The Attitude of Management Toward Research ROBERT E. WILSON, Standard Oil Co. (Ind.), Chicago, 111. T r a i t s of a r e s e a r c h d i r e c t o r , v a r i o u s p h i l o s o p h i e s of h a n d l i n g research projects, a n d f u n c t i o n s o f all u n i t s from t h e top down are here tackled w i t h a view t o b e t t e r organization and r e t u r n s f r o m i n v e s t m e n t of t i m e , m o n e y , a n d p e o p l e J.Ν THE petroleum industry, and in most industries today, there is no question in management's mind as to the value of a n d the necessity for research. We have seen many examples of w h a t research can d o . Indeed, the present-day petroleum indus t r y is largely a demonstration of what r e search has done to improve products, i n crease yields, c u t costs, and increase t h e scale of operations. If we tried to evalu ate what research has done for our indus try, it would r u n into the billions of dol lars; and a t t h e same time research h a s done even more for t h e consumers of p e troleum products. We have also seen w h a t happens t o com panies t h a t do not carry on research. They gradually drop o u t of t h e picture b e cause they can not keep u p with their competitors' better products and lower prices—and this competitive process is t h e basic cause of t h e growth and vigor of American industry. Research
Growing
More
Difficult
Present-day management also has n o doubt as t o t h e need for expanded r e search facilities. We must have more a n d better organized expeditions into the u n known if we are to bring back results. T h e change t h a t has taken place during t h e p a s t 25 years of research is a good deal like t h a t which has taken place in geographical exploration during t h e past 200 years. T h e time has long passed when a Daniel Boone, with a rifle a n d a pack horse, could a d d greatly to our store of geographical knowledge. T o make significant additions today requires a well-organized Byrd expe dition, going t o extremely remote places with a plenitude of scientific equipment a n d observers. I n t h e past 25 years many lines of research have reached about t h e same point. T h e processes in our plants today have been worked over Κ iozens of good chemists, physicists, and engineers to see what could be done t o reduce costs. T h e efficient b u t costly equipment they have designed for us h a s increased our i n vestment so much t h a t it is difficult t o find a new process enough better than t h e old one t o justify scrapping the present equipment. Our fixed charges a r e a far larger p a r t of total costs t h a n they once
274
were, and operating charges are a smaller part. W e recognize, therefore, t h a t we must expand research if progress is to continue. T h a t means increasing the research facili ties, especially in view of t h e long dry spell during the war, when we could not do much in t h a t direction. There is, however, some question in management's mind as to whether t h e law of diminishing returns may not be starting to take hold. Re search costs can not continue to go up for ever as they have in the past. The costs have been justified to date—have been more than justified. B u t some time we are going to reach a place where we can not af ford to go further in expanding our re search organizations, because the rest of the operations will not be able to carry the load. Some d a y there will not be enough yield for each dollar expended in research. Today we are approaching t h e point where we m a y save three workers here and a few dollars there, b u t we have t o add a chemist in operations and a chemist in the labora tory to keep u p with things, and the result ing economy is not very great! So, re search must realize that i t faces a harder and harder j o b to justify its costs. I n some cases research laboratories seem to make a fetish of having the latest and fanciest equipment. When a new electron microscope comes out, n o matter how much it costs and no m a t t e r how little use there is for i t in a given laboratory, every laboratory seems to think it must have one. Before major investments are made in a p p a r a t u s , we should be pretty sure they are really going to fit into an in telligent research program directed to the company interests. Laboratories in a given region should cooperate in permit ting one another to use such special equipment. Research has long done a fine job in studying t h e costs of operations of all sorts, b u t t h e time has come when it must give some attention to cutting its own costs. Physician, heal thyself! Research today is big business and expensive busi ness. M o s t of m y thoughts on the question of t h e a t t i t u d e of management toward re search may b e covered under two general
heads: first, what h a s management t h e right to expect of a research director, and second, what does a research organization have the right to expect of management. What Is Expected of a Research Director? T h e first requirement is unquestionably that he be a good administrator. T h a t does not necessarily mean t h a t he m u s t be a brilliant scientist. While there is n o t any necessary exclusion between the two, certainly not many of our most brilliant scientists are good administrators and not many of our best administrators could be classed as brilliant scientists. We do, however, expect t h e research director t o make contributions t o t h e work. He should be a m a n who has come u p through the ranks a n d who knows how re search is done, what it takes t o make a good research man, and he m u s t appre ciate the brilliant scientist. T h a t brings up the problem of how to use t h e m a n who is a very able scientist, b u t who is either not a good administrator or not interested in administrative work. I t is i m p o r t a n t to have an organizational p a t t e r n and salary scale t h a t are flexible enough to pro vide such a m a n with an adequate reward in salary, even if he feels he works best b y himself or with only one or two extra pairs of hands. I remember very well my first industrial research experience. I was amazed t o find t h a t a m a n of D r . Langmuir's out standing ability did most of his work with one assistant, in a single laboratory. H e was, of course, available and widely used for consultation, b u t his outstanding contributions were made b y carrying out his experiments with his own hands. I t is well known how much he contributed, and what a great asset he h a s been to the General Electric Co. I t is necessary to have what might be called both a staff and a line organiza tion in research. Some men, as they grow older in experience and possibly a little less inventive, are most useful in consult ing staff positions, as associate directors, or something of t h a t kind. Other associate directors furnish the drive, and the d a y to-day direction of most of t h e research is p u t under them. T h e second requirement of a good re search director is t h a t he be fast on his feet. H e must be a m a n who realizes t h a t we are in a competitive struggle and t h a t it is up t o him t o keep us ahead of t h e other fellow. H e should see new trends sooner
CH EMICAL AND
ENGINEERING
NEWS
than management can. His j ob is t o h>e on the watch, to cull the literature, fco be on his toes in every respect. Being fiust on his feet also means that h e must ciat off lines of work t h a t have ceased t o b>e promising. There is nothing worse th-an a research director who starts a m a n to woirk on a certain problem, and then lets him g o on and on without a frequent re—evaluation of t h e promise of his work. O n e way w e avoid that is t o have a few m o r e problems t h a n men. I t is an effective w a y to k e e p manpower from being wasted. T h e appraisal ratio, applied to a g i v e n research project, is the product of tfcie probable value to t h e company of a s u c cessful result, multiplied by t h e probable chance of success, and divided b y tlie estimated cost of the research a r i d develoipment—not merely the researeli cost. AJU these factors involve estimates, but a g o o d research director is in t h e best position "to make these estimates. Obviously t h e ratio must fc>e substantially greater than u n i t y or t h e probl&xn should n o t be tackled. Sorae problems will arise t h a t are of such trenaendous p o tential value to t h e company that t h e y should be studied even though -the chances of success are relatively small. Conversely, there will be many problems tb^at are not of tremendous value, t>ut t h a t a,re justified because t h e chance o f success is large and t h e probable cost is small. J u s t w h a t the ratio should b e fora pn>t>lem to be undertaken is hard to say. A good deal depends, of course, o n the op-fciiuism of t h e man who makes up> the figures a n d on t h e comparison with "the possible other projects. If all of the problems in the laboratory are rated, it will be surprising how widely t h e ratios differ—indicating t h a t i n many cases too m u c h effort is being put on certain lines of w o r k a n d n_ot enough effort on other lines. T h e third requirement of a research director is t h a t he keep the company "business and potentialities in the forefront of his thinking a t all times. T o o ma,n3r r e search directors have private hobbies. T h e y may have a reputation "to maintain in a certain field, and they \va,nt to continue to publish papers in it. Such considerations must n o t enter i n t o the d i r e c tion of research. W e can not, i n industrial research, afford a lot of hobby-riding. T h e fourth and possibly t l x e most i m portant function of a research cLirector is t o form a courageous two-way channel b e tween management a n d the laboratory. I say "courageous" because i t takes g u t s to go up and tell the management when, it is making a mistake; and y e t managem e n t appreciates a m a n who ^will do thiat a n d keep hammering away u n t i l h e bias t h e m convinced, or else has decided t h e r e is no hope. B u t I do n o t thinJk it h a p p e n s very often, t h a t t h e director a n d the m a n agement can n o t see eye to e y e when tlxey make a real effort t o understand o n e a n other's viewpoint. Overcoming initial o b jections often takes courage, hiowever, a.nd V O L U M E
2 7,
NO.
5*
-
m a n y research directors seem t o be deficient in t h a t . T h e y growl a n d complain t h a t management has n o t done this or t h a t , but they really have n o t exerted themselves adequately t o get their points across. Equally i m p o r t a n t — a n d I t h i n k possibly more research directors fail in this respect—they should t a k e back the viewpoint of management a n d try t o sell it to t h e men in the laboratory. W e do not h a v e the t i m e to tell all t h e employees our problems, to explain w h y we can not do certain things, or why we must put more emphasis on certain problems and less on others. W e can and should tell t h e research director, and t r y to satisfy him. T h e research director should n o t go back a n d say, " T h e brass h a t s insist on so and s o , " without explaining t h e reasons. He is the man who m u s t carry the explanation d o w n to t h e fellows working o n the problem. A research worker can hardly be enthusiastic if he is working on something the importance or value of which h e does not understand. Fifth, t h e research director should be conscious of the personnel problems of the large modern laboratories. W h e n I first w e n t into a n industrial laboratory, where t h e r e was a small group, the entire staff t a l k e d over all t h e problems in the laborat o r y a t a single conference at least once a week. I realize now w h a t a fine, broad c h a n c e t h a t was to get a n understanding of t h e company business, in comparison with t h e restricted view research m e n h a v e in our large present-day laboratories. Professors tell me t h a t students, coming back after a year or two in large industrial laboratories, are generally rather discoura g e d . T h e y do not k n o w w h y t h e y are working on things, a n d they feel t h e y are m a k i n g slow progress. T h e y know only one little corner of t h e company's business. In addition, t h e y have gone from a beautiful campus to an industrial commun i t y with unsatisfactory housing conditions, and they are naturally discouraged. I sometimes think it is best never t o hire a man straight o u t of school b u t to let him g o t o another company for a year or so, w h e r e he can learn t h a t working for a living i s n o t the joyful thing his professors may h a v e painted it, and realize t h a t no working conditions are entirely satisfactory. B y t h e t i m e he gets to his second or third j o b , he appreciates some of these facts, b u t h e seldom does on his first. T h e research director, however, can do a great deal to ease t h a t situation. T h e n e w man comes in as a stranger in a big organization. H e needs personal friends a n d social contacts. T h e research direct o r can n o longer do a s we used to do— h a v e parties or picnics t h a t include the e n t i r e laboratory. B u t every supervisor w h o has a group of m e n working for him should assume t h e responsibility for gett i n g the new m a n properly introduced, m a k i n g some social contacts, having some activities—bowling t e a m s or baseball
JANUARY
3 1,
1949
teams, or something of t h a t kind—so t h a t the new men can feel they h a v e a chance t o make friends. When a new m a n is hired, it would be well to send everyone in t h e laboratory a brief s t a t e m e n t of wrho he is, where he is from, w h a t schools h e went t o , and on what he is going to work. There are sure to be two or three people in the laboratory who came from his school or his vicinity. They should be encouraged to look h i m up. Otherwise, in some of our larger laboratories, they might n o t realize for 5 or 10 years t h e y had a fellow alumnus in t h e organization. There must b e a consciousness, n o t only in t h e research director b u t down through the entire staff, of t h e importance of trying t o make the newcomer feel at home. Obligations of Management to Research So m u c h for w h a t we expect from our research directors. Now for t h e obligations of management to research. In t h e first place, it is o u r No. 1 job t o furnish good research direction. M a n a g e m e n t does not find t h a t easy; it is a difficult job. It is highly desirable, of course, to develop a good research director within your own organization. If t h a t can be done, it is a little easier for the m a n to sell himself. He has been sized up a n d a p praised, and if he h a s risen to t h e top in a competitive struggle he is p r e t t y sure to be good. Sometimes, however, i t is n o t feasible t o select a director from within t h e organization. In picking an outsider, " great care must be exercised. I am reminded of one of my first consulting jobs with a company in N e w York, when I was working in the research laboratory a t Massachusetts I n s t i t u t e of Technology. I found this company was making a product t h a t involved a lot of technical problems. I said, " H o w can you people possibly get along without a chemist?" They said, "Well, we have h a d very unsatisfactory experience with chemists." " H o w many have you h a d ? " "We have h a d three, a n d none of them worked out well." "Well," I skid, " T h a t is certainty a surprise to me. I do n o t see how you can get along without one. How did you select these chemists?" " O h , " they said, "we advertised in t h e New York Times.13 Second, the research m e n are entitled t o a d e q u a t e compensation. During t h e war we were particularly hampered in bringing u p research salaries to proper levels, b e cause they were under government control. T h e wage earners, represented by unions, could g e t approval for almost a n y t h i n g ; b u t it w a s very difficult to get approval for a d e q u a t e salary changes. However, t h a t situation has been largely remedied. Salaries h a v e been going up rapidly, a n d I do not think most 'managements begrudge this when they understand t h e problem. Third, the research men are entitled to a d e q u a t e facilities and good working con275
ditions. That does not mean that they should be set off in a park with a country club around them. I believe a laboratory should be close to the plant, partly for the effect of plant contacts on the laboratory men, and partly for the effect of laboratory contacts on the plant men. It is a great mistake, in my opinion, to isolate most of the research work from the plant, although there may be a small separate research laboratory for very fundamental problems. Fourth, management should be expected to supply "patient" money, as John Teeple picturesquely described it. Management must not be too impatient for quick returns. Beware of the company that has just recently "got religion," or that has some new vice president who is anxious to make a record and thinks research will give a quick answer to all the problems. I have seen a half-dozen cases where a big research organization was built up, with more men than problems, and miracles were expected within the first two or three years—and then the whole organization broke up. A steady, slow growth is needed to ensure a stable organization. I t is much like starting out today to do wildcatting in petroleum. For two, four, or five years, a company would have a hard time getting into the black, because it would not have any backlog of discoveries made five and ten years ago to help carry expenses. It is the same way with research. Usually the results of research do not make themselves felt in the first two or three years. But that background of research may crop out and be of tremendous value several yeard laterFifth, the research director should be given a reasonably free hand. My own philosophy of handling research has been to give a good research director a stack of chips and tell him to get into the game and do his best. It is a gambling proposition. I do not think management can contribute a great deal to telling him which hand to back, though it should keep him advised of the facts he needs regarding the economics of the situation, and the plans of the company. I do not favor the budgeting of individual projects—that is, saying at the beginning of the year, "We are going to spend so much on such and such a problem and no more." For one thing, it keeps a research director from being fast on his feet. I t keeps him from shifting men from one problem to another. Also, it is almost bound to cut down the amount of fundamental work he does. A good research director can be counted on to keep a balance of fundamental work going on if he does not have to budget his problems, because he knows that without fundamental work the laboratory results will tend to peter out. H e is much more likely to keep a proper balance if he does not have to account on each problem. In addition—and most important—it is an awful accounting headache all the way along the line. Why 276
spend so much money to find out exactly what this, that, and the other problem costs when you can not accurately appraise its value in any case? More time and effort is wasted on trying to keep accurate account by problems in a research laboratory than almost any other field. I do not quite know how our laboratories at the Standard Oil Co. (Ind.) escaped such control because we have had a very close accounting control of everything. When I was research director, I had to argue the point two or three times. However, we have escaped it; and we are all happy that we have done so. Of course we do budget our expenses in the three main categories of research, technical service, and control.
PLUTONIUM If this world isn't going to pot As quickly as you think it ot, Just leave its problems to plutonium. {And, brother, that is no balonium. ) R.T.S.
Sixth, management should build up the organization in such a way t h a t research workers are relieved as far as possible of routine work. It is a disgrace that some laboratories do not have enough dishwashers and errand boys, and use research men for such work. It is a waste of technical manpower. Fortunately, with research workers so scarce^ this viewpoint is becoming more and more appreciated. Seventh, relieving research workers of routine applies also to patent applications. Most research workers are, understandably, not particularly interested in the patent aspects of their work. Patents are not the most important thing; but if a company is engaged in industrial research, it must build up a reasonable patent picture to protect its own operations, aside from any possible royalties. There should be a separate well-staffed organization to do all the searching, to do practically everything except write the original notes of what is discovered. In our case, we even found it desirable to have a small section of laboratory men who do what we call patent research. Some man may have made some unusual observation that is not right along his line of work. If we insist he follow it up, he may be annoyed, and say, "I haven't time to bother with that type of thing." But with a group of men interested in all possibilities and available to do brief experimental work, research workers are much more willing to bring to the company's attention tilings that may turn out to be important to it in fields remote from the individual's line of research. Eighth, management should be ready to give prompt and courageous attention to worth-while ideas. There is nothing so devastating in a research organization as to develop a process or product and have C H E M I C A L
it turned down or given the glassy stare— or to have it lie in a report without notice for four or five years—until some competitor comes out and makes a great splurge with the same idea. Often management, if it does not have a good memory, will call up and say, "Why in blazes didn't you boys get that idea?" Of course, in such cases, that is just what every research director hopes mane cement will do. Management should be more willing to back the horses developed by its own laboratory, rather than something developed outside. Too many managements are carried away with the glamor of things that some outsider describes in glowing terms, while they hesitate to support the developments of their own laboratories which are presented with a more honest appraisal of pros and cons. Sometimes the real blame for this situation lies on the research directors, who have not done as effective a job of selling as they might ; but that is not the only reason. Sometimes there is the attitude that, "Well, I knew John when he was just a lab boy, and I don't believe that can be such a wonderful idea." But if somebody across the street brings it out, then they are ail hot and bothered about it immediately. We have all found that even the best research organizations will make occasional mistakes and will bring in some sour ideas; but if real money is put into a process or equipment, although it may not work in just the way it was supposed to, every research man is on his toes to see that an adequate return is obtained, and it usually is. Having research men know that management is ready to back their good propositions promptly puts them on their mettle. I t increases their enthusiasm, and it makes the research men a little more careful about the things they propose if they realize their proposals will be taken at full face value. Ninth, there is a responsibility on the part of management not to divert too many men from research t o sales technical service and manufacturing technical service. Research directors should not resent the requests that come in, as some directors seem to do. Such requests show that the company is becoming conscious of what technical men can contribute. But when a man who is doing a good job of manufacturing or selling gets interested in what the laboratory can do and sees some examples, he is likely to become more and more demanding. There has to be a limit somewhere. We have generally found it desirable t o set up a special staff of men for technical service work. The man who directs such work is free to say whether these men shall work on this, that, or the other problem; but he is not free to take men away from fundamental research. For this reason I consider it undesirable to have a research department report to the vice president in charge of either manuAND
ENGINEERING
NEWS
facturing or sales. I t usually turns out, in the course of time, that if this vice presi dent is doing the job as he should from his particular viewpoint, most of the men are likely to be diverted to sales technical serv ice or to manufacturing technical service, as the case may be. Or perhaps the vice president is trying to double as a director of research, which he should not do. He has other responsibilities. Tenth, it is the obligation of manage ment to have an ever-open ear. This does not mean it has to be open all afternoon to the research director; but, if he is ready to make a concise and factual presentation, management owes it to him to see him and to give him the time he really needs. I think management is coming more and more to that viewpoint. Sometimes a research director may think he is unfortu nate in working for a man who is very busy, but he may find that after all he is better off if the boss does not have time to tr}' to run his job. Management should always make it a point to attend important research confer ences, or have some representative there and give a brief talk, so that the younger men will know that management is inter ested in research. Also, when an impor tant new development comes up for deci sion, the research directors should bring up the men who worked on the problem and give them a chance to present their case, or at least be present when it is talked over. Eleventh, management owes the re search worker, even those down the line, a reasonable opportunity for technical con tacts. Of course, there has to be some limit. A company can not send half its re search staff to a chemical society meeting; but when the meeting is in the neighbor hood and there is not much expense in volved, a fairly large number of the men can be released for a day or two—reducing the number progressively, of course, as the meetings get farther and farther away from home. However, I am. always suspi cious of the research workers who do not bother to attend local section meetings and then have a great urge to attend a na tional convention a long way off. I do not think it is wise to allow research to be done for the purpose of presenting papers. That can become a never-ending proposition, with people more interested in getting a national reputation on some particular subject than in doing something useful. When a man has done a job for a company, however, and wants to present it to a technical society, he should be allowed to do so—after adequate patent protection has been obtained. Research men should have that right and should be encouraged to use it, even though it may take a cer tain amount of company time to write up the results in suitable form for a paper. Such presentations do have substantial indirect value to a company, which gets a reputation for being willing to have its men present papers and for being progres sive. I t is not just a matter of humoring V O L U M E
2 7,
NO.
5
»
.
the man, but can usually be justified from the company viewpoint. The men in the research organization, when they are interested and willing to do it, should be given a reasonable amount of time co take an active part in either local or national professional society activities. I know this participation helps to keep pro fessional men more contented, and it also helps to make the societies better informed and provided with abler officers and direc tors. Finally, management owes it to research not to swing back and forth from one ex treme to the other, depending on how the stock market behaved in the previous few months. We all have seen instances in
which companies have built up big labora tories in boom times, and then in poorer times have liquidated them. This does not mean that the research laboratory can expect to be exempt from everything that happens in times of curtailment, but we should look at research as a long-pull prop osition. Frequently research is more needed in depression than in boom times. While management and research di rectors are bound to view some things dif ferently, a sincere effort on the part of each to understand the other's viewpoint will help them both and greatly benefit the whole enterprise. T H E above article is based on a talk given at a meeting of the Industrial Research Institute.
Southern Student Affiliates Meet A STAFF REPORT Ν SPITE of the fact that many schools Iwere holding semester examinations and their students were unable to attend, what is believed to be the first all-Student Af filiate convention covering a wide area drew an attendance of 103 students from 12 colleges and universities in the South and Southwest to New Orleans, Jan. 14 and 15. Conceived and sponsored by the Loyola University Chapter of Student Affiliates, the meeting was held in the Roosevelt Hotel. Local representation was from Tulane University, Loyola, Ursuline College, and St. Mary's Domini can College, and half the registrants were "out-of-towners" from the University of Florida, Sam Houston State Teachers College a t Huntsville, Tex., Birmingham Southern, Auburn, Texas A&M, and St. Mary's University a t San Antonio. During the day and a half technical pro gram, presided over by Charles Jarreau of the Loyola chapter, there were 10 papers by student affiliates and two invited faculty guest speakers. J. E. Muldrey of Loyola spoke on "Frontiers of Chemistry" and Hans Jonassen of Tulane on "Gradu ate Training in Chemistry in the South." The first student paper, on fluidized catalysts, was delivered with demonstra tions by Robert Longmire of Tulane. R. D. Wales, Texas A&M, in discussing the effects of pH and dissolved oxygen vari ation on oxidation-reduction potential^ of marine waters, showed that the redox po tential is decreased by an increase in pH and increased in logarithmic proportion by an increase in dissolved oxygen. Kenneth Wilson, University of Florida, reviewed the peacetime applications of atomic energy, and a history of the alums and their preparation was given by Joan Glynn of Ursuline College. Proving the value of experience which may be gained in summer employment, Edward H. Gause of Loyola described his development of a potentiometric titration of sulfuric acid in the presence of alum to an empirical
J A N U A R Y
3 1,
1949
end point so that the titration can be com pleted without the precipitation of alum interfering with the analysis. Highlight of the evening was the mixer, during which Paul F . Bailey, moderator of the Loyola chapter, prepared the famous New Orleans beverage "café brûlot." Review papers given on Saturday morning were by John Yardley of Birmingham Southern on "Cesium, the Wonder Metal," Louise Oser of Dominican College on "Acetylene in Industry," and R. Fritz of St. Mary's on magnesium production from sea water. M. M. Duncan, Auburn, concluded the technical session with a discussion of the various types of chemical bonding. Saturday afternoon was given over to two plant trips. At the banquet that evening Sewell Oertling, chief technologist at Shell's Norco refinery, spoke on research in the oil industry and opportunities for chemists in that field. Chairman Charles Jarreau of Loyola pours cufé brûlot at the mixer
277
