapted to the use to 
 which it is put. Sometimes niceties of trimming and 
 equipment are sufficient. Sometimes a new and more 
 convenient style of package is used. Sometimes the 
 distributor builds up an atmosphere of good taste about 
 the goods or a reputation for constant quality which 
 insures the consumer against dissatisfaction. Some- 
 times the distributor depends upon "service" or special 
 conveniences to the consumer provided as collateral to 
 the commodity. Always, however, the aim is to isolate 
 his product from the stock commodity of substantially 
 like nature. And nearly always the distributor utilizes 
 trade-marks or trade names to identify his product as a 
 distinct commodity and to make sure of the repeat 
 order when the purchaser enters the market again. 
 
 He must then convey knowledge of his differentiated 
 product to those consumers whose subjective ratio of 
 
 £48
 
 PRICE POLICIES 
 
 exchange would have led them to pay a higher price for 
 the stock commodity before transferring their demand 
 to other goods. By calling attention to the superior 
 qualities, or convenience, or constant reliability of his 
 
 ^L' 
 
 Selling at the Market Pius 
 
 rf \ 
 
 ;' -«- 
 
 
 "^ M 
 
 Chart V 
 
 This chart illustrates the effect of the price policy termed "selling at the market plus." 
 On the ordinate ox is laid off a scale of prices for a staple commodity. The abscissa oy shows 
 the number of purchasers. 
 
 The demand curve LM indicates the number of purchasers at a given price, growing less 
 as the price increases and greater as the price decreases. Then if oa represents the market 
 price of the staple commodity, oc will represent the number of purchasers. Now if the 
 merchant-producer differentiates his product from the staple commodity, and stimulates a 
 demand for it, the effect is to increase the number of possible purchasers at each price level. 
 Thus the demand curve LM is replaced by the demand curve L'M'. 
 
 Obviously the merchant-producer may dispose of the differentiated product at a price oa', 
 higher than the price oa, without reducing the number of purchasers, oc. In other words, he 
 can profit by the increased demand through raising his price rather than by increasing his 
 sales. 
 
 differentiated product, he transfers to it a portion of the 
 demand that formerly found expression in the purchase 
 of the stock commodity. How this altered demand 
 develops is indicated graphically in Chart V. 
 
 The marketing of hats furnishes an illustration of this 
 development. If derby hats were distributed as a 
 staple, unbranded and at a single market price for a 
 given quality, many consumers would pay $3 for a staple 
 hat whose individual ratio of exchange would render 
 
 249
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 them willing to pay more than $3 for a hat rather than 
 go without. But certain producers have distinguished 
 their hats from the staple kinds by their brands. By 
 calling attention to niceties of trimming and finish 
 and by emphasis upon the correctness of their styles 
 some of them have built up a demand for their hats 
 at $5. 
 
 These trade-marked hats and the staple hats selling 
 at $3 are substantially the same commodity, but are 
 differentiated by detail modifications. These detail 
 differences render the well-to-do consumer willing to 
 pay a higher price for the trade-marked hat. No doubt 
 the demand for the more expensive hat depends in part 
 upon the consumer's feeling of security that his hat will 
 be of good quality and of proper shape if it bears the 
 name of a producer who has built up this market pres- 
 tige. This feeling of security forms a part of the sub- 
 jective valuation placed upon the hat. No doubt, too, 
 motives of emulation sometimes enter in, and the con- 
 sumer derives part of his gratification from the mere fact 
 that he purchases a hat which is known to sell at a 
 higher price than those purchased by his less well-to-do 
 neighbors. 
 
 It is of interest to note that other manufacturers of 
 branded hats have in recent years fixed their prices at 
 $4 and $6, appealing to consumers upon different price 
 levels from those reached by prior distributors of trade- 
 marked hats. Thus they reach with a $4 hat a group of 
 consumers not available to the distributors of $5 hats 
 because their subjective ratios of exchange did not ren- 
 der them willing to pay $5 for a hat. And with a $6 hat 
 they draw from the distributors of $5 hats certain of the 
 consumers whose subjective valuation of a hat renders 
 
 250
 
 PRICE POLICIES 
 
 them willing to pay more than $5 for one that pleases 
 them. 
 
 The policy of selling at the market plus makes the 
 most severe demands on the ability of the distributor. 
 To succeed he must have an unusual equipment, in- 
 cluding knowledge of human nature, of the psychologi- 
 cal reactions of the individual consumer, and must be 
 able to give proper weight to such motives as social emu- 
 lation and all the varied factors that enter into the con- 
 sumer's subjective ratio of exchange. 
 
 The process resulting from the increasing adoption of 
 this price policy and the increasing differentiation of 
 commodities at various price ranges is closely analogous 
 to the creation of new commodities. When the hat 
 trade splits up into a number of individual brands, prac- 
 tically distinct commodities at diflFerent price levels, 
 the situation from a social point of view is little dif- 
 ferent from that arising from the creation of new com- 
 modities which are not merely modifications of existing 
 commodities. 
 
 If the safety razor be regarded as a new commodity 
 rather than as a modification of the old-stj'le razor, it 
 provides us with an opportunity to test the social justi- 
 fication for the creation of a new commodity. When the 
 first widely advertised safety razor was put upon the 
 market at $5 a considerable margin of profit was left 
 to the producer. It was said at the time that the ac- 
 tual cost of manufacture was less than $1. Now this 
 wide margin of profit made possible an extensive adver- 
 tising campaign which brought the new device to the 
 attention of the entire consuming public. Everyone, 
 whether in the large centers or remote districts, learned 
 of the safety razor and its uses. Great numbers pm-- 
 
 251
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 chased the razor because the subjective valuation which 
 they placed on the new commodity exceeded the price 
 asked. The large reward received by the distributor may 
 properly be regarded as compensation for bringing 
 about a better adjustment to meet human needs. 
 
 To-day the safety-razor demand is well established, 
 and those consumers whose individual ratios of exchange 
 do not render them willing to pay $5 for a safety ra- 
 zor are able to gratify their conscious need by a choice 
 of many similar products at prices ranging as low as 
 twenty-five cents. It is an interesting problem for the 
 producer whether, in view of the enormous demand for 
 blades, the initial price of $5 might not well have been 
 reduced in later years and selling eflFort concentrated 
 on the production and distribution of blades. Every 
 extra razor sold would mean the creation of an added 
 and more or less permanent demand for blades. This 
 is the sales policy followed in marketing many office 
 and factory appliances; the initial price of the device 
 is little more than its delivered cost and the profits 
 come from the sale of supplies necessary to its op- 
 eration. 
 
 Now when the producer of a commodity already mar- 
 keted by other producers sets off his commodity from 
 others of like kind, and sometimes even by minor im- 
 provements is enabled to build up a demand for it on a 
 higher price level than that of the stock commodity, he 
 has made a more accurate adjustment in supplying 
 human wants and has brought the possibility of this 
 more accurate adjustment to the attention of consumers. 
 The purchaser of a trade-marked hat at $5 would buy a 
 staple hat for $8 if the $5 hat did not give him equal or 
 greater proportional gratification, taking into account 
 
 252
 
 PRICE POLICIES 
 
 the differing objective ratio of exchange. Obviously, 
 the consumer who buys a trade-marked hat does so be- 
 cause he prefers to pay $5 for such a hat rather than $3 
 for an unbranded hat, and to criticise the payment of the 
 additional $2 for the differentiated product because the 
 modifications are not substantial is to attempt to sub- 
 stitute for the subjective valuation of the consumer as 
 a basis of exchange an external social standard. The 
 more highly differentiated the scale of commodities is, 
 the more accurately will it be possible for the individual 
 consumer to satisfy his material wants. 
 
 For true value is not objective, but subjective. The 
 practical basis of exchange is the extent to which the 
 article will satisfy the desire of the purchaser. The 
 price you can secure depends on the intensity of this 
 desire, not on the material value of the finished product, 
 such as its calories of food value or its wearing quali- 
 ties. The problem, then, is to intensify the demand for 
 your goods, to add to the material value a psychic value 
 in the gratification the customer experiences in its color, 
 flavor, design, or style. 
 
 The manufacturer who is successful in establishing a 
 differentiated product as a distinct commodity on a new 
 price level is, for a time, in the position of having a mo- 
 nopoly as to the differentiated commodity. Such com- 
 petition as he has is the indirect competition of the 
 similar staple. His position often enables him to obtain 
 temporarily a margin of profit out of all proportion to 
 the cost of the actual improvements in the differen- 
 tiated product. This, again, may be justified as a re- 
 ward for turning the resources of nature to the better 
 satisfaction of human wants. As advertising is the 
 price we pay for instruction in better methods of satis- 
 
 253
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 fying our wants, so this is the price we pay for the inven- 
 tion of better methods. Of course, if the poHcy of sell- 
 ing at the market plus results in a higher price on a 
 staple necessity for the marginal consumer, it goes be- 
 yond all justification. But this, I believe, will rarely 
 if ever happen. In the main, the policy affects only the 
 specialties. And even from these the large percentage 
 of profit will decrease as other producers, observing 
 the pioneer's gains and following his example, differ- 
 entiate their products from the original staple in the 
 same or in different ways. This rise of competition at 
 the new price level and the premium put on features 
 which make selling points will tend to force in the dif- 
 ferentiated commodities the substantial improvements 
 in quantity, quality, or service warranted by the 
 higher price. 
 
 254
 
 CHAPTER XVI 
 
 ORGANIZATION OF DEMAND CREATION 
 COMBINATION OF AGENCIES 
 
 rr^HE fundamental purpose in demand creation is ef- 
 -*- fective transmission of ideas about the goods to 
 prospective buyers at the lowest unit cost. Analysis of 
 the market will locate these probable customers. A wise 
 price policy will establish a balance between the widest 
 practicable distribution and maximum profit. Labora- 
 tory tests of the materials of demand creation will de- 
 termine which selling points have strongest appeal and 
 the quantity and sequence in which they are most con- 
 vincing. Similar studies and "try-outs" will show the 
 relative efficiency of the middleman, the direct sales 
 force, and the various kinds of advertising in communi- 
 cating ideas to possible purchasers. And a survey of 
 the considerations involved in the location, construc- 
 tion, and equipment of the sales plant will show how 
 best to turn these factors to the main purpose. 
 
 All these activities are preliminary. They produce 
 results only when the agencies of demand creation are 
 set in motion. Before this can be done, or at least be- 
 fore the forces of distribution find equilibrium of great- 
 est profit, one further problem must be solved — what 
 combination of agencies is to be used to reach the mar- 
 ket and what part of the work of demand creation is to 
 be performed by each. 
 
 There are manj^ marketing situations, to be sure, 
 where a single agency of demand creation is adequate. 
 
 i55
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 If stress is put upon the combinations of two or more, it 
 is because relatively few business men recognize to what 
 extent this is a possible problem for them. The major- 
 ity still gamble on their business instinct and experi- 
 ence; and the success or failure of the campaign is the 
 only evidence they have whether these were or were not 
 safe guides. If their previous experience has indicated 
 that advertising or a direct sales force is an efficient 
 agency of demand creation, they are likely to adopt it 
 for distributing the new product although the latter 
 may be quite different in character. If they make any 
 studied comparison of the various agencies, it is prob- 
 ably on the basis of average cost of selling, without any 
 serious attempt to analyze the market from the view- 
 point of its accessibility to each agency or to consider 
 each agency from the angle of its ability to transmit the 
 particular kind of ideas about the goods involved. 
 
 In fact the business man too seldom realizes how in- 
 tricate is the problem of determining the agency or the 
 combination of agencies which is exactly adapted to 
 reach the various geographic or social and economic 
 strata of his market. Too often he adopts one method 
 and becomes an advocate of it, disregarding other meth- 
 ods entirely or taking them up in a perfunctory way. 
 While it is true that the method adopted may be more 
 efficient than any other single method, for his particular 
 work of demand creation the solution of the problem 
 by no means stops here. For the agency which is rela- 
 tively efficient in reaching one geographic area may be 
 inferior to another method in reaching another area. 
 And a system of distribution which has proved effective 
 in reaching buyers on one economic level may be com- 
 paratively ineffective when employed to reach another. 
 
 256
 
 COMBINATION OF AGENCIES 
 
 The problem of finding the most effective combination 
 of agencies, then, is most complicated. Each distinct 
 area and economic and social group must be treated as 
 a separate unit for study and analysis. 
 
 The manager selling through salesmen ordinarily 
 finds a decreased selling cost as he increases his sales 
 within his immediate territory. But as he widens his 
 market, the selling cost usually increases. 
 
 Here a combination of salesmen and direct advertising 
 may cut his cost of demand creation. He may, for in- 
 stance, reduce the number of salesmen's visits by one 
 half and either prepare the way for them or supplement 
 their efforts by a series of "follow-up" letters or by per- 
 sonal correspondence. In areas more distant or with a 
 widely scattered population, it may be profitable or even 
 necessary to eliminate the salesmen entirely. Selling by 
 mail through direct advertising may be the most effec- 
 tive method, cost considered. The important thing al- 
 ways is that no one agency should be adopted to the 
 exclusion of all others merely because its use has proved 
 profitable in previous ventures. There may be sections 
 where it is positively unprofitable and where it can be 
 supplemented or entirely supplanted by other methods 
 of demand creation. 
 
 The method of sale is a factor of importance in the 
 problem of agency selection. If sale is in bulk, the 
 purchaser seeing the actual goods before buying, dis- 
 tribution through a series of middlemen is generally 
 most feasible. In such a situation expense ordinarily 
 prohibits direct handling. Only in exceptional cases, 
 where, for example, small household appliances are sold 
 by door-to-door canvass, is the handling of the product 
 by exclusive salesmen practicable. Sale in bulk is also 
 
 257
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 possible by either direct or general advertising, where 
 the size and nature of the product permit its sale and 
 shipment on approval or subject to trial by the pur- 
 chaser. 
 
 If sale by sample is the method best adapted to the 
 commodity in question, middlemen or direct salesmen 
 are likely to be the most eflBcient agencies of demand 
 creation. Such commodities, indeed, are distributed 
 through both middlemen and direct salesmen, the sale 
 at each stage being by sample until the final movement 
 from retailer to consumer, when the sale becomes one 
 of bulk. Direct salesmen in the majority of cases sell 
 from sample. And even demand creation by direct ad- 
 vertising is in some cases practically sale by sample. 
 For the distributor by mail of a commodity which is not 
 bulky may send for inspection and test a sample of the 
 commodity with his direct advertising. 
 
 Where sale by description is used exclusively, direct 
 or general advertising is likely to be the most effective 
 agency. Increasing knowledge and skill on the part of 
 advertising managers, writers, and illustrators, coupled 
 with recent advances in the mechanical processes of 
 reproduction and printing, have enlarged the scope and 
 effectiveness of advertisements until they seem alone an 
 adequate and economical means of demand creation for 
 a multitude of commodities and staple specialties. The 
 establishment of a domestic parcel post equipped to 
 convey twenty -pound packages to any part of the coun- 
 try and to make collections on delivery has added fur-: 
 ther effectiveness to advertising by providing a cheap, 
 popular, and universal agency for the supply of goods 
 which have been marketed by printed and written 
 symbols. 
 
 258
 
 COMBINATION OF AGENCIES 
 
 It is possible, though generally not economical, to dis- 
 tribute a commodity through a series of middlemen, the 
 sale at each stage being accomplished by description. 
 And the use of salesmen in selling by description is very 
 common. Take the cases where furniture, heavj' ma- 
 chinery, hardware, and like commodities are sold from 
 photographs, from catalogs, and more recently through 
 the use of moving pictures. 
 
 The extent to which any one agency of demand crea- 
 tion can be used must not be decided alone, however, 
 on theoretical grounds. In many cases the middleman 
 is already strongly intrenched and the whole business 
 structure is based upon his activities. His relations with 
 his customers are long established, his retention of many 
 of his traditional functions, such as financing operations, 
 sharing the risk, and even transporting the goods, 
 clinches his hold on his trade territory and makes him a 
 necessary factor in reaching the dealers or consumers 
 he serves. 
 
 Choice of the agency for demand creation also reacts 
 upon the materials of demand creation (ideas about the 
 goods) and upon the goods themselves. Under the or- 
 thodox type of distribution, the basis of the sale to the 
 middleman is his opportunity to dispose of the goods 
 at a profit. Because of this emphasis on price and sala- 
 bility, the tendency of the orthodox system in market- 
 ing unbranded commodities is to turn the energies of 
 the producer toward lowering the cost of production 
 and reducing the price. The influence of satisfaction or 
 dissatisfaction on the part of the consumer comes to him 
 only indirectly through a chain of middlemen. jMore- 
 over where the goods are not differentiated by trade- 
 mark or trade name, their identity is often completely 
 
 ^9
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 lost in the successive stages of distribution. Even the 
 retailer in many communities and neighborhoods con- 
 cerns himself rather with salability than with ultimate 
 satisfaction to the customer. And if the customer is 
 not perfectly satisfied and accepts the product only be- 
 cause it is the best he can get, his objections must or- 
 dinarily be relayed through two or more middlemen 
 before they can reach the producer. 
 
 Only marked defects in quality, therefore, are likely 
 to be brought to the attention of the producer. Thus 
 he fails to establish that touch with consumers which 
 would guide him toward improvements in quality and 
 service in his goods. Since his attention is not directed 
 primarily to those elements, leakage of demand is likely 
 to result. He is not manufacturing the precise com- 
 modity which the market desires and would buy with 
 the least outlay of money and energy for demand 
 creation. 
 
 The differentiated, trade-marked product, on the 
 other hand, encounters a barrier to a wider market in 
 the fact that it competes for the middleman's interest 
 and selling effort directly with the latter's own "house 
 brands" and with articles of lesser quality or prestige 
 which allow him a greater unit margin of gain. Few 
 wholesalers consider themselves established with their 
 trade until they control several lines marketed under a 
 firm name or symbol. Nor is the "house brand" any 
 longer the exclusive possession of the jobber. The larger 
 retail stores feature many trade-marked "staple special- 
 ties" of their own, if the phrase "staple specialties" is 
 permissible to describe commodities only slightly dif- 
 ferentiated from the standard market types. 
 
 The "house brand" may be regarded as a logical 
 
 360
 
 COMBINATION OF AGENCIES 
 
 outcome of current conditions in distribution, of suc- 
 cessful attempts by manufacturers to market advertised 
 products through the regular channels of the trade. 
 The wholesaler, arguing that his salesmen do an indis- 
 pensable part of the actual work of demand creation, 
 carrying knowledge and samples of scores of lines into 
 remote villages which few single manufacturers could 
 afiFord to cultivate, sees no ethical or practical obstacle 
 in the way of bringing out similar articles under his own 
 name and pushing these instead of the manufacturer's 
 advertised brands. 
 
 From the viewpoint of the jobber and the retailer 
 alike, the advertised product presents an element of 
 danger in that its sale cannot be controlled. After con- 
 tributing what seems to him an important factor in the 
 creation of demand and the securing of general distribu- 
 tion for such articles, the wholesaler may see his dis- 
 count cut; and the retailer may be compelled either to 
 accept an increased sales quota or to give up a line which 
 has become identified with his store through his adver- 
 tising and through selling talks with his customers. 
 The "big stick" policy in handling the trade is so recent 
 in the memories of many middlemen that it still colors 
 their attitude toward those newly advertised products 
 and advertising manufacturers whose reputations for 
 fair treatment of their distributors have not been sub- 
 jected to decisive tests. 
 
 It is only a nationally exploited line of clothing or 
 shoes, for instance, which can secure mention of its 
 individual trade-mark in the advertising of certain im- 
 portant stores necessary to its distribution scheme in 
 the larger cities. The policy generally adopted by these 
 stores in connection with advertised articles not ex- 
 
 S61
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 clusively controlled is to stock and supply them when 
 called for, but to give every display advantage to their 
 private brands and to offer these when the customer 
 does not request a specific product. This is certainly 
 fair merchandising — presuming that the house brands 
 are the equals of the trade-marked specialties in quality 
 and service. Yet the manufacturer who has blazed the 
 way in a particular field may properly feel that such 
 practice deprives him of some of the fruits of his cam- 
 paign of demand creation. 
 
 The opposition of middlemen to any reduction of their 
 margin is general enough to present a serious problem to 
 the producer. Often the latter postpones taking over 
 the function of communicating ideas about his goods 
 directly to the user because he sees that he must con- 
 tinue to pay the middleman for that work, whether or 
 not it is performed by him, or else develop independent 
 agencies of physical supply. The latter step involves, 
 of course, the outlay of considerable capital for the es- 
 tablishment of branch warehouses and organizations or 
 of branch retail stores in such centers of population as 
 promise suflicient volume of potential sales. 
 
 On the other hand, retailers in certain fields have been 
 educated, through observation or demonstration of the 
 effect of advertising on sales, to acceptance of a lowered 
 margin or trade-marked goods in exchange for repeated 
 campaigns of demand creation. This may be witnessed, 
 for example, in clothing, or more recently in shoes, where 
 national exploitation of one trade-marked line has given 
 it an exclusive agent in almost every town of importance 
 in which the company does not maintain a branch store. 
 In other lines, producers have found it advantageous to 
 undertake the wholesaler's work of physical supply 
 
 262
 
 COMBINATION OF AGENCIES 
 
 rather than compensate him for a selhng function no 
 longer performed. 
 
 In directly assuming the burden of demand creation 
 the manufacturer has choice of two agencies, an ex- 
 clusive sales force and general or direct advertising. 
 Since their function is the same, the transmitting of 
 ideas about the goods not only to the ultimate users but 
 also to such middlemen as it may be advantageous to 
 employ in distribution, the analogy between them is very 
 close. In many fields they may be used as substitutes, 
 one for the other; yet each retains its zones of special 
 utility and both reach their highest efficiency, as a rule, 
 when they supplement one another in the work of 
 demand creation. It is a significant fact, in this connec- 
 tion, that some of the largest mail-order houses, com- 
 mitted to a policy of direct advertising, have found 
 it profitable in certain districts to employ salesmen to 
 take orders for their merchandise. 
 
 The combination of sales force and advertising is often 
 advantageous because the personal equation of the in- 
 dividual "prospect" is an unknown quantity in demand 
 creation. You may discover and standardize and record 
 a thousand selling points about your product. From 
 this material you may construct a sales argument whicli 
 is logical and decisive, and then reduce it to a written 
 or printed form which disarms criticism and defies bet- 
 terment. The special and exclusive virtue of advertis- 
 ing, indeed, is that it allows you to transmit to the 
 reader exactly those ideas you wish to convey in the 
 exact form that seems best for their transmission. But 
 unless you are intimately acquainted with every possi- 
 ble customer you address and are able to adapt your 
 selling talk, your pictures, your diagrams, to his wnme- 
 
 263
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 diate state of mind, your standard argument will shoot 
 wide much oftener than it hits the mark. If your circle 
 of prospective users is broad enough, you can content 
 yourself with this average result. If the range of your 
 product is narrow, you must find some way to adapt 
 both approach and selling talk to the attitude and sit- 
 uation of the individual buyer. 
 
 Adaptability is the salesman's strength, just as ac- 
 curacy in the transmission of ideas is a dependable 
 quality of advertising. The average business man, as 
 an individual prospect, does not exist. His moods, his 
 interest in proposals not plainly relevant to the task 
 he happens to have in hand change with the passing 
 hours. The approach or selling talk which would merely 
 rouse impatience at a time when he is absorbed in a 
 problem might secure quick recognition and interest 
 when that problem is disposed of. 
 
 But advertising cannot choose the moment of its ap- 
 proach except in a very limited or a very general way. 
 If direct or local, for example, it can avoid Monday 
 mornings and the crest of "rush" seasons like the weeks 
 before Christmas and Easter in retail stores. If general, 
 it can choose its mediums with the aim of presenting 
 ideas at the time or in the place where the prospect is 
 most likely to entertain them and it can concentrate on 
 the seasons when buying is general. But all advertising 
 lacks the salesman's ability to sense immediate condi- 
 tions which may be either propitious or adverse and to 
 base action on what is thus perceived. 
 
 The same capacity in a seasoned man allows him to 
 feel out the prospect's attitude toward the product or 
 service he represents and to shape his talk so that he 
 can avoid touching convictions or prejudices which 
 
 S64
 
 COMBINATION OF AGENCIES 
 
 might interfere with the sale. Moreover he can discover 
 the objections which occur to the prospect as he con- 
 siders purchasing. By meeting them he can turn them 
 into buying arguments; at the least he can take them 
 away with him for further study and future answer. 
 Any specialty salesman of experience will agree that he 
 cannot forecast the workings of a new prospect's mind 
 or determine beforehand just what objections will 
 emerge. Where advertising stakes all its chances on 
 finding the average buyer in an average mood, the 
 salesman can bring as much intelligence, tact, and skill 
 as he possesses to the problem of avoiding the unfavor- 
 able occasion and inducing a receptive attitude toward 
 what he has to offer. 
 
 Once the prospect's attention is engaged, the sales- 
 man has the appeal of personal presence to give warmth 
 and color to his assembly of facts and figures. For 
 personality still has much to do with the transactions 
 of business. Few products have an actual monopoly 
 of value or utility; the enthusiasm or tact or resource 
 of the salesman frequently is the decisive factor in in- 
 fluencing the purchase. There is more than one in- 
 stance on record where a sales force which had analyzed 
 and developed to the full the consumer uses of an in- 
 ferior product has held the market against a better 
 article until the factory experimental rooms, working 
 night and day, could match the merits of the competing 
 product. 
 
 In the distribution of many staple specialties like 
 furniture, shoes, dry goods, and the like, the assurance 
 of a straightforward, competent, and likeable salesman 
 that the "goods are right" often carries greater weight 
 with the retailer than the guarantee of the house behind 
 
 265
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 them. So true is this that within their territories many 
 travehng men are virtually in business on their own ac- 
 count; by grace of long acquaintance and fair and 
 friendly dealing they have acquired a hold on their trade 
 which it is no easy matter to overcome. Recognition of 
 this fact has prompted many intensive advertising cam- 
 paigns to establish a direct contact with dealers or con- 
 sumers and emphasize the house rather than the sales- 
 man as the responsible force behind the merchandise. 
 
 Because of his personal contact and adaptability, the 
 salesman can also build a convincing argument from ma- 
 terials which advertising would present much less ef- 
 fectively. He can stage his sale. If a sample of the 
 product is the key argument (it usually is in the case of 
 a specialty), he can introduce it at the exact moment 
 when the other man is prepared to concentrate upon it. 
 If it is a machine of any sort, he can demonstrate those 
 functions which would be of importance in the prospect's 
 business. Better yet, he can lead the latter to operate 
 the thing himself and thus come a step nearer to under- 
 standing and accepting it. If it is a new fabric, an 
 unfamiliar food product, a differentiated garment or 
 line of garments, he can analyze it from the standpoint 
 of the individual he is addressing and shape his sales 
 talk to drive home its advantages either for direct use 
 or for resale to consumers. 
 
 More important still, when the salesman has created 
 a demand for his goods he is at hand to close the order 
 immediately. Experience or intuition tells him when 
 this moment comes, and his will power goes into the 
 balance to overcome the buyer's inertia, his tendency 
 to put off decisions when decision means paying out 
 money. So generally is this negative, defensive motive 
 
 266
 
 COMBINATION OF AGENCIES 
 
 recognized that sales managers urge on their men at all 
 times the supreme importance of closing the order. To 
 overcome this inertia when advertising is the agency 
 used, the demand created must be much more impera- 
 tive. There is no salesman at hand to seize the propi- 
 tious moment, and some positive action must therefore 
 be suggested to the prospect, the signing of an inquiry 
 card, telephoning the local agent, or telling him of a 
 store where the goods can be secured. 
 
 The direct sales force and producer advertising, then, 
 are different agencies for accomplishing the same end. 
 This is to transmit, or to secure the transmission of, 
 ideas about the goods to the prospective buyer's mind 
 more directly and more accurately than the orthodox 
 chain of middlemen will transmit them. 
 
 The character of his product may permit a manufac- 
 turer to send his salesman or his advertising direct to 
 the ultimate user and to make direct deliveries by ex- 
 press, freight, or parcel post, or through his own branch 
 houses or stores. Again, distribution through retailers 
 may be the only method possible, and the nearest his 
 salesmen can come to influencing the prospective con- 
 sumer is to teach dealers and their clerks its outstanding 
 selling points and to train them to use and demonstrate 
 these intelligently. Here he can supplement this in- 
 direct contact with advertising aimed at the consumer, 
 the dealer, and the clerk. Or, needing both the whole- 
 saler and retailer for physical supply of his tooth paste, 
 gloves, watches, breakfast food, or whatever his product 
 is, he may put the burden of demand creation on ad- 
 vertising, backing it up with introductory or supple- 
 mentary campaigns of sampling and demonstration to 
 prospective users. 
 
 267
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 In comparing salesmen with different forms of ad- 
 vertising, the business man often judges both agencies 
 upon the direct returns over a short period. This does 
 not take account of all three classes of demand aroused 
 by selling effort — expressed conscious demand, unex- 
 pressed conscious demand, and subconscious demand. 
 The direct and immediate return from selling efforts de- 
 pends solely on expressed conscious demand. But the 
 business man must consider the unexpressed conscious 
 demand and the subconscious demand. Take the case 
 of an advertised collar. A man notices the advertise- 
 ment, reads it, and decides that at some future time he 
 will try it. Months later perhaps he does so. His 
 purchase is not reflected in the immediate returns, yet 
 clearly it is a result to be reckoned with in any compar- 
 ison. Or suppose the man merely notices the adver- 
 tisement. At a later date, when purchasing collars, he 
 is shown the advertised brand along with another. The 
 brand being vaguely familiar from the advertisement, 
 he purchases it in preference to the others. Here, too, 
 the aroused demand is of a degree not reflected in imme- 
 diate returns, yet it is of value to the distributor. 
 
 It is obvious, then, that in balancing the advantages 
 of selling through salesmen against selling through ad- 
 vertising in whole or part, the business man must con- 
 sider not only the demand reflected in the direct imme- 
 diate returns alone but also the lesser degrees of demand 
 which, while not immediately effective, bring later re- 
 sults and contribute to make demand continuous. 
 
 Thus a salesman might make fifty calls at an expense 
 of $100, and ten sales might result from his efforts. Or 
 for the same $100, 5,000 pieces of direct advertising 
 could be mailed, resulting perhaps in only eight sales.
 
 COMBINATION OF AGENCIES 
 
 Or, again, if the same $100 were used in the insertion 
 of a page advertisement, in 100,000 of the circulation 
 of a standard magazine, only six would result. Now it 
 is apparent that, judging by the direct results, the 
 salesman is the most efficient agency of distribution, 
 the direct advertising next, and the magazine adver- 
 tising least efficient. But the manager will bear in mind 
 that, while the salesman made ten sales, he had only 
 . forty opportunities to create the lesser grades of de- 
 mand. The direct advertising gave 4.992 opportunities 
 for the creation of demand falling short of immediate 
 expression, and the magazine advertising, perhaps, 
 49,994 such opportunities, assuming that one person 
 saw the advertisement in half the copies printed — not 
 an improbable assumption since each copy of a maga- 
 zine is usually read by several persons. 
 
 Aside from furnishing a basis for laboratory tests of 
 ideas, direct advertising, especially the written com- 
 munication, has certain distinct advantages. It is pos- 
 sible to convey the exact shade of meaning, in detail and 
 elaboration. The letter writer has absolute supervision 
 over the written symbols he is using. Style, substance, 
 manner of presentation are under his direct control. 
 He can balance the time devoted to description and 
 argument on the one hand against the time devoted to 
 the anticipation of objections on the other. He can 
 present his exact proposition, avoiding misrepresen- 
 tation or misunderstanding. He can be courteous or 
 olunt, ceremonious or informal, conservative or forcible, 
 as the occasion may require. 
 
 The advantages and limitations of direct advertising 
 cannot, of course, be reduced to direct measurement. 
 Yet direct advertising, general advertising, and the use 
 
 269
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 of salesmen can be measured on the basis of a common 
 unit — cost, as measured in the effectiveness of calls. 
 Say, for example, that the cost for 100 calls by the sales- 
 man is $100, the cost for 100 letters is $3.35, and the 
 cost for 1,000 readers of a page advertisement in a 
 magazine is $1. A hundred dollars in the first case 
 reaches 100 prospective customers, in the second case 
 3,000, in the third case 100,000. Which method pro- 
 duces the most orders on a given expenditure.'^ That is 
 the test. 
 
 A sound selling policy, then, must be built up on a 
 careful analysis of the market by areas and strata and 
 upon a detailed study or the proper agency or combina- 
 tion of interrelated agencies to reach each area and 
 stratum. The economic generalizations expressed in 
 the law of diminishing returns must be recognized. A 
 sound selling policy, furthermore, must weigh not only 
 the direct results obtained from the use of one or the 
 other agency over a short period, but also the less 
 measurable results represented by the unexpressed con- 
 scious demand and subconscious demand which go to 
 aid the selling campaign of the future. And at every 
 point the elusive human element is present and cannot 
 be left out of the account. 
 
 In any study of the agencies of demand creation the 
 drift toward a shortened cycle of distribution presents 
 one of the most interesting developments in modern 
 business. It is significant of broad changes and read- 
 justment in our market machinery and it is also a factor 
 to be carefully considered by the producer in selecting 
 his agencies of demand creation. 
 
 Starting with the manufacturer's assumption of va- 
 rious functions performed by the middleman, it has 
 
 270
 
 COMBINATION OF AGENCIES 
 
 brought about a counter movement with the same end 
 among both wholesalers and retailers. On the one side, 
 many manufacturers of shoes, clothing, and silverware 
 (to mention only the most typical lines) have virtually 
 eliminated the wholesaler, selling direct to the retailer. 
 Many others have dropped all middlemen and deal with 
 the consumer in person, either by the mail-order route, 
 by exclusive stores, or by special salesmen. 
 
 To compensate for this loss in potential volume, the 
 natural growth of the country keeping up his accus- 
 tomed average of sales, the wholesaler in his turn has 
 developed his super-profitable "house brands" and re- 
 duced the chain by one member, turning manufacturer 
 himself. And even here there is divergence in method. 
 Of two great merchandising houses in Chicago, one has 
 adopted the policy of ownership of many important 
 sources of supply, while the other follows the radically 
 different course of control of many small producers 
 by furnishing them with working capital, patterns or 
 style suggestions, and a market for their total output, 
 thus allowing them to go back to the original function 
 of the manufacturer and concentrate on production 
 problems alone. The large retailer, therefore, has short- 
 cut distribution as effectively as the advertising manu- 
 facturer, by owning or controlling sources for several 
 of his principal lines. The sum of these parallel move- 
 ments has been in the direction of simplification and 
 economy for a relatively small but significant section of 
 business. 
 
 Out of the antagonistic yet concurrent efforts of the 
 producer, the wholesaler, and the retailer to simplify 
 distribution by elimination will probably come a class 
 of eflBcient, broad-gauged middlemen (already indicated 
 
 271
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 by the rise of such individuals in every field) who will 
 recognize that each function in distribution is best per- 
 formed by its appropriate agency and will cooperate 
 with the producer in getting rid of duplicate effort and 
 in securing the greatest number of turnovers per dol- 
 lar of capital and expense. The integration of trade 
 and industry can be effected by the coordination of 
 functions and individual efforts. It is not dependent 
 on the concentration of ownership and control. 
 
 «72
 
 SECTION n 
 PHYSICAL SUPPLY
 
 CHAPTER XVII 
 
 PLANT AND OPERATION POLICIES 
 
 "pHYSICAL supply of the goods is the final step in 
 -*- distribution, supplementing the work of demand 
 creation and making it effective by putting the product 
 in the consumer's hands. Through the various agen- 
 cies developed for the purpose, it applies further mo- 
 tion to the material already changed in form by the 
 activities of production. In this application of motion 
 to change the place of the transformed material, demand 
 creation may seem to have the more important part. 
 But both are necessary to bring about the new place 
 and ownership utilities which are the essence of selling; 
 and both are so interdependent in their relations that 
 policies and methods cannot be fixed for one without 
 keeping in mind their influence on the operations of the 
 other. 
 
 The problem of physical supply can be simply stated. 
 How shall the maximum of aroused demand be satisfied, 
 with minimum demand leakage and least expense, pres- 
 ent and future sales both being considered? Time, 
 convenience, and service are the three things the pro- 
 spective buyer looks to; the producer's task is to provide 
 the goods when they are wanted, where they are wanted, 
 and in the condition and volume which the market at 
 the moment requires. 
 
 Minimum leakage of demand is one of the main con- 
 siderations. Aroused demand is not only a useless thing, 
 
 275
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 but is actually detrimental unless the product can be 
 delivered before the edge of curiosity and desire is dulled. 
 The woman who asks her grocer for a new naphtha soap, 
 whose lasting qualities she has seen advertised weekly or 
 monthly, will soon lose patience and interest in the 
 quest if the grocer assures her that he cannot find it in 
 the market. If she trusts her grocer — and no manu- 
 facturer can ignore the fact that customer confidence is 
 the solid basis of retail trading — she will accept his 
 statement and dismiss the claims of the new soap, par- 
 ticularly if inquiries at other stores show that they also 
 have not stocked it. From expressed conscious demand, 
 she drops down the scale of interest to the zero level ; 
 and disappointment induces a negative attitude which 
 must be overcome before the process of demand creation 
 can be begun again. 
 
 If the product is suflSciently differentiated from the 
 staple offerings, like, for example, a naphtha soap which 
 will not dissolve too rapidly in hot water, appeals to the 
 economy motive may renew her desire to make a trial 
 purchase. But for a commodity only slightly differ- 
 entiated from the standard, an initial breakdown in 
 physical supply like this forms a serious hindrance to 
 distribution. There is lost ground to be regained; the 
 materials of demand creation lack their first fresh ap- 
 peal; and the prospect's experience has made her skep- 
 tical of the manufacturer's promises of service. 
 
 It is true that consumers' requests are the best evi- 
 dence of aroused demand, and that middlemen generally 
 require some such evidence before they will purchase 
 a specialty competing with an item already in stock. 
 Hence a delay, and thus a dilemma for the producer. 
 It is, indeed, a serious problem so to coordinate the 
 
 «76
 
 PLANT AND OPERATION POLICIES 
 
 demand-creation campaign and the plan for putting 
 the goods within reach of prospective buyers that the 
 interval between the expression of aroused demand and 
 the delivery of the product shall be as short as possible. 
 The ideal, of course, is immediate delivery; and there 
 are practicable ways of attaining this ideal or of dis- 
 counting the effect of the delay on the prospective buyer. 
 We shall say more about them a little later. 
 
 The point to be emphasized at the outset is that the 
 leakage of demand due to defective physical supply 
 puts a heavy tax on the operations of distribution. How 
 great this waste is cannot be estimated with any degree 
 of accuracy; but it is fair to say that a large proportion 
 of unsuccessful advertising campaigns have failed be- 
 cause provision was not made for physical supply of the 
 goods on a scale as broad as the advertising plan. Leak- 
 age of demand resulted because the vital principle of 
 balance was not observed. 
 
 While sales were made in bulk, and in the many cases 
 where they are still made in bulk, the problem of phys- 
 ical supply, as distinct from demand creation, cannot 
 arise. Delivery is made at the time of purchase; de- 
 mand creation and physical supply are coincident. But 
 where sale by description or sale by sample is employed, 
 a demand is created which is effective only if the goods 
 can be made readily available to the prospect. If a con- 
 siderable expenditure of energy is required of the cus- 
 tomer or if there is a considerable lapse in time before 
 the demand created is satisfied, a large share of the 
 aroused demand will be lost. Time and convenience are 
 thus essential elements. If any rule is to be formulated 
 it is that there must be the least possible interval 
 between exciting demand and satisfying it. Demand 
 
 «77
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 creation without effective physical distribution repre- 
 sents pure waste. 
 
 Approaching his problem of physical supply, the man- 
 ufacturer finds that its activities break up, as in produc- 
 tion and demand creation, into two main groups. There 
 are plant policies, having to do with the location, con- 
 struction, and equipment of his warehouses and their 
 attendant shipping facilities; and there are operating 
 policies which govern the problems of materials, agen- 
 cies, and organization. The classification is indeed 
 more obvious here than in demand creation for the 
 reason that the factors dealt with once more are tan- 
 gible and measurable things. 
 
 The materials, for example, are the goods themselves. 
 The agencies are either: (1) functional middlemen in 
 the transportation field, railroads, express companies, 
 and the parcel post; (2) areal middlemen, wholesalers 
 and retailers; or (3) agencies of direct supply, branch 
 houses, and exclusive agents or retail stores. Supple- 
 menting the agency of transportation are the pro- 
 ducer's own delivery facilities for transportation of the 
 goods for part or all of the way from the main plant or 
 branch house to the place where the consumer comes in 
 contact with them. To the traditional groups of areal 
 middlemen also might be added the special types more 
 recently evolved, such as wholesale and retail catalog 
 houses, buying associations of dealers and consumers, 
 chain stores, and warehousing and delivery concerns. 
 
 It appears at a glance that here are three distinct 
 economic functions : (1) transporting the goods to create 
 place utility, (2) storing the goods to create time util- 
 ity, and (3) physical transfer of the goods to create 
 ownership or possession utility. The problem of phys- 
 
 278
 
 PLANT AND OPERATION POLICIES 
 
 ical supply is complex because the agencies which the 
 manufacturer must use are not aligned according to 
 these economic functions but have developed along 
 trade and sectional lines as well as those of convenience 
 and individual initiative. 
 
 To get relatively wide distribution at minimum cost 
 he may require only the middlemen of the orthodox 
 system. To supply the largest possible percentage of 
 aroused demand without leakage and with reasonable 
 expense, however, he may be obliged to use several 
 agencies to reach different economic or geographic 
 groups of consumers. The sacrifice of an economical 
 method of demand creation may be necessary in order 
 to insure thorough physical distribution of the product. 
 Or, on the other hand, the upbuilding of an exclusive 
 organization of supply agencies may be required to 
 take advantage of the demand-creation plan on which 
 the success of the business depends. 
 
 It is the existence of these problems, the difficulty of 
 establishing a balance between the opposing require- 
 ments of demand creation and physical supply and of 
 maintaining this balance in the face of changing con- 
 ditions, that give the business man his function and 
 opportunity and justify his, at times, exaggerated 
 gains. In my view of things, the man who effects a 
 new and more economical combination of forces in 
 production or distribution is entitled to the savings he 
 effects until such time, at least, as competition has 
 learned to utilize or match his discovery. 
 
 Society must pay for its short cuts, either in market 
 rewards to individual innovators or in public appropri- 
 ations for the study and recommendation of the most 
 eflBcient processes by which it is fed and clothed and 
 
 279
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 housed. The latter would be the simplest, most direct, 
 and most economical method of arriving at results which 
 thousands of business men are attempting to approach 
 by roundabout, detail ways. This would not involve 
 the quenching of individual incentive and the initiative 
 which springs from it, by the socializing of our machin- 
 ery of production and distribution. Instead, by more 
 speedily standardizing the conditions, methods, mate- 
 rials, and processes on which all business men must 
 concentrate until they have been standardized, organ- 
 ized national investigation of these common factors of 
 business would so much the sooner release for creative 
 purposes the enormous funds of capital and human 
 energy now dissipated in unproductive motions. 
 
 From the consumer's standpoint, service is the key 
 to physical supply. A commodity is of dominant value 
 to him only when and where it will satisfy some material 
 or psychic want. Its appeal decreases in proportion to 
 its lack of time and place utility. The problem of the 
 producer is to anticipate the consumer's need, to bring 
 the product within easy reach at the moment the de- 
 mand develops and to keep on bringing fresh units to the 
 point of contact as long as the demand continues. The 
 location, construction, and equipment of the supply 
 plant, therefore, are essential considerations, governed in 
 the main by policies analogous to those which apply to 
 similar problems in production and demand creation. 
 In discussing the other policies, indeed, many of the 
 factors which count in physical supply were anticipated, 
 by reason of the mutual interdependence of the three 
 groups of activities. 
 
 It is enough, then, to suggest the main elements in 
 location. Accessibility to the market is obviously the 
 
 ^0
 
 PLANT AND OPERATION POLICIES 
 
 first of these, with emphasis on transportation facilities 
 and charges. If sale is by sample or by description, 
 the warehouses may be placed with an eye single to serv- 
 ice, remote from both sales offices and factory if speed 
 in delivery is thereby increased. This is one of the two 
 motives for branch warehouses at strategic points, the 
 other being the savings made in freights on carloads to 
 these market centers and package shipments in the trib- 
 utary zone. The advertising value of stock rooms 
 adjacent to the sales department is a negligible quan- 
 tity, now that the motor car has put the downtown 
 office within easy reach of the warehouse district. 
 
 Except in the case of machinery and other heavy 
 products where the customer may wish to inspect the 
 actual unit to be delivered, bulk sales by the manufac- 
 turer involve no special plant problems in physical 
 supply. The retail store in fact remains the one agency 
 in distribution where sales in bulk retain their impor- 
 tance. Even here, in the larger cities, the desire to secure 
 a selling location in the high-rent zone has separated 
 the salesroom from the stock room and made selling by 
 sample common in furniture, grocery, and department 
 stores. In groceries, perhaps, this advance is most typ- 
 ical of the new viewpoint and new spirit of analysis 
 that are making themselves felt in business. 
 
 Speed and economy of handling likewise dictate the 
 policies of construction and equipment. The character 
 of the goods, the building standards, the material re- 
 sources of the locality, and the amount of capital avail- 
 able are the factors which the manager has to consider. 
 In a business of any size, he has the technical knowledge 
 of his works executives to draw upon and can safely 
 leave to them the detail execution of the policies de- 
 
 281
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 termined with their aid. Lacking such counsel, the 
 head of a small business is still able to secure specific in- 
 formation on every phase of construction and equip- 
 ment from outside specialists or from authoritative 
 books and articles in trade and technical periodicals. 
 
 The small business has its compensations in that it is 
 usually compact, and handling operations can be kept 
 under the personal control of the manager. Its internal 
 transportation and storage problems are too simple to 
 require elaborate mechanical conveyors, tiering ma- 
 chines, electric "mules," overhead trolleys, and other 
 labor-saving devices to keep costs on a reasonable plane. 
 Its outside transportation is an equally simple matter. 
 If the product is bulky, a switch track serving both the 
 receiving and dispatching platforms and connecting with 
 the local railroads is the sum of requirements. For val- 
 uable small products, goods made of delicate materials, 
 or articles shipped in small lots, the advantages of a 
 switch track would be next to nothing. The problem of 
 outside transportation was considered at some length, 
 however, in an earlier section of this volume in the chap- 
 ter on factory location. It will suffice here to repeat 
 that the character and speed of the service required to 
 insure deliveries on the customer's terms is the main 
 factor, though the cost of providing for such service 
 balanced against the financial situation of the business 
 may postpone adoption of the most desirable policy. 
 
 In his choice of agencies for physical distribution, the 
 manufacturer encounters certain limitations. The na- 
 ture of the product, its unit price, and the buying habits 
 of consumers may indicate the middleman as the most 
 efiFective and economical means of supply. For staples, 
 this is a matter of course, since the work of demand 
 
 282
 
 PLANT AND OPERATION POLICIES 
 
 creation is largely in his hands; but this agent serves 
 also for branded foods and trade-marked specialties in 
 dry goods, women's wear, notions, hardware, furniture, 
 toilet and medicinal preparations, men's clothing and 
 furnishings, and many other lines of semi-staple neces- 
 sities and luxuries which lend themselves to this method 
 of physical supply, even when the middleman is not the 
 chief agent of demand creation. 
 
 At the same time, every one of these lines furnishes 
 numerous instances of physical distribution without the 
 aid of the middleman, the producer assuming all the 
 work of supply save actual transportation. The retail 
 catalog houses have blazed many trails in this par- 
 ticular field, but their experience would be an unsafe 
 guide to the manufacturer of a single product or a lim- 
 ited line. Despite their control of production sources, 
 the mail-order concerns must rank as middlemen, their 
 exercise of the important middleman function of assem- 
 bling, storing, and assorting many kinds of commodities 
 being an essential element in their success. 
 
 In analyzing his supply problem, the manager looks 
 first to the buying habits of his prospective customers 
 and the incentive he can ofiPer them to change, if a change 
 be necessary to give him maximum distribution at min- 
 imum expense. Does his product fall within the wide 
 range where the consumer puts convenience first, either 
 because the amount of money involved is small and 
 values are standardized or because the nature of the 
 product puts a premium on frequent small purchases 
 close at home? 
 
 It would be a doubtful experiment, for example, to 
 attempt to distribute a new brand of flaked cereal other- 
 wise than through retail stores. The custom of the aver- 
 
 283
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 age housewife is to order factory-cooked breakfast foods 
 from her grocer in small quantities to insure freshness 
 and original jflavor. To overcome this fixed habit and 
 the belief that cooked breakfast foods deteriorate with 
 age would probably be a much more difficult task than 
 to divert a good share of existing demand to a new brand 
 marketed through orthodox channels. The money in- 
 centive that could be offered for the direct purchase 
 of a dozen packages at one time, too, would be lessened 
 by transportation charges out of all proportion to the 
 value of the shipment. 
 
 Admitting that this is an extreme example of "con- 
 venience goods," there is nevertheless a great number 
 of commodities either low in price or of only occasional 
 use which must be placed within easy reach of the con- 
 sumer at the moment his demand takes form in order to 
 achieve adequate sales. The stock of any neighborhood 
 dry goods, hardware, drug, or grocery store will furnish 
 hundreds of such articles, many of them trade-marked 
 and highly differentiated from competing products. 
 Differentiation, trade-marking, and vigorous campaigns 
 of demand creation are responsible, indeed, for this wide 
 physical distribution and for the consumer's recollec- 
 tion of the trade name or brand when subconscious de- 
 mand for the article is turned into conscious demand 
 by his need of it or sight of it displayed in a store. 
 
 Accentuate this differentiation by closer adaptation 
 to the consumer's wants and keep up or increase the sell- 
 ing campaign, and the subconscious demand for time and 
 convenience goods can be transformed into conscious 
 or expressed demand to satisfy future wants or wants 
 suggested by the advertising as immediate or impending. 
 
 For products put forward in this manner, the retailer 
 
 284
 
 PLANT AND OPERATION POLICIES 
 
 and jobber with their waiting stocks are not necessary 
 agencies of physical supply; the parcel post or a rail- 
 road or express company will furnish the single missing 
 element of transportation. It is primarily a question of 
 demand creation. If your ideas about the goods have 
 sufficient force and appeal to dominant buying motives, 
 you can make your sale entirely by description and 
 pay a transportation middleman for delivering the 
 product. 
 
 This is the method of the catalog houses, whether 
 they specialize in one line like women's garments or 
 attempt to cover all lines of merchandise. Price is the 
 basis of their differentiation of products, the aim being 
 to offer each customer the kitchen apron or winter hat 
 which comes nearest to her notion of what she wants to 
 pay. Lowered prices and large assortments are, indeed, 
 their chief materials of demand creation. Adopting the 
 policy of selling at the market minus, they adjust their 
 production policies to their price scale, ownership or 
 control of sources making them in this respect manu- 
 facturers rather than merchants, and thus observe the 
 principle of balance and the principle of interdependence 
 which should govern in all the activities of production 
 and distribution. 
 
 An interesting sidelight on the changes taking place 
 in distribution is the successful effort of many manu- 
 facturers seeking maximum distribution in the larger 
 cities to convert their trade-marked specialties into 
 "convenience goods." Limited in their downtown 
 sales by the tendency of the big stores to push house 
 brands or competing products with wider profit mar- 
 gins, they have turned to the neighborhood stores as 
 supplementary outlets. Kayser gloves, Arrow collars, 
 
 285
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Community silverware, and a long array of advertised 
 products are recognized by consumers as the same, no 
 matter where they are purchased. Buying them close 
 at hand when they are needed is so much easier than 
 having to foresee and provide for future wants that the 
 time and place utilities thus created are superior to the 
 attraction of the larger assortments downtown. Not 
 always, to be sure, but often enough to make the neigh- 
 borhood store an increasingly important factor in sup- 
 plying urban demand. The tendency is accentuated by 
 the local retailer, who exploits the articles in the neigh- 
 borhood to prove that his stock is composed of standard 
 merchandise. 
 
 The producer who is tempted to break away from 
 orthodox channels of distribution must bear in mind 
 also that any scheme of direct supply involves problems 
 of financing, handling, and shipping of which the middle- 
 man system would practically relieve him. The elimina- 
 tion of the wholesaler alone, though it has been effected 
 in nearly every line of business by individuals who 
 deal direct with their retailers, causes a transfer of cap- 
 ital burden which might neutralize in a specific case all 
 the advantages of direct supply. The conditions and 
 problems set up in the packing and shipping rooms, in 
 the credit, collection, and accounting departments are 
 vastly more complex, of course, in a business having one 
 or two or three thousand dealer accounts than in one 
 which seeks its market through fifteen or thirty or fifty 
 wholesalers. 
 
 The manufacturer who deals direct usually is able to 
 place his line with one of the best retailers in each lo- 
 cality and to establish close relations with him. If his 
 product appeals to all classes, however, this sacrifices 
 
 286
 
 PLANT AND OPERATION POLICIES 
 
 not a little aroused demand which might be supplied 
 by dealers of lesser standing to whom the jobber, by 
 reason of his nearness, can safely extend credit and 
 dispatch orders, in combination shipments, which would 
 not repay the producer's handling and administration 
 costs. 
 
 In seasonal trades, again, manufacturing policies are 
 greatly simplified by distributing through the middle- 
 man. Instead of driving machines and men at excessive 
 speeds during the actual selling period, advance sales 
 to jobbers can be used to equalize production, thus 
 increasing eflBciency and reducing unit costs. When 
 the wholesaler is not able to finance the advance deliv- 
 eries himself, these can be made the basis of banking 
 credit, whereas the piling up of large quantities of unsold 
 stock at the factory would be counted by bankers a 
 somewhat speculative undertaking for any business not 
 long established and without abundant working capital. 
 
 One of the factors not to be overlooked is that the 
 middleman in a great many lines seems solidly estab- 
 lished. His adherents claim with considerable show of 
 authority that nearly ninety per cent of the output of 
 our small factories is distributed through the jobber 
 and that eighty per cent of the products turned out by 
 our larger concerns seek consumers through his hands. 
 Despite the success of the innovators and their short 
 cuts to the consumer, therefore, the individual manager's 
 supply problem does not hang solely on the elimination 
 of the wholesaler or of all middlemen. The profitable 
 approach is just as likely to be that of utilizing the mid- 
 dleman for physical distribution and cooperating with 
 him in the performance of his other function of demand 
 creation. It is not safe to ignore his control of the im- 
 
 287
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 mediate market for a multitude of commodities, or to 
 overlook the fact that, if he is not already distributing a 
 passable substitute for your product, development of 
 demand for it will surely bring about the addition of 
 such a substitute to his lines. By trying to eliminate 
 him, you divide your market and make necessary a more 
 intensive campaign of demand creation to overcome his 
 competition. 
 
 The direct agencies of physical supply hardly require 
 definition; their special fields and functions are obvious. 
 The branch house, the exclusive agent, and the manu- 
 facturer's chain store are all instruments to the same 
 end; they aim to control the supply function at every 
 stage and insure delivery of the goods to the ultimate 
 user in perfect condition and with the least possible loss 
 of time and convenience utility. The net result is min- 
 imum leakage of aroused demand. 
 
 Branch houses and chain stores are the concomitants 
 of sales volume. Either the aroused demand is so great 
 that the facilities of local middlemen are inadequate to 
 supply it; or the demand is of such character that it 
 may easily be diverted to other products by the middle- 
 man; or the volume of sales makes it profitable to take 
 over the supply function as an auxiliary business. 
 Whatever his motive, the manufacturer establishing 
 branch houses or chain stores must realize that in the 
 first case he is multiplying his administration problems 
 and in the latter is assuming the risks and attacking the 
 problems of a new and radically different range of busi- 
 ness activities. 
 
 The exclusive agent has more the character of partner- 
 middleman operating in a special territory, furnishing 
 part or all of the necessary working capital, and bound 
 
 288
 
 PLANT AND OPERATION POLICIES 
 
 up in his interests with the interests of the parent con- 
 cern. The warehousing concern, on the other hand, usu- 
 ally limits itself to the function of receiving and storing 
 goods at some strategic market center, and of assorting, 
 packing, and shipping orders in accordance with in- 
 structions from the producers or merchandisers it serves. 
 
 As in demand creation, the producer's subtler and 
 more important problems have to do with the organiza- 
 tion of his supply system — the choice of the agencies 
 to be used, assignment of the work to be done by each, 
 and coordination of their activities to cut out duplicate 
 efforts and to promote economy of stocks, reduce hand- 
 ling and transportation expense, and yet provide such 
 service as will conserve the maximum of aroused de- 
 mand. In our earlier discussions of the middleman's 
 part in demand creation, enough was said about his 
 function as a supply agency to indicate his capacities in 
 this direction and the situations and lines in which he 
 can be employed. 
 
 The average manufacturer takes the line of least re- 
 sistance or solves all his problem of supply at a stroke. 
 He chooses the middleman and accepts as necessary the 
 limitations which the choice puts on his business. Or 
 he swings to the other extreme and decides to market his 
 product by mail and make deliveries through the parcel 
 post or some other functional middleman of transporta- 
 tion. But the constructive business man breaks up his 
 supply problem into several unit problems and proceeds 
 to solve each on its merits, keeping an eye meanwhile 
 on the effect of the solution on his other activities. 
 Knowing the capacities of the middleman and the ad- 
 vantages and disadvantages of direct systems of supply, 
 he analyzes his market and chooses the agency which 
 
 289
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 will best serve each physical area and economic level 
 without incurring undue expense or interfering with 
 demand creation. 
 
 One well-known maker of men's and women's shoes, 
 for instance, distributes the bulk of his output through 
 a chain of retail stores maintained in the larger cities. In 
 lesser places, he avoids the internal competition between 
 lines in the usual shoe store by giving the agency to a 
 high-grade clothier who is willing to put in a shoe de- 
 partment, stocking full lines of the advertised men's lines 
 and taking special orders for women's shoes until de- 
 mand has developed. Local conditions may sometimes 
 point to a department store as the most promising dis- 
 tributor; or a shoe dealer's control of the town's best 
 trade and his desire to identify his store with a line so 
 well advertised may dictate his selection as the advan- 
 tageous agency. In territory served by neither chain 
 stores nor retail agents, mail orders are accepted and 
 the shoes are delivered by parcel post. Provision for 
 supplying aroused demand is thus made wherever the 
 manufacturer's advertising penetrates, the combination 
 of agencies depending on an analysis of local require- 
 ments or opportunities instead of a hard-and-fast policy 
 of dealer representation alone or of direct supply. 
 
 This typical case suggests the flexibility of plan es- 
 sential to effective distribution. Neither in demand 
 creation nor in physical supply does the wise manager 
 attempt to solve his problem as a whole and evolve a 
 standard policy applicable to all the economic and geo- 
 graphic divisions of his market. Instead, he recognizes 
 each group as a unit requiring individual analysis, call- 
 ing, perhaps, for a special combination of supply agen- 
 cies, if leakage of demand is to be avoided. In serving 
 
 290
 
 PLANT AND OPERATION POLICIES 
 
 a field relatively small as compared with the national 
 market, he will encounter widely varying conditions in 
 the larger centers and their suburbs, in cities having a 
 preponderant industry or class of industries, and in 
 towns and villages on which agricultural populations 
 depend for their merchandise needs. And even when 
 he has classified and grouped these district and local 
 markets, he will find case after case where the person- 
 ality, imagination, or selling ability of a single merchant 
 has upset all the conventions of trade and brought under 
 his control lines usually considered as having no affinity. 
 
 Initiative of this sort, based on searching analysis, 
 is responsible for the intensive distribution of to-day 
 which has multiplied supply sources in an effort to turn 
 highly developed specialties into convenience goods. 
 Putting dollar watches into hardware stores to reach 
 classes of prospects who never enter a jeweler's illus- 
 trates this tendency. Pickles, preserves, and marma- 
 lades are now displayed in the long unused windows of 
 butcher shops. Household remedies and disinfectants 
 can be had in many groceries. "Side lines" of special- 
 ties for the use of men are found in cigar stores, pool 
 rooms, and other places where they resort. The corner 
 drug store becomes a neighborhood trading center be- 
 cause it is open evenings, Sundays, and holidays, has 
 made convenience goods of a multitude of commodities 
 which appear also in the stocks of its hardware, dry 
 goods, candy, stationery, jewelry, and haberdashery 
 neighbors. 
 
 Up and down and across, the whole fabric of distri- 
 bution has been searched for points of approach to 
 groups of prospects inaccessible by traditional methods 
 and for new and more effective contacts with prospects 
 
 391
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 still unsold. Demand creation is transformed into a 
 problem of physical supply. The problem of offering 
 the commodity to the consumers at the time and in the 
 place where it is needed or is able to suggest its need 
 influences all the prior activities of the business. The 
 safety razor, for example, becomes a gold-plated birth- 
 day or Christmas gift and thus breaks into the jewelry 
 store where such gifts are bought. 
 
 392
 
 PART III 
 
 THE PROBLEMS OF ADMINISTRATION
 
 CHAPTER XVIII 
 PLANT AND OPERATION POLICIES 
 
 GOING back to our fundamental classification of 
 business activities, it will be recalled that these 
 activities divide into three main groups according to 
 their purposes. In production the object is to change the 
 form of the materials, in distribution to change their 
 place, and in administration to facilitate these other 
 changes. Despite the sharp cleavage of functions, it is 
 only in exceptional cases that a clean-cut division of 
 activities into three coordinate groups is found in 
 actual practice. 
 
 The "general" departments which concern them- 
 selves severally with finance, purchasing, employment, 
 credits, collections, accounting and auditing are fre- 
 quently recognized as a separate administrative group. 
 Seldom, however, is it realized that their distinctive 
 function of facilitating operations demands unit hand- 
 ling under a common executive, with authority corre- 
 sponding to that of the factory superintendent and 
 sales manager in their respective fields. Instead, the 
 departments report individually to the head of the busi- 
 ness; more often than not the work of one or more of 
 them is under his immediate supervision. 
 
 This retention of control is a survival from the time 
 when small-scale production allowed the manager to 
 guide all the operations of his business. The man who 
 
 295
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 owns and runs a neighborhood saw mill, for example, 
 performs all important functions himself, except perhaps 
 actual direction of the sawing. He buys and measures 
 his logs himself or takes orders for sawing on commis- 
 sion. Daily he indicates what timber is to be con- 
 verted into lumber and what sizes are to be produced. 
 He "waits" on customers and receives their cash in 
 payment or agrees to give them credit until they market 
 their wheat or corn or hay. He makes a memoran- 
 dum of the sale, carries a ledger account with the pur- 
 chaser, and either sends him statements until the bill 
 is paid or goes out collecting when he needs funds to 
 meet his pay roll or some other necessity. He "hires 
 and fires" helpers as occasion may demand. He secures 
 personal loans from the bank to tide over his busy sea- 
 son or makes shift to worry through on his own capital. 
 If he has a problem in office management, it is only to 
 decide whether or not he needs a girl to answer the 
 telephone and keep his few accounts. 
 
 As a business develops from this simple type, the 
 problems of administration, of facilitating and keeping 
 control of all important operations, grow in number and 
 significance. Because head sawyers and mill foremen 
 are relatively easy to find, the lumberman whose busi- 
 ness is expanding surrenders first his supervision of 
 manufacturing activities. Next, if the growth con- 
 tinues, he finds himself so occupied with buying, financ- 
 ing, credits, and the like that he turns the marketing 
 over to a sales manager. But even when his undertak- 
 ing has grown to the point where individual executives 
 look after most of the administrative functions, the 
 owner will probably continue personal buying of stump- 
 age and financing of new operations without consider- 
 
 296
 
 PLANT AND OPERATION POLICIES 
 
 ing how concentration on these activities forces him to 
 neglect others equally essential. 
 
 In a word, as the division of labor is carried further 
 and further in the activities of production and of distri- 
 bution, the control of operations becomes increasingly 
 complex. The pieces on the chessboard of business 
 are multiplied and the final result depends more and 
 more on the way in which each piece is manipulated to 
 further the general plan. The necessity of coordinating 
 and supervising the countless resulting details compels 
 the management to deal with paper representations of 
 the materials, motions, and relations involved instead 
 of directly with the things themselves. 
 
 The possibility of thus keeping in touch with the sig- 
 nificant activities of a business has the further effect of 
 enlarging organizations and the scale of operations. And 
 finally, the scope and complex character of business and 
 the distance of the controlling executives from the things 
 controlled put such a premium on efficiency in all these 
 facilitating activities, that the principle of the division 
 of labor is extended and the classification and organiza- 
 tion of functions follow as a matter of course. 
 
 The introduction of machinery for compiling and as- 
 sorting statistics, for wTiting reports, tabulating infor- 
 mation, and checking mental calculations indicates how 
 great is this increase in administrative activities. The 
 need, therefore, is imperative for an organized method 
 of bringing them all together, of establishing their rela- 
 tions, and of holding them at such distance from the 
 executive that no one of them will exercise undue influ- 
 ence on his decisions — in short, a method of keeping 
 them in balance with all the other activities of the 
 business. 
 
 297
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 The tardy discovery of a common purpose in their 
 functions is probably due to belated analysis. It is only 
 within the last twenty years, indeed, that they have 
 begun to take on real importance and acquire a sepa- 
 rate entity. In the small or medium-sized business the 
 usual practice is to associate some of them with the fac- 
 tory or sales organization and to retain others under the 
 eye of the executive. Only the larger concerns, facing 
 the urgent problem of keeping them all effective and un- 
 der control, have set them apart as a group of auxiliary 
 functions directly related with but impossible to clas- 
 sify as activities of either production or distribution. 
 The Taylor system puts extraordinary emphasis on 
 facilitating activities, particularly those which directly 
 affect production. In the average factory, however, a 
 sharp line is drawn between "productive" and "non- 
 productive" labor and effort is made to keep the latter 
 at a minimum. 
 
 Confronted with this multiplicity of motions and 
 operations, the manager, as well as the student of busi- 
 ness, finds that some scheme of classification is necessary 
 to make their relations clear. When we inquire their 
 purpose, their common character as the facilitating ac- 
 tivities of business comes out immediately. Pursuing 
 the analysis along the lines already made familiar, we 
 find that they break up into two groups concerned with 
 plant and operating functions. As before, the first 
 group subdivides according as the functions relate to 
 location, construction, or equipment of what might be 
 called the "oflSce plant." 
 
 Adopting again the method of approach used in ana- 
 lyzing the activities of production and distribution, the 
 manager finds that many of the policies governing this 
 
 298
 
 PLANT AND OPERATION POLICIES 
 
 first group of activities were solved when the corre- 
 sponding policies for the factory and sales units were 
 determined, because the mutual relations of adminis- 
 tration and the other two groups are even closer than 
 those between production and distribution. In plant 
 location, for instance, we saw in an earlier chapter that 
 nearness to sources of materials and supplies was a posi- 
 tive requisite in a great many industries. Automati- 
 cally, then, the location of the purchasing department 
 would be fixed by the same considerations. In no other 
 way can such effective coordination between manufac- 
 turing and purchasing policies be secured as when the 
 production heads and the buyer are in daily and friendly 
 contact. 
 
 Even when the important markets for materials are 
 distant, the sales departments of the supply concerns 
 are alert to anticipate the wants of customers and pros- 
 pects and to keep them informed of all changes in prices 
 and conditions. The isolated purchasing agent, it is 
 true, misses his full share of the occasional bargains in 
 staples which the emergency needs of supply houses pro- 
 vide for the man at the central market. He loses also 
 the benefit of contact and exchange of ideas with other 
 buyers both in trade gatherings and in the functional 
 associations which are doing so much to standardize 
 everyday practice. 
 
 By proper organization, however, he can unload the 
 purchase of all but his most important lines on assist- 
 ants and hold himself free for buying trips to the sources 
 of these latter before he makes his seasonal or annual 
 contracts. Meanwhile his presence at the factory allows 
 him to draw on the counsel and information of the en- 
 tire organization, from the manager's newest interpre- 
 
 299
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 tation of a general buying policy to the gossip picked up 
 by some department head whose knowledge of the mar- 
 ket is too slight for him to give it relevance and value. 
 
 But the purpose of this chapter is no more than to 
 analyze broadly the activities of administration and 
 illustrate the manner in which the suggested approach 
 can be applied to their solution. It is enough therefore 
 to point out that the essential factors in facilitation — 
 speed, accuracy, and minimum expense — are at their 
 best when the administrative departments are physi- 
 cally near to the making or selling departments which 
 they check and serve. Ability to have orders approved 
 by the credit department with the least possible delay 
 is an important aid to the sales force. Collection poli- 
 cies are likewise inseparable from future sales and from 
 the present finances of the business. Employment ac- 
 tivities must be responsive to the daily, almost hourly, 
 requirements of both factory and sales divisions. Ac- 
 counting, auditing, and statistical departments by the 
 character of their functions, must keep close on the 
 track of all the other units in the three major divi- 
 sions. In all these cases distance and elapsed time re- 
 duce efficiency and increase costs. Finance is the single 
 exception to the rule of advantageous nearness, and 
 this only when the amounts involved are too large to 
 be handled safely and legally by any but metropolitan 
 bankers or commercial-paper houses. 
 
 Sound business policy envisages many compromises, 
 however. WTien conditions prescribe a separation of 
 the manufacturing and marketing units, a corresponding 
 grouping of the administrative departments according 
 to the special affinity of each may be necessary, or 
 frank division of each into local organizations for pur- 
 
 SOO
 
 PLANT AND OPERATION POLICIES 
 
 chasing, employment, and those recording functions 
 common to both production and distribution activities. 
 Where branch houses are maintained, this division is 
 carried even further, though the "home office" usually 
 retains control over the district organization. 
 
 Construction and equipment policies in administra- 
 tion may be dismissed with a line. In the main, they 
 parallel those of distribution so closely that there is no 
 need of repeating that analysis here. 
 
 When we turn to the operating activities of adminis- 
 tration, we see that they also split up in the accustomed 
 way and range themselves under materials, labor, and 
 organization. But there is this important difference: 
 the materials of administration or facilitation — that 
 to which motion is applied — are the paper represen- 
 tations of the functions which are performed. These 
 paper representations are either causal (purchase or 
 employment requisitions or shipping requisitions) or 
 resultant (outgoing invoices, vouchers, and general and 
 department records) . 
 
 How this classification squares itself with current 
 business practice may be easily tested. Purchase requi- 
 sitions refer to plant and equipment or to materials and 
 supplies, while employment requisitions have to do 
 with labor. They originate in either the manufactur- 
 ing, the distributing, or the facilitating activities of the 
 business. The first group of functions centers in the 
 purchasing agent, and labor brings us back to the em- 
 ployment bureau or to the officials who do the hiring 
 for the production, sales, and office divisions. Besides 
 the factory or finished stock departments which supply 
 the products required to fill customers' orders, the ship- 
 ping requisitions directly involve the credit man, who 
 
 801
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 must approve the order before it is filled. On the re- 
 sultant side, the outgoing invoices are the affair of the 
 credit and collections departments. The vouchers rep- 
 resent the responsibilities of the financial end of the 
 business, while the accounting, auditing, and statistical 
 departments occupy themselves with the house records. 
 fi These various departments, or the executives and 
 employees who perform the facilitating functions with 
 which each is charged, are the agencies of administra- 
 tion. They correspond to labor in production, and to 
 middlemen, direct salesmen, and advertising in demand 
 creation. In considering their activities, the manager 
 finds himself engaged with three different sets of 
 problems. 
 
 The first group has to do with the broad policies in- 
 volved in determining the "what" and "how" of facili- 
 tation. These are his affair, since he alone is able to 
 decide what shall be the relations between the admin- 
 istrative department and the two other great divisions 
 of the business. Nor can anyone else establish the 
 proper balance between the speed, accuracy, and cost of 
 the facilitation service. The object, of course, is to 
 coordinate the work of each department with all the 
 remaining activities of the business and to direct its 
 internal activities only so far as is necessary to effect 
 this coordination. 
 
 These internal departmental activities constitute the 
 second group of administrative problems. They are 
 detail in character and grow out of the need of express- 
 ing the manager's broad policies in the facilitating 
 functions performed by each department and in its 
 external contacts with all other departments. They 
 are of lesser import to the manager for the reason that a 
 
 30^
 
 PLANT AND OPERATION POLICIES 
 
 competent department head usually assumes all respon- 
 sibility for them. Once the main policies to be observed 
 are laid down, he will be able, as a rule, to accommodate 
 his internal policies to the same purposes. 
 
 The third class of problems is concerned with organi- 
 zation of the mechanics of administrative effort, with 
 the smooth and effective working of the routine by 
 which the general policies of the manager and the detail 
 policies of the department heads are carried out. Per- 
 haps because so many of the operations involved are 
 the same or nearly the same in all the departments, the 
 value of organization and standardization, with a 
 common executive to look after the routine, has been 
 generally recognized and an office manager given 
 supervision over all but the special facilitating func- 
 tion of each department. This not only secures effi- 
 ciency of operation, but also relieves the manager of the 
 burden of settling minor disputes between executives 
 reporting directly to him. 
 
 The manager's chief problem, of course, is to reconcile 
 the frequently conflicting purposes of the administrative 
 departments and of his factory and sales divisions. This 
 is a task closely paralleling that of maintaining a uni- 
 form standard of cost, quality, and service in the 
 activities of production and distribution when opposi- 
 tion of interests occurs. 
 
 By way of illustration, the purchasing agent, to whom 
 price is a primary consideration, may insist on substi- 
 tuting for some specified material, tool, or supply requi- 
 sitioned by the factory another which the state of the 
 market allows him to buy at a bargain. Here either 
 executive may be right. The cheaper material may not 
 be fit for the use to which it will be subjected or the proc- 
 
 303
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 esses which it must undergo in manufacture, yet the 
 production head may not be capable of demonstrating 
 this fact to the buyer. On the other hand, the latter, 
 through wider acquaintance with market resources, is 
 often able to provide a better machine or material at 
 the same price or one of equal quality at a lower price, 
 but cannot persuade the factory executive to accept his 
 judgment. 
 
 Both are specialists, concentrating on their charac- 
 teristic functions, and with the specialist's inclination 
 to see the transaction from no other viewpoint than his 
 own. Friction naturally ensues unless the manager 
 steps in and fixes a policy that overlooks no buying ad- 
 vantages which can be conserved without sacrificing 
 factory eflBciency or quality in the product. The recur- 
 rence of such situations and the consequent necessity of 
 interpreting the same basic policies in terms of each 
 fresh dispute constitute strong reasons for the organi- 
 zation of the facilitating departments into a separate 
 division under its own executive, with authority co- 
 equal with that of the sales manager and the factory 
 head. 
 
 This, of course, is one solution of the manager's prob- 
 lem of organization, of keeping control of the operations 
 of administration without undertaking a load of detail 
 or abandoning his strategic position apart. Though the 
 general problem parallels that of organization in distri- 
 bution, the nature of the primary departmental func- 
 tions makes the matter of control more complicated. 
 In distribution — and in production, too, for that matter 
 — each activity leads up to the next in order, and all 
 focus on the final transfer of the goods to the consumer. 
 In administration each department is an independent 
 
 304
 
 PLANT AND OPERATION POLICIES 
 
 unit dealing direct, so far as its main function goes, 
 with the making or selling group. 
 
 Within the department the technique of this func- 
 tion is generally well understood. It remains for the 
 manager to define the policies which shall govern its 
 application, to determine how the department shall 
 divide or concentrate its operation, and to lay down 
 rules which will insure cooperation with all the other 
 units of the organization. 
 
 The character and scale of the business help to define 
 the scope and function of each department. As the 
 organization increases in size, these activities take on 
 added importance both because of the volume involved 
 and because there is less of personal contact between de- 
 partment heads. Hence it is that many concerns have 
 taken the initial step toward the ideal here suggested 
 and have transferred the departments handling them 
 from production or distribution to an administration 
 division. 
 
 Nothing more than a glance at these functions and 
 the policies which guide them is possible. The purchas- 
 ing department, for instance, has as its chief responsi- 
 bility to keep available at all times a suflScient supply of 
 the materials and equipment required by production. 
 The kind of materials, the qualities, the quantities to 
 be kept on hand, even the terms on which they may 
 be bought are all dictated, wholly or in part, by the 
 management's policies in production, distribution, and 
 finance. 
 
 Absolute standards may be set not only for the kind 
 and quantity of materials but for all classes of equip- 
 ment and supplies. Or within the limits fixed by the 
 general policies of the management on cost, quality, 
 
 305
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and service to the customer, the buyer may exercise 
 wide discretionary power in his choice of materials or 
 his substitution of one machine or supply for another. 
 He may be a mere clearing house for quotations or he 
 may be a powerful factor in production costs. 
 
 Sometimes this depends on the latitude allowed by the 
 manager, but oftener on the buyer's own departmental 
 policies, on the adequacy or inadequacy of his records, 
 on his knowledge of market shifts and of the basic con- 
 ditions behind them, on his acquaintance with technical 
 processes, his progressiveness in adapting new materials 
 to the uses of his plant, his information about the indi- 
 vidual situations of his suppliers, his ability to keep his 
 stocks in balance and to secure the low market price 
 without overbuying. The human factor also counts. 
 His attitude toward salesmen is reflected in their formal 
 dealings with him or in "good buys" and friendly 
 "leads" on sources in non-competing lines. His tact 
 and skill is shown in reconciling the standards fixed by 
 "the front oflBce" with the demands of department 
 heads. All these are matters of internal, departmental 
 policy, it is true, but none the less they are expressions 
 of the wise manager's buying policies. 
 
 So, also, with the employment head. Like the pur- 
 chasing agent, his first function is supply, to establish 
 and maintain touch with sources from which he can 
 draw all the various classes of labor required by the 
 production, sales, and administrative divisions, in both 
 executive and in clerical or workman types. Besides 
 this initial contact, he has in many organizations a con- 
 tinuous contact with such of the rank and file as are 
 susceptible of development through training courses and 
 through shifting to more responsible positions as their 
 
 306
 
 PLANT AND OPERATION POLICIES 
 
 capacities increase. In the average organization, he 
 surrenders the function of development to the depart- 
 ment executives, retaining only authority to transfer 
 men who are obviously misplaced and to find other 
 positions for men with whom departments are dis- 
 satisfied. 
 
 It is not within his province to designate how men 
 shall be managed or made more eflScient or paid. These 
 are policies which the manager of the business deter- 
 mines, as a rule, in direct consultation with the produc- 
 tion, distribution, and administration executives. It 
 is the practice in a great many organizations, however, 
 to give the employment head a degree of negative con- 
 trol over these relations between bosses and men. No 
 workman, for instance, can be discharged without a 
 review of the case and the reasons for his discharge by 
 the employment head. If injustice has been done him, 
 the fault is remedied, whether the routine or the indi- 
 vidual boss was to blame. Labor "turnover" is rec- 
 ognized as having a significant relation to production 
 eflSciency; one of the functions of the employment de- 
 partment is to hold it down and make the most of 
 available labor by transfers and "lay-offs" instead of 
 dismissals for extra hands. 
 
 As to the qualifications of the men to be hired, the 
 manager may well set up standards to be observed, 
 physical, mental, and moral. Quality men make for qual- 
 ity products and satisfactory relations with customers. 
 Here, again, the character of the business and the kind 
 of work to be done are imperative factors in the choice. 
 Above all, the wise manager will insist that the men to 
 be hired must harmonize with the prevailing type al- 
 ready in the business; otherwise, friction is sure to arise. 
 
 307
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Many organizations, indeed, make it a rule to fill all 
 positions of even minor responsibility by promotion from 
 within, sacrificing the occasional advantage to be had 
 by hiring competent executives or specialists from out- 
 side in order to secure harmony in operation, observance 
 of house policies, and the incentive to self-development 
 which a policy of promotion always supplies. It is in- 
 teresting to note, also, that in a business where every- 
 body is congenial and intensely interested in his work, 
 the outside man of different character or standards who 
 will not fall into line and adopt the organization spirit 
 is in many cases indirectly "fired" by the other em- 
 ployees. 
 
 It is difficult, in considering the various agencies of 
 administration, to indicate even the outstanding prob- 
 lems and policies, without exceeding the bounds set 
 for a single chapter in the book. The functions of the 
 credit department, in fact, alone would furnish mate- 
 rial for a chapter, so many and intimate are its rela- 
 tions with the activities of distribution and so important 
 its influence on finance. It is the primary office of the 
 credit man not only to keep selling safe, but to hold the 
 number of rejected accounts at a minimum consistent 
 with an average loss from bad debts agreed upon with 
 the management as a standard allowance. His function 
 is, indeed, not negative, but positive and causal, since 
 favorable action in every possible case is the aim and 
 favorable action results in the exchange of goods in the 
 finished stock room for the customer's promise to pay. 
 This promise becomes an asset in the business, its value 
 depending, first, on the customer's ability to pay, and 
 second, on the customer's willingness to pay promptly 
 when the obligation falls due. 
 
 308
 
 PLANT AND OPERATION POLICIES 
 
 Now the same customer's ability to pay and willing- 
 ness to pay promptly may vary widely on different credit 
 bases. Here the general policy of the manager comes 
 in as a defining element. This may be one of great lib- 
 erality, with terms and payments so arranged that the 
 chief concern of the credit man is with the honesty 
 and business ability of the buyer rather than with his 
 immediate assets. On the other hand, caution and 
 exact adherence to trade usages may be the rule. Be- 
 tween these extremes, any one of many combinations of 
 strict dealing and consideration for the customer may be 
 required by the nature of the product, the character of 
 customers, general or seasonal market conditions, and 
 the internal situation of the business itself. 
 
 To make the first policy practicable, for instance, 
 ample working capital, a wide margin of profit, a 
 strongly competitive selling situation, a limited number 
 of prospects, and salability or earning power on the 
 part of the product might all be necessary. A steam 
 shovel or motor truck might be sold to a young contrac- 
 tor on terms which would allow earnings to take care 
 of all payments after the initial deposit, if the buyer's 
 character were above question and his record indicated 
 experience in excavation or transportation work. The 
 product in this case would not be consumed and might 
 be reclaimed on default of payments. An inflexible 
 credit policy might be required when the profit margin is 
 small, the unit price low, a great number of customers 
 involved and capital none too abundant; or, again, 
 when consumer advertising has created such demand 
 for the product that dealer-customers have little 
 choice but to stock it and pay for it on the manufac- 
 turer's terms. 
 
 309
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 Departmental policies likewise reflect the ideals or 
 decisions of the manager. A friendly attitude toward 
 customers and prospects, for instance, tact in making 
 inquiries, in explaining the conditions on which credit 
 can be granted or the reasons why it must be refused, or 
 advice to buyers on the conduct of their businesses and 
 the elimination of practices and risks which limit their 
 buying power. Again, speed in passing on orders to 
 expedite service and aid the sales department, mainte- 
 nance of competent records and use of every available 
 source of information to determine the exact status of 
 each proffered account, including, perhaps, interchange 
 of ledger records with other firms, either direct or 
 through trade or local associations. And, finally, co- 
 ordination of credit policies with those of the sales de- 
 partment, to the end that no order shall be refused when 
 a plan can be devised for making its acceptance safe, 
 and no prospect or salesman shall be antagonized 
 through his failure to understand the reason for the 
 refusal of an order when such a course is necessary. 
 
 Between credits and collections the relations are so 
 close and the interdependence so absolute that both 
 functions are usually handled by a single department 
 until the volume of transactions makes their separation 
 advisable. When a credit account passes maturity, it 
 becomes a collection; but the policies which direct its 
 subsequent handling are much the same as in credits. 
 There is, however, this important difference: consid- 
 eration for the past-due customer stops short of that 
 shown the prospect seeking credit. In the latter case 
 only potential profits, balanced by an element of risk, 
 are involved; in the former, the assets of the seller are 
 directly at stake and his right to recover them is recog- 
 
 310
 
 PLANT AND OPERATION POLICIES 
 
 nized as paramount even by the debtor. The process of 
 "gentling" ceases, therefore, as soon as it becomes evi- 
 dent that some unexpected cause, like a crop failure, an 
 industrial disturbance, or some personal misfortune is 
 not to blame for the dehnquency. Recourse is had to 
 sterner methods, ending with an appeal to the law, if the 
 amount involved is large enough and the debtor is pos- 
 sessed of assets suflScient to satisfy the expected judg- 
 ment and costs. 
 
 The handling of past-due accounts, indeed, puts for- 
 ward problems as difficult and important as the grant- 
 ing of credits. To help the debtor to pay, rather than 
 to exact his pound of flesh, is the policy of the forward- 
 looking manager. At no stage of the negotiations, short 
 of the appeal to the courts, does he leave action to the 
 discretion of any outside agency; or, indeed, entirely 
 to the decision of his own collections head. 
 
 Of all the activities of administration, control of 
 finance is the last which the average manager surrenders. 
 The money end of the business has long been recognized 
 as his peculiar province, and it is only in very advanced 
 organizations that he has the courage and wisdom to 
 put it in its proper departmental place. His policies, of 
 course, must guide the treasurer or whatever official as- 
 sumes the function of providing and protecting the 
 necessary capital at the lowest possible cost and the 
 corollary function of maintaining the right balance be- 
 tween too much capital in dull periods and too little 
 in rush seasons. 
 
 The financial needs of a business are twofold: (1) 
 plant or fixed capital, which usually goes into the laud, 
 buildings, machinery, and other fixed assets necessary 
 to the business; and (2) operating capital, need for 
 
 311
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 which grows out of the activities of other administra- 
 tive divisions, purchasing, employment, and credits. 
 Fixed capital obviously is invested money; operating 
 capital is usually composed of invested money, loans 
 made from banks or private persons, and frequently 
 also undivided earnings. 
 
 After the first capital needs of his business are satis- 
 fied, the problem of the manager is to establish the most 
 efficient balance between the invested and borrowed 
 portions of his working capital. The latter must vary 
 with the seasonal rise and fall of sales and changing gen- 
 eral conditions which may require long credits to cus- 
 tomers one year and allow them to discount their bills 
 the next. It is recognized as safe and profitable policy 
 to let borrowed capital carry the peak loads provided 
 by the seasonal and extraordinary demands and to 
 maintain the invested portion at the point where it will 
 take care of ordinary needs and permit of at least an 
 annual "cleaning up" of loans. Banks usually require 
 this, indeed, except in the case of rapidly expanding 
 undertakings or concerns with an overwhelming pre- 
 ponderance of security in the way of fixed assets and 
 raw or finished stock. Not a few businesses, with re- 
 serves large enough alone to carry the peak loads, keep 
 these invested in marketable securities as an emergency 
 fund and make short-time loans to satisfy temporary 
 extra requirements. The larger the element of risk 
 involved, the higher the rate of interest this temporary 
 capital will command. 
 
 All this assumes, of course, adequate initial capital for 
 the undertaking. It is not too much to say that four in 
 every five American industries are launched on the 
 basis of hopeful selling possibilities rather than on that 
 
 312
 
 PLANT AND OPERATION POLICIES 
 
 of employing capital seeking an outlet. The manager's 
 big financial problem for many years is to secure money 
 to meet immediate requirements and to care for neces- 
 sary extensions at the factory and in the field. 
 
 It is not uncommon to find all the other activities of 
 the business subordinated to this one problem of keep- 
 ing a working balance in the treasury. The plant is 
 rented, power is taken from a service company, equip- 
 ment and materials are bought on terms rather than 
 price, castings are purchased outside to avoid invest- 
 ment in a foundry, the economies of lot production are 
 sacrificed to a made-to-order schedule, the products 
 turned out are those which can be sold in the shortest 
 time, heavy discounts are given for cash, all except the 
 indispensable functions of distribution and administra- 
 tion are eliminated or performed in some fashion by the 
 manager himself. The average small venture, indeed, 
 offers a striking demonstration of the interdependence 
 of all the activities of business and the influence of that 
 which seems most remote on the problem immediately 
 in hand. 
 
 Besides the function of securing capital, the manager 
 must lay down broad policies for the protection of that 
 which he is using. His long look ahead must include 
 the future needs of the business. Funds must be pro- 
 vided beforehand to meet contemplated expenditures 
 for extensions in production or sales, else working capital 
 will suffer. Insurance must cover factors of risk such as 
 fires, accidents, and customer failures, in order to safe- 
 guard capital and command the lowest market rates on 
 loans. 
 
 For the same reason, such a balance must be main- 
 tained between liabilities and quick assets as will sat- 
 
 313
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 isfy the banks that their loans are safe and a low rate of 
 interest justified. That other important element in 
 finance, credit with supply sources of materials, equip- 
 ment, and service must also be protected by prompt 
 payment of all accounts, since the resources thus made 
 available more than compensate for the obligations in- 
 volved. The governing policies of the manager neces- 
 sarily include the coordination of activities in credits, 
 purchasing, employment, and all other departments 
 whose conduct affects finance. The routine by which 
 these guiding policies are carried out may be left to the 
 treasurer or other official responsible for them. 
 
 It is hardly possible to speak definitely of managerial 
 policies toward the records of a business. These fall 
 into four groups each having a common function to 
 purpose. (1) Accounting has for its objects the more or 
 less mechanical recording of the operations of a business 
 in terms of a common unit and the matching of expendi- 
 tures for motions and materials with the results of those 
 motions and materials in their converted form. (2) 
 Auditing provides a system of checks and queries on 
 expenditures to make sure that they have been properly 
 recorded and charged to the department benefiting by 
 them and to guarantee that a fair equivalent has been 
 received for each dollar paid out, whether for materials 
 or the motions applied to them. (3) Cost-keeping in- 
 volves a classified analysis of expenditures for unit oper- 
 ations and groups of operations and a comparison of 
 such expenditures with the returns therefrom. The 
 form of this comparison shall emphasize variations from 
 the normal standard, since these demand the attention 
 of the manager. Occasional "sample" analyses may 
 suffice for minor processes; daily costs on the major or 
 
 314
 
 PLANT AND OPERATION POLICIES 
 
 governing operations are necessary to keep the manager 
 informed of significant changes. (4) Statistics consider 
 the records, cost of operations, and the commercial re- 
 turns from a prophetic rather than a historical point of 
 view. Their purpose is to forecast the future in the light 
 of the past, to trace the relation of past costs to possible 
 future costs and attainable results, and to fix standards 
 and quotas for unit and grouped operations, expendi- 
 tures, and incomes. 
 
 As constructive factors in administration, all these 
 records are developments of the last twenty years or less. 
 Even in the late '90s, it was the exception rather than 
 the rule to find a cost-keeping system which was more 
 than a collection of accurate figures on materials and 
 arbitrary guesses at the value of the labor and "over- 
 head" expense involved. Accounting had hardly been 
 carried past the stage where its records meant any- 
 thing but a stupid history of purchases and sales, of 
 outgo and income. I recall examining one set of pur- 
 chase books in which every item bought for nearly 
 fifteen years had been entered in chronological order: 
 item, one Remington typewriter; item, one car load 
 of pigs' stomachs; item, one hundred pounds of Bab- 
 bitt metal. But no attempt had ever been made to 
 classify purchases or to bring together materials or 
 supplies of the same character in order that the manage- 
 ment might know the relations between total purchases 
 of any kind and the total product resulting from them. 
 
 Even to-day it is common practice to maintain ac- 
 counting, cost, and statistical records which have no 
 logical place or purpose in the controlling scheme of the 
 business. Yet purpose here is the decisive test, as it is 
 in determining the value or worthlessness of any other 
 
 315
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 motion or operation in business. With such a test, 
 "What is the purpose of this record?" it is possible to 
 classify and to coordinate or combine all useful records 
 and to eliminate those which have no practical value. 
 
 Nine years ago, when cost accounting was on the first 
 crest of its vogue, the president of a large mid- western 
 company reduced his cost department from twenty-five 
 to five men by proposing this test of purpose for every 
 record that was being kept. Instead of a separate cost 
 ticket on each standard machine made, for instance, 
 sample costs were run on only one job in every fifty to 
 check and insure manufacturing efficiency, since no 
 other reason existed for cost-keeping in this particular 
 case. And so on through the whole top-heavy structure, 
 conceived in a specialist's enthusiasm for recording every 
 detail, whether it was useful or merely curious. 
 
 The most important thing in any system of records, 
 then, is that it should be a means, not an end. The 
 purpose being facilitation, no method or record should 
 be employed which does not actually forward the activi- 
 ties of the business. The records should fit the business, 
 not the business the records. They should grow out of 
 its operations and necessities rather than be superin- 
 posed from the top. At the same time, it must be borne 
 in mind that the scope and functions of records increase 
 so rapidly as a business expands that sooner or later the 
 point is reached where the recording functions must be 
 organized and coordinated if the manager is to retain 
 control of operations. 
 
 The strategic position for the manager, as we have 
 already seen, is one free from the routine of any depart- 
 ment yet in touch with the significant details of all. 
 From no other viewpoint can he secure a clear vision of 
 
 316
 
 PLANT AND OPERATION POLICIES 
 
 his business, protect himself from department bias, and 
 overcome a tendency to lay too much stress on the ac- 
 tivities which are most familiar or most interesting to 
 him. I do not mean that the head of a business must 
 withdraw himself from contact with all details in order 
 to keep his perspective. In innumerable small busi- 
 nesses the owner-manager is of necessity the sales 
 manager or factory superintendent, the advertising 
 manager or the purchasing agent, to say nothing of his 
 handling the finance and perhaps of credits and col- 
 lections. 
 
 It is to just such a man that the classification of busi- 
 ness activities here proposed should be most suggestive, 
 since it emphasizes as of equal importance with the 
 organization of production and distribution the coor- 
 dination of those facilitating activities which are fre- 
 quently administered haphazard because the conse- 
 quences of neglect are slower to appear. In the large 
 business also the management deals with details, but 
 only with symptomatic details. The popular concep- 
 tion of the head of a big enterprise is of a man engaged 
 in transactions involving millions of dollars. Yet in a 
 business of great size it may be a mere matter of rou- 
 tine to put a note for one hundred thousand dollars in 
 the bank, while the question of making a change in the 
 product which would add or subtract only a fraction of 
 a cent in cost may be a significant detail which the head 
 himself must handle because it might aft'ect some func- 
 tion of the product or some equality which commends it 
 to its public. The big business man, therefore, is con- 
 stantly concerned not with routine trifles but with 
 symptomatic details. 
 
 Acceptance and use of this classification does not 
 
 317
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 necessarily mean the formal partition of an organiza- 
 tion into production, distribution, and administration 
 divisions, though such a plan of organization in effect 
 is successfully employed by leading American indus- 
 tries. The purchasing agent may remain a factory 
 official; the experimental department may continue to 
 work under the sales manager; and so on. The impor- 
 tant thing is that the manager himself should recognize 
 that these activities are of interest not only to the 
 factory and the sales department but to the entire 
 organization. 
 
 Viewed thus, their oversight and direction are as es- 
 sentially the manager's concern as supervision of making 
 and selling operations. By placing them, as facilitat- 
 ing activities related to the whole business, on a par 
 with the activities of production and distribution, the 
 scheme of classification set forth in this volume gives 
 the manager the correct focus and a rational approach 
 to the problems arising from them. The first service of 
 any good classification, indeed, is to allow the man using 
 it to back away from the things with which he is imme- 
 diately occupied and to see all the activities involved in 
 their true relations and proportions. This strategic po- 
 sition, aloof from details (or at least recognizing them 
 as details), leaves him free to sense the broader prob- 
 lems which changing conditions and increasing social 
 control of private enterprises are proposing for his 
 solution. 
 
 318
 
 CONCLUSION
 
 CHAPTER XIX 
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 TF the vital relations of any business were charted, its 
 ■■• internal and external activities might be represented 
 as two circles impinging upon one another, with the man- 
 agement on the alert at the point of contact. The inter- 
 nal problems would arise out of the interrelations of the 
 activities of production and distribution and of what 
 we have been calling administration, for lack of a better 
 name. The external problems would have to do, first, 
 with the special public the business is concerned with, 
 its customers and prospective customers, its direct and 
 potential competitors, and the general body of labor from 
 which it draws its workers, executives, salesmen, clerks, 
 and factory operatives. And outside this circle, another 
 much larger might be traced to indicate the relations of 
 the business with the general public. Included in this 
 larger group also would be the individuals making up the 
 special public of the business and the members of its 
 organization. For the attitude even of employees is 
 affected by their social judgment of the business, its 
 methods, and its aims. 
 
 Unless he is blind or obstinate, the business man of 
 to-day must realize that he is no longer autocrat of a 
 private undertaking, able to base his policies on personal 
 whim or desire or an entirely selfish conception of busi- 
 ness. Just as society is asking of its college men what 
 return in community service they are making for the 
 
 321
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 exceptional training advantages accorded them, public 
 opinion is beginning to put the same query to business 
 men in sharper and more specific fashion. 
 
 "Here," society says, "you have been given oppor- 
 tunities and advantages in your business such as no 
 body of men at any other stage of the world's progress 
 ever enjoyed. You have raw materials, fuel, and sup- 
 plies that are cheap, abundant, and of excellent quality. 
 You have machinery of surpassing ingenuity and ca- 
 pacity, able to perform almost any task you set for it 
 at a cost amazingly slight. You have labor of excep- 
 tional intelligence to direct and supplement the func- 
 tions of these machines. You have a comprehensive 
 transportation system which allows you to locate where 
 land is low in price and buildings can be cheaply con- 
 structed, yet makes a great continent available as a 
 market for your product and a source for your supplies 
 of labor, materials, and equipment. 
 
 "Almost every necessary element and favorable con- 
 dition you could ask are provided for you. You have 
 only to supply the organizing ability which will pick out 
 the right product to make or sell, and choose the right 
 materials, the right equipment, the right location, the 
 right method of marketing. Over against your organ- 
 izing ability and capital, I set these contributions of 
 mine. What, besides the market minimum of value 
 and service at the price, are you going to supply as my 
 profit on my investment of community machinery and 
 opportunty.f^" 
 
 Without formulating its approach to the problem, the 
 public has been applying more and more inclusively the 
 basic test of purpose to the activities of business men. 
 It has come to feel that the large margin of profit 
 
 322
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 hitherto allowed permits them to practice many of the 
 mere "arts of commerce." It has sensed the existence 
 of useless motions, observed the needless duplication of 
 essential functions, and in some cases, as, for example, 
 the establishment of a government parcel post to carry 
 package freight in competition with the express com- 
 panies, it has sought a drastic short cut to a fairer 
 balance between service and cost. 
 
 In other instances it has adopted milder measures of 
 regulation and adjustment, as in the work of the Inter- 
 state Commerce Commission, the Federal Trade Com- 
 mission, the Federal Reserve Board, and the activities 
 of state and local boards for the supervision of trade 
 and industry. It is from this attitude of society, crys- 
 talizing after long agitation into definite policies of reg- 
 ulation, that arise what may be termed the external 
 problems of business. They are of growing importance, 
 of immediate importance, indeed; and no discussion of 
 business policies would be complete \\athout some effort 
 to analyze them and suggest how their solution may be 
 approached. 
 
 In dealing with both the general public and his 
 special public, the manager is not so much concerned 
 with the immediate activities observed as with the 
 actuating motives which they express. When he begins 
 studying the public and its relations with his own under- 
 taking, he finds that he must first take account of the 
 motives likely to influence the various social groups or 
 strata to action favorable or unfavorable to his pur- 
 poses. The comfort his employees enjoy in pleasant, 
 well-lighted, and well-ventilated workrooms, for in- 
 stance,Us likely to reflect itself in a community esteem 
 which j^will have a favorable influence on local sales 
 
 323
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 and on the attitude of that section of the public from 
 which he must draw his workers. A practical classifi- 
 cation of the public, then, might be based, first, on the 
 relations of the business with the various strata, and 
 next on the common motives which will permit group 
 handling and group control. 
 
 All this has to do with the business as an individual 
 enterprise, apart from all other concerns and responsible 
 only for its own actions. But the manufacturer or mer- 
 chant must reckon now with the public's changing 
 attitude toward all business, toward the particular 
 industrial or trade group to which his concern belongs, 
 toward the special form of organization which he has 
 adopted. For a long time the fullest measure of com- 
 petitive freedom for the individual was the rule, on the 
 theory that society was best served when every man's 
 initiative was given full play. As the possibilities of or- 
 ganization, coordination, and integration in the indus- 
 tries and trades became clear, the wisdom of an extreme 
 laissez faire policy began to be questioned. As a con- 
 sequence, the management to-day faces a whole series 
 of external problems which did not exist until these new 
 social forces were set in motion. 
 
 To solve the problems, he must understand the mo- 
 tives which control them and must be able to establish 
 their specific relations with his undertaking. The func- 
 tion of the classification proposed in this volume, in- 
 deed, is not only to allow him to see all the activities of 
 his business in proper perspective and to allocate his in- 
 ternal problems as they arise, but also to enable him to 
 value the external conditions affecting them and to ad- 
 just his internal policies before the possibility of recon- 
 ciling them has passed by.
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 In determining the policies which shall govern his 
 relations with the general public, the business man will 
 recognize three modifying influences — or three phases 
 of the same powerful influence — to which he must 
 accommodate his activities. Of these, the hardest to 
 understand, yet the real force to reckon with, is public 
 opinion, the reaction of contemporary thought or emo- 
 tion or ethical sense upon the activities of business and 
 the conditions under which they are carried on. 
 
 The law is the second influence. But law is simply 
 the crystallization of public opinion into a definite enact- 
 ment; while the government, the third influence, is no 
 more than an administrative agency for putting this 
 formulated public opinion into efiFect. 
 
 The wise man is the man who keeps abreast of public 
 opinion, who detects the changing viewpoint of the 
 country or the community and so modifies his individual 
 business practice that he later cooperates with rather 
 than opposes the operations of the law or the machinery 
 of the government. The average business man, how- 
 ever, is so submerged in the details of his undertaking 
 that he either fails to catch the drift of public opinion or 
 does not realize its vital bearing on his business. Not 
 until the law formulating this public opinion is enacted 
 or is about to be enacted does he awake to its purport. 
 Then he discovers that the people, through various civic 
 organizations, have been studying the underlying ques- 
 tion much longer than he had suspected. 
 
 Public opinion is the fundamental force, yet even now 
 business men are more intent apparently upon the in- 
 terpretation of the law and the attitude of the govern- 
 ment than upon developments in the public mind. 
 Watching all the legislative, judicial, and administrative 
 
 S25
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 machinery in motion, the manager of a business may 
 too readily conclude that his important external prob- 
 lems are concerned chiefly with it, not with smoking car 
 debates, the gossip of a switch shanty or the discussion 
 of a woman's club. But it is this talk and the convic- 
 tions which emerge from it that are the things he must 
 consider. 
 
 The passage of the Interstate Commerce Act in 1887, 
 for instance, was the first positive break in a long-con- 
 tinued policy of federal encouragement of private enter- 
 prise, particularly evident in the subsidies and land 
 grants to railroads and the protective tariff for the fos- 
 tering of industries. The Sherman Anti-Trust Law 
 followed in 1890, but a dozen years were to elapse before 
 court decisions, defining the jurisdiction of the Inter- 
 state Commerce Commission and the inclusive scope 
 of the Sherman law, emphasized the principle of control 
 underlying both enactments. 
 
 No better illustration could be cited of the power of 
 pubhc opinion to modify the established law as inter- 
 preted by the courts and administered by the govern- 
 ment. The Commerce Act as adopted expressed the 
 current feeling that discriminations in railroad rates for 
 or against individual shippers or communities consti- 
 tuted a menace to the business of the country. The 
 Sherman law was quite as definite in its attack on con- 
 tracts and combinations in restraint of trade. But 
 neither effected any radical change in the practices they 
 were aimed at until the courts and the national govern- 
 ment began to feel the pressure of a more thoroughly 
 aroused public sentiment a full decade later. Had the 
 railroads taken cognizance of this public opinion and 
 reshaped their rate and traffic policies in accord with it, 
 
 326
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 they might have been spared some of their troubles in 
 the last seven or eight years. 
 
 This pressm*e has continued and grown stronger. 
 Witness, among other effects, the decisions of the United 
 States Supreme Court dissolving the Northern Secur- 
 ities Company and the oil and tobacco combinations, 
 canceling the exclusive anthracite coal contracts, 
 affirming the right of the Interstate Commerce Com- 
 mission to fix general or zone rates on its own initiative, 
 and declaring the constitutionality of the long and short- 
 haul clause of the Commerce act. On the administrative 
 side, also, there are the defensive suit pending against 
 the International Harvester Company, which hinges on 
 the ability of that concern to throttle competition rather 
 than on any actual efforts to that end, the inquiry into 
 the internal efficiency and policies as well as the trade 
 conduct of the United States Steel Corporation, the 
 successful attacks on the methods of discouraging com- 
 petition employed by the "bathtub trust," and other 
 virtual monopolies maintained through control of basic 
 patents. 
 
 And, finally, the legislative program of the Sixty-third 
 Congress included the Clayton law, defining unfair 
 trading practices, and the act establishing the Federal 
 Trade Commission, the latter standing toward business 
 in general in much the same relation as the Interstate 
 Commerce Commission stands toward the railroads 
 and the shippers of the country. As this chapter is writ- 
 ten (November, 1915), the first important suit to enforce 
 the Clayton law has been begun at St. Louis, the sales 
 contracts and price-fixing methods of the United Shoe 
 Machinery Company being under fire as unlawful prac- 
 tices that lessen competition and promote monopoly. 
 
 327
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 This increasing concern of the federal and state gov- 
 ernments with business practices opens up a new vista 
 of external problems to the manager. They have to do 
 with four general groups of business relations now under 
 scrutiny or already under control : (1) the form and size 
 of organizations and their relations with other organi- 
 zations, outside the buying and selling of products or 
 services; (2) the relations of a business with its em- 
 ployees; (3) its relation with customers; and (4) the 
 responsibility of individual oflBcers for corporation acts. 
 
 Under the first heading come the suits for the disso- 
 lution of combinations like the American Tobacco, 
 Standard Oil, and Northern Securities companies, the 
 prohibition of interlocking directorates and of owner- 
 ship through single stock control of competing business, 
 the forbidding of pooling and price-fixing agreements 
 between normally competing concerns, and the limita- 
 tion of new security issues in the public utilities field 
 by the various state boards for the regulation of such 
 companies. 
 
 Government control of the relations of a business with 
 its employees has taken shape largely in welfare legis- 
 lation, like the workmen's liability and compensation 
 laws now in force in a majority of the states, in the 
 child-labor laws, in the laws limiting the hours of labor 
 for women, and in the fire and sanitary building codes 
 of the states and cities. The agitation for a minimum 
 wage law based not on the abihty of the worker to pro- 
 duce but on his living needs is evidence of a radical 
 trend in public opinion which will have to be taken into 
 account. Really an economic question, involving the 
 elimination of the unfit worker "at the margin," it is so 
 confused with human values that, as with so many 
 
 328
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 other business problems, it may have to be decided on 
 other than economic grounds. 
 
 The pure food law and the rate-making provisions of 
 the Interstate Commerce Act are typical of the stat- 
 utes directly affecting the relations of a business with 
 its customers. The supervision of price-fixing, which 
 is one of the main issues involved in the St. Louis 
 federal suit against the United Shoe Machinery Com- 
 pany, the right of manufacturer to fix the resale price 
 of his product, and the direct regulation of prices of 
 products or services where the monopoly factor or a 
 question of public policy enters in, are all questions 
 still in process of settlement but offering serious external 
 problems for the consideration of the business man. 
 
 How far the last function may go is illustrated by 
 the recent Oklahoma law empowering the state Corpora- 
 tion Commission to regulate the price of petroleum and 
 the volume of its production within the state. The Com- 
 mission is directed to fix the actual value of crude oil 
 by comparing the market prices of products with the 
 cost of refining plus a fair profit, to limit production 
 whenever the market price of crude oil fails to yield a 
 profit on this basic figure, and to allot to each well 
 owner his share of the current production allowed. In 
 a word, the cooperative action forbidden to the well 
 owners by the federal anti-trust laws is here assumed 
 outright by the state and the control of his production 
 and distribution activities is taken from the individual 
 producer. That the intention is to aid him and pre- 
 vent the demoralization of an industry does not decrease 
 the significance of the act. 
 
 In the same category of undefined hazards is also the 
 final attitude of the government and the law as to the 
 
 3^9
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 responsibility of individual officials for the acts of the 
 corporations which they control. On this point the law 
 certainly lags behind public opinion; in time it is likely 
 more closely to reflect public sentiment. 
 
 This trend toward government regulation of economic 
 activities is a matter of growing concern to every man- 
 ager. In the past, the average man has left the thresh- 
 ing out of proposed restrictive legislation or adminis- 
 trative orders to the larger commercial and industrial 
 units, on the theory that their interests in the matter 
 were so great that they could be trusted to represent 
 his interests also, or for the more specific reason that 
 usually he did not realize the importance of the question 
 until it was brought to his attention by the action of 
 these larger concerns. In the future, however, the atti- 
 tudes of the national and state governments toward 
 business are bound to have an increasing influence on 
 the conduct of every factory and store. 
 
 No longer can the head of a small business leave the 
 adjustment of his relations with the law, the govern- 
 ment, and public opinion to the grace of his big neigh- 
 bors or competitors or the zeal of the men who run 
 his trade association. He must cooperate with his trade 
 rivals and associates in conveying to the public in gen- 
 eral and to the men who make and execute the laws an 
 understanding of the activities, the relations, and the 
 necessities of business. Otherwise ill-conceived or doc- 
 trinaire legislation will hamper the operations of trade 
 and industry to such a degree that the burden of added 
 expense, under the law of competitive business, will 
 have to be passed along to the consumer. 
 
 Regulation must be intelligent; it must be at the 
 hands of men who know the technique of business, who 
 
 330
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 can distinguish between essential and non-essential 
 activities, and can further the first while they are elim- 
 inating the latter. In self-defense, therefore, as well 
 as for the sake of progress, every business man must 
 begin to consider the shaping of legislation and the 
 choice of administrative officials as external problems 
 of his business. 
 
 The slogan "good enough is the enemy of the best" 
 is written large upon the wall of one great American 
 factory. It expresses the idea that must be carried, 
 sooner or later, into every activity of business. Indi- 
 vidual initiative will effect this standardization, just as 
 it has accomplished the cheap refining of petroleum and 
 steel. But society through its legislative and admin- 
 istrative machinery could provide wholesale short cuts 
 to efficiency by organizing this questioning of methods, 
 materials, and results, by directing and financing an un- 
 ending search for the useless motion and extending it 
 to all the operations of business. 
 
 There has been no lack of public inquiries, recently 
 directed at various phases of trade and industry. They 
 effected little, however, because they dealt only with 
 phases, as in the recent government investigation of 
 price maintenance, for instance, and failed to consider 
 the problem in its broad relations with other problems. 
 This because the investigators lacked an approach to 
 the problems they were trying to solve, such a classi- 
 fication, perhaps, as we have been using here to deter- 
 mine the relations of business activities and to make 
 our problems understandable. 
 
 We have seen, too, how this classification embodied 
 the germ of efficiency by providing a test which will 
 differentiate necessary from useless motions. It might 
 
 331
 
 AN APPROACH TO BUSINESS PROBLEMS 
 
 also be employed in the government's analysis of trade 
 and industry as a whole, in order to determine for its 
 own use and for the individual business man's use what 
 motions are essential and what ones may be spared. 
 This, I think, is the big, undefined purpose behind our 
 unrelated public inquiries into business operations. 
 There is the feeling that waste and useless motions hold 
 the cost of doing business above the rightful level; 
 and investigation after investigation is launched in 
 response to public pressure to discover and remove the 
 cause. 
 
 Society, indeed, needs to know more about the activ- 
 ities of business, to realize their relations and interde- 
 pendences. Individually its members must understand 
 that if useless motions are to be tracked down and 
 eliminated in order to reduce costs, consumers must 
 cooperate. Frequently they demand so much service 
 that the business man, forced to comply, is obliged to 
 increase his mark-up on delivered cost to keep his margin 
 safe. Such service means extra motions, extra equip- 
 ment, extra labor. If a community or a class insists on 
 this extra service, it should understand that it must pay 
 for it. 
 
 When it exercises its collective function of law- 
 making, too, society must consider lest it lay on busiuess 
 a burden of costly reports, unnecessary building or labor 
 restrictions, inequitable taxes and the like which busi- 
 ness in turn must pass back to its members in their roles 
 of individual consumers. When a satisfactory level of 
 efficient operations is worked out for trade and industry, 
 it will come through the common effort of business and 
 society to eliminate the countless useless motions that 
 exist to-day. 
 
 332
 
 THE EXTERNAL PROBLEMS OF BUSINESS 
 
 Summed up, then, the business man has two distinct 
 groups of problems to consider. In approaching and 
 solving them he must keep always in mind the universal 
 application of the principles of balance and of interde- 
 pendence. He must remember that no question can be 
 dealt with in the light of departmental requirements 
 alone, or even as one affecting only production or dis- 
 tribution or administration policies. Instead, it must 
 be handled as a matter involving many, perhaps all, of 
 the other activities of his business. But more is neces- 
 sary. He must not only perceive these contacts and 
 establish right relations between activities within his 
 organization and within the circle of his special public; 
 he must also accommodate his policies and his practice 
 with the convictions, and at times the sentiments, of 
 the general public as finally expressed by the law and 
 the administrative acts of the government. In a word, 
 the external problems of business thus become internal 
 problems in the sense that no intra-organization policy 
 can safely be determined without taking into account 
 the attitude of society toward the activities involved. 
 
 333
 
 PRINTED AT THE HAHVARD TJNI\T;RSrrT PRESS 
 CAMBRIDGE, MASS., U.S.A.
 
 UNIVERSITY OF CALIFORNIA LIBRARY 
 
 Los Angeles 
 This book is DUE on the last date stamped below. 
 
 Form L9-25jn-7,'63(D8618s8)444
 
 UC SOUTHERN REGIONAL LIBRARY FACILITY 
 
 AA 001 018 443 
 
 
 "o<r
 
 
 iliiliilii