Read an Excerpt
By John P. Kotter
Free Press Copyright © 1986 John P. Kotter
All right reserved.
The General Management Jobs: Key Challenges and Dilemmas
It is well recognized in a general sense that the world of the typical manager has changed considerably during this century as organizations have grown larger, more diverse, more geographically dispersed, more technologically sophisticated, and the like. But I wonder if we really appreciate in a more specific sense how these trends, which continue today, affect the nature of managerial work.
With respect to general-management jobs, these trends seem to have made most of these jobs extremely demanding, difficult, and complex in both an intellectual and an interpersonal sense. These jobs today put a person in a position where he is held responsible for a complex system which he cannot directly control and cannot entirely understand. They demand that he identify problems and solutions in an environment where behavior-results linkages are unclear, that he cope with the fact that thousands of diverse issues and problems could absorb his time and attention, that he balance the short and the long run despite pressures to ignore the latter, that he somehow motivate good performance and deal with bad performance on the part of large numbers of subordinates, that he keep a very diverse group of people working together harmoniously and effectively, and that he get a lot of other busy people over whom he has no formal authority always to cooperatewith him.
These same trends have also helped create more kinds of GM jobs, and have made the key demands associated with those jobs less and less similar in different contexts. As a result, two general-management jobs today can be very different in terms of the key tasks involved, and thus in terms of the demands they make of the general manager. Even GM jobs that look very much alike on the surface can present the incumbent managers with a very different set of challenges and dilemmas.
In this chapter we will examine in some detail the basic nature of the demands associated with all the general-management jobs in this study. Further, we will explore how and why these demands can be different in different settings.
The Job, the Context, and the Emergent Demands
Like most "Jobs" in modern organizations, the GM jobs in this study tended to be defined, sometimes formally and sometimes not, in terms of a set of responsibilities and a set of relationships. Specifically, the responsibilities and relationships associated with these jobs were:
1. Long run -- for setting some or all of the basic goals, directions, and priorities for an organization, including deciding what business or businesses to be in, and how to secure key resources.
2. Medium run -- for deciding how to allocate resources effectively to that business or those businesses so as to achieve long-run goals.
3. Short run -- for the efficient use of the human, financial, and material resources employed in that business or those businesses, including some profit responsibility.
1. Up -- reporting to a GM boss (or a board of directors).
2. Lateral -- sometimes (but not always) having to rely on other internal groups for support (e.g., corporate staff) or having to coordinate groups that are associated with the business but do not report to this GM position.
3. Down -- authority over what is usually a very diverse set of subordinates (not just specialists in a single function).
As the above suggests with such words as "some," "usually," and "sometimes," there was variety in how these jobs were defined, and we will explore that variation later in this chapter. But despite that variety, this statement of responsibilities and relationships basically describes all the GM jobs in this study.
These jobs were located within broader business and organizational contexts that were almost always quite complex, owing to factors such as business uncertainty and the large number of people involved. As a result, each of the job responsibilities and relationships tended to be magnified and shaped into important and difficult sets of demands, challenges, and dilemmas.
Job Demands I: Challenges and Dilemmas Associated with the Responsibilities
1. Key Problem/Challenge #1: Setting basic goals, policies, and strategies despite great uncertainties. In the typical GM job in this study, the long-run task was fraught with great uncertainty. The number of factors relevant to this type of strategic decision making was generally enormous. Knowledge of how those factors interacted was generally very limited. And tools for forecasting those factors into the future were generally crude. Yet despite all this uncertainty, the GM job was usually charged with overall responsibility for making the long-run decisions for some organization.
Dan Donahue, for example, had been in the process of reexamining and adjusting the basic direction of his organization when I first met him. This reexamination occurred because his division (which he had just recently joined) had been losing money. The reexamination had proved to be a very difficult task for two major reasons. First, Dan lacked clear information regarding the past and present state of affairs in his division and in that industry. He could not clearly identify what, if any, distinctive competence and comparative advantages his division had over companies with whom they competed. Different individuals in the company had varying opinions, none of which could be objectively verified with available information. Second, forecasting future opportunities and risks was hampered by dozens of important unknowns. Even the most sophisticated information gathering, analysis, and forecasting could offer only vague guesses to such questions as:
* Will there be any breakthroughs in the next decade in the two or three technologies they used most often? If yes, how will these affect product design and manufacturing economics?
* How will changes in the demographics of the labor force, in family patterns, and in disposable income affect consumer demand for their products? What impact will inflation have on consumer demand? How bad will the inflation be?
* Will any major new competitors enter their industry in the next decade? If yes, who are they likely to be (foreign or domestic) and where will they try to position themselves?
* Who is likely to win the U.S. elections in 1980 and 1984? What effect might that have on the regulation of their industry?
* What will probably be happening in the parent corporation over the next decade? How will that affect their inclination to provide resources to this division?
Although the long-run decisions that Donahue faced were extremely complex and the uncertainties very large, his situation was not at all atypical in this study. Indeed, at least half the other GMs had to deal with a long-run task which seemed to be as or more complex and uncertain. Furthermore, all the evidence of which I am aware suggests that the same is true in general for these kinds of jobs in corporations today.
2. Key Problem/Challenge #2: Achieving a delicate balance in the allocation of scarce resources among a diverse set of functional and business needs. Not allowing short-run concerns to dominate long-run ones, or marketing issues to stifle production needs, etc. Because of growth, ambitious goals, performance problems, and the like, resources were usually scarce in the situations in this study. Indeed, none of the fifteen GMs had extra cash for which there was no clear need. This scarcity made resource allocation an especially important task. Furthermore, the typical situation had a diverse set of activities that required resources because of the different products, markets, functions, and technologies involved. This diversity made resource allocation a complex task. Taken together, scarcity of resources and diversity of needs made the resource allocation task a most demanding balancing act. Under these conditions it was easy for short-run concerns to dominate long-run issues, or one product line to starve another, or one functional area to stifle another.
When I was with John Thompson, the United States economy had just gone into an economic downturn. Because sales were slipping, he had to reduce the resource budgets that had been previously planned in order to maintain some minimum level of profitability. In commenting on this, he told me:
Sometimes it is very difficult to judge how much to cut and where in a situation like this. If I cut too much, we do well this year, but it will hurt us in the future. If I don't cut enough, we can be hurt badly this year. If I overdo the cuts in operations, we could end up with business that we cannot handle. If I overdo the cuts in sales, we could end up with excess capacity in operations. It really is a tough balancing act.
Frank Firono talked about this same basic issue in this way:
In our business it is easy to crank out the short-term sales and relatively easy to get short-run profits. It's also relatively easy to get one store really performing well. What is difficult is to achieve acceptable short-run numbers while maintaining or increasing the quality of the business (a key long-term objective), and to get most or all of your stores performing pretty well.
To some degree, all the GMs in this study faced this problem. Evidence from elsewhere again suggests that this is probably the case for GM jobs in general.
3. Key Problem/Challenge #3: Keeping on top of a very large and diverse set of activities. Being able to identify problems ("fires") that are out of control and to solve them quickly. Because the buck stops at the GMs' desks, any problem associated with their businesses can become their problem. Any task that is not being accomplished effectively or efficiently can eventually create serious problems for them. But because of the typical scope of a GM's job, spotting fires that are out of control can be extremely difficult. And because of the diversity and complexity of these activities, figuring out how to put the fires out can also be very challenging.
Some of the GMs in this study were responsible for operations that spanned the entire globe. Some were responsible for the manufacturing and selling of hundreds of different kinds of products. Some were responsible for operations that employed many different technologies. In the case of a typical GM, thousands of people, most of whom were not physically located close to him, were somehow involved in his operations on a daily basis. Under these circumstances, simply trying to monitor daily or weekly operational activity can be extremely difficult. The most impressive information-systems technology available today cannot monitor all this activity quickly and accurately. Even if it could, a GM could spend twenty-four hours a day simply trying to digest that information. Furthermore, under these circumstances, the sheer volume of relatively minor short-run problems can be enormous. B.J. Sparksman echoed the sentiments of many of the people in this study when he told me that "sometimes this job is just a never-ending supply of little problems."
Furthermore, the complex nature of the operational activities associated with most of these jobs can make it very difficult to know what to do when a "problem" is seen. During my visits with the GMs, I saw numerous instances of this. In one fairly typical case, the company involved was experiencing difficulty making shipments. The general manager, Richard Papolis, was faced with two questions: first, how important was this problem (and how much attention, if any, should he give to it); second, why was this happening (what was the underlying problem)? Papolis' subordinates had varying opinions regarding these questions. Some felt the problem was due to the poor performance of two individuals in manufacturing and could be corrected fairly easily. Others felt it was more complex, systemic, and important. They argued that the entire manufacturing function had not been keeping up with the company's growth. Still others felt the problem was caused mostly by the marketing department, which had been having trouble forecasting orders accurately. A few initial discussions on these issues brought Papolis a lot of information -- both facts and opinions -- but no clear answer to either of his questions. Such was often the case.
Again, to some degree, this kind of problem was a part of all the jobs in this study. Evidence from elsewhere also suggests that the same is probably true for most or all GMs.
Job Demands II: Challenges and Dilemmas Associated with the Relationships
In addition to responsibilities, the GM jobs placed the incumbents in a web of relationships which both influenced and were influenced by those responsibilities. Each major type of relationship usually created its own set of challenges and problems.
4. Key Problem/Challenge #4: Getting the information, cooperation, and support needed from bosses to do the job. Being demanding with superiors without being perceived as uncooperative. Like other managers, the GMs were not able to do their jobs without some support and cooperation from their superiors. Their bosses could supply critical resources, information, and rewards. Because of this, because their superiors were human (not "perfect" bosses), and for still other reasons, another important job challenge or problem related to managing relationships with a boss or a group of bosses.
Gerald Allen and Dan Donahue, both of whom were located a few levels below their corporate CEOs, had relatively weak and unrespected immediate bosses. In both cases, this made the task of reporting to top management and getting its support more difficult. Without extra effort on the parts of Allen and Donahue, messages from the top sometimes did not arrive clearly or on time, and their ideas or requests did not receive enough top-management attention. In addition, because their bosses could offer them so little, just dealing with them on a daily basis was often frustrating and took time away from more important matters.
Terry Franklin and Bob Anderson were physically located more than 1,000 miles from their bosses, and their businesses accounted for less than 10 percent of their bosses' responsibilities. Franklin only saw his boss two or three times a year. These factors gave Franklin and Anderson considerable daily autonomy but made it difficult to get their bosses' attention, understandably, or help.
Paul Jackson reported to a very strong corporate CEO who had once been his peer (and rival). Because their management styles were also very different, Jackson found dealing with his boss to be most difficult. Others in the company reported that his boss had reprimanded him loudly in public on a number of occasions. At one point, Jackson told me that, because of his boss, his job "simply was not any fun anymore."
Other GMs in the study had still other problems that made managing their relationships to their bosses difficult or frustrating or both. Even in those cases where it was not a "problem," the task of "managing up" was taken very seriously by the GMs. They all recognized that, to some degree, current job performance and future career success depended on it. Such appears to be the case not only in other GM jobs, but in most managerial jobs.
5. Key Problem/Challenge #5: Getting corporate staff, other relevant departments or divisions, and important external groups (e.g., big unions or customers or suppliers) to cooperate despite the lack of any formal authority over them; getting things done through them despite resistance, red tape, and the like. Most of the GM jobs in this study had to interface with some corporate staff groups. Some also were required to coordinate functional groups that related to their businesses but that did not report directly to them. Still others had to deal with unions or other external groups because of their size and importance to the business. Many of these lateral relationships were somewhat adversarial in nature, and they often created problems for the GMs. Sometimes the problems were extensive.
When I asked Paul Jackson to tell me about the most difficult decision he had had to make in the past few years, he quickly replied that "making decisions is easy, but getting them implemented is sometimes nearly impossible. I have to work through so many people who do not report to me that it can be terribly difficult at times." John Thompson put it this way:
This would be a much easier, and I think, more fun job if we were left alone to do our work. Instead, there are so many people that want to get into the act: corporate staff, other divisions, people from Washington, unions, etc., etc.
When I was visiting Jack Martin, he learned that actions made by two others in his corporation, neither of whom reported to him, would cost him about $350,000 on his bottom line. In neither case was he consulted or warned in advance; he was simply informed after the fact. And (obviously) he was furious! I watched Gerald Allen sit in the office of a staff manager (who did not report to him) for about thirty minutes while that manager was blatantly rude and offensive to him. But Allen sat calmly until he received assurances that something he needed would be done. In still other cases, I saw GMs struggle with corporate advertising departments who provided little service at a big cost, with important customers who wanted unreasonable favors, and with a wide variety of other similar situations.
Evidence from elsewhere suggests that these problems posed by lateral relationships are common today in general-management jobs, especially those in divisionalized firms. Some evidence even suggests that they are a large part of most managerial jobs today.
6. Key Problem/Challenge #6: Motivating and controlling a large and diverse group of subordinates. Dealing with inadequate performance, interdepartmental conflict, and the like. The general management jobs in this study gave the incumbents authority over what was usually a large and diverse yet interdependent group of people. The people typically had different points of view, different stakes in organizational decisions and outcomes, and sometimes quite different personalities. But the GMs had to depend upon them all, because they directly affected areas for which the GMs were responsible.
The GMs in this study often talked about motivating good performance, and nearly half said that the most difficult decision they had to make in the past few years was in regard to replacing a key subordinate. In each of these cases, the subordinate was not performing adequately, and the GM had to make difficult judgements regarding whether the person could ever improve, how much longer it might take, and the costs and risks of inadequate performance in the interim. I even watched Michael Richardson agonize over such a problem. In that situation, two of the key people reporting to one of his subordinates had recently quit, and some of those who remained were very upset about the way their boss was managing things. Richardson had already tried a few solutions to "the problem" that had not worked, yet he did not want to make an irreversible decision unless it was absolutely necessary. He had known this person for fifteen years and had put him in his current position. The whole situation was agonizing for the subordinate, for Richardson, and for those around him, some of whom were pressuring him to act in very different ways.
The GMs in this study also frequently commented about conflict and communication problems among subordinates and their departments. Bob Anderson told me that "The editorial side of the organization and the business side of the organization are two different worlds. Like oil and water, they don't mix well." Richard Poullin pointed out that "The. creative people and the operating people would be at each other's throats in five minutes if we didn't constantly work to avoid these problems."
I witnessed a number of instances in which a GM had to become involved in a problem between subordinates from different departments. On some occasions, the conflicts were based on relatively simple misunderstandings. But in some cases, the problems were heated, very complex, and not at all easy to resolve. When I visited John Cohen, two of his young and ambitious subordinates were almost at each other's throats. Each sincerely believed the other person to be the source of conflict. One felt that the other was attacking him for political gain, the other saw the first as creating big problems for the company and refusing to listen to intelligent advice. After speaking to both of them, it was not at all obvious to me how to cool them off and resolve their differences without considerable effort.
Managing subordinates was a moderate-to-very-severe challenge in all the GM jobs in the study. Considerable evidence from elsewhere suggests that this is probably the norm for GM jobs, as well as for most other middle and upper level management jobs?
The Overall Demands: A Summary
All of the demands associated with GM job responsibilities and relationships are summarized in Figures 2.1 and 2.2. Any one of the problems and challenges shown in these figures could make a job difficult. All six together add up to what can be a particularly stressful situation and a very difficult time-management problem. Many of the general managers in this study worked sixty or more hours a week, and some confided that the job could absorb 120 hours if they did not set some limits on themselves. Moreover, they usually worked in a rapid-pace, high-pressure environment. John Thompson, one of the GMs who had recently started in his first general-management role, expressed the problem this way:
Before (in his previous job) it was easier to stop at 6:00, to turn off the job in my mind, and to go home. Now, even if I leave at 6:00, I find myself thinking or worrying about something on the way home and at home. There is always something.
Tom Long put it this way:
It never ceases to amaze me how many people want to see me each day, and how many different problems they come up with. We could schedule meetings all day and all night if we didn't try to control it.
Of course, general-management jobs are not unique in the time they require or in the pressures involved. Other professional and managerial jobs can be very time-consuming and stressful. Nor can the demands listed in Figures 2.1 and 2.2 be found only in GM jobs. To a degree, almost all management jobs have somewhat similar demands; but general-management jobs seem to be unique with respect to the overall diversity of demands.
No other managerial or professional job places a person in a position where he or she must deal with long-, medium-, and short-run tasks, as well as with a very diverse group of specialists in a variety of different kinds of relationships. All other jobs somehow limit the diversity of demands. For example, lower-level management jobs do not have the long-run responsibilities. Other higher-level management jobs do not have the same variety of specialists as subordinates. Staff jobs and traditional professional jobs seldom have many subordinates at all. Only GM jobs contain all this task- and people-related diversity. Ultimately, therefore, it is the diversity of complex demands that makes the job a general management job, that makes it different, and that makes it particularly difficult.
Differences in Job Demands
Although the GM jobs in this study shared the same six basic demands, there was considerable variation among them in terms of the overall intensity of those demands, the relative importance of the six problem areas, and the exact nature of each demand. For example, although all of the GM jobs in this study presented the incumbent with difficult decision-making problems, figuring out what to do was much more complex in some settings than others. Likewise, although all the jobs contained difficult implementation problems, getting things done was clearly much more demanding in some situations than in others.
In general, it appears that at least two major factors help create these kinds of differences associated with GM jobs (see Figure 2.3). First of all, differences in the job itself appear to be important. There seem to be a number of different kinds of GM jobs that are common today, each of which has a slightly different configuration of responsibilities and relationships. Second, differences in the business and corporate contexts seem to be very important. Running a banking business, for example, can present very different problems from running an automobile parts business. And running that business in a small, young, western company can present very different problems from running it in a large, old, eastern corporation.
Seven Different Kinds of GM Jobs
Before World War I, for all practical purposes there was only one type of GM job in the United States: a chief executive officer (CEO) reporting to a board of directors, with full responsibility for a functionally organized company. Even up to World War II, probably 95 percent or more of all GM jobs were of this single type. But in the past forty years, as businesses have grown larger, more diverse, and more complex, different kinds of GM jobs have emerged in some significant number. These jobs have been "invented" along with more and more complex organizational structures to help corporations manage their large size and diversity. Today there are at least seven different, commonly found types of GM jobs. These include what I will call the functional CEO; the multidivisional CEO; the group GM; the autonomous division GM; the semiautonomous division GM; the product/market GM; and the operations GM. Briefly, the seven types can be described as follows:
1. CEO in a functionally organized company: This is the "traditional" GM job, which reports to a board of directors (or chairman) and has functional managers reporting to it.
2. Corporate CEO in a multidivisional company: The most obvious difference between this and the first type is that general managers, in addition to some staff functional managers, report to this position. Furthermore, the multidivisional CEO usually has fewer short-run responsibilities.
3. Group GM: This type of GM job reports to a general manager and has GMs reporting to him. A typical group GM, for example, might report to a multidivisional CEO and have six or seven division GMs reporting to him. This job also tends to have fewer long-run responsibilities than a CEO and fewer important external lateral relations (e.g., to bankers).
4. Autonomous division GM: This type of GM is in many ways like the traditional job (#1), except that it reports to a GM, not a chairman or a board of directors. Like the group GM, it tends to have fewer long-run responsibilities than a CEO, more short-run responsibilities, and fewer corporate external lateral relations. Often the key responsibility in this job is for profit.
5. Semiautonomous division GM: This GM job is like the last, except that it has fewer downward but more internal lateral relations (to corporate staff), it reports more closely upwards, and it tends to have slightly fewer responsibilities overall. For example, a typical semiautonomous division GM might report to a Group GM (who has several other divisions with related products/services/markets) and have to rely to some degree on corporate (or group-level) personnel, legal, accounting, public relations, and financial staff.
6. Product/market GM: This type of general manager tends to have even fewer types of subordinates and more lateral relations. Typically, for example, mostly marketing people will report to this job, but the GM will be responsible for coordinating the manufacturing and engineering people associated with his business (or businesses). This job also has still fewer long-run responsibilities.
7. Operations GM: This final type of common GM job tends to have the least overall long-run and the most short-run responsibility. It tends to have mostly manufacturing or sales/service personnel reporting to it and to have some lateral relations (which, unlike the P/M GM, it does not have to coordinate closely). A typical operations GM might be the manager of a plant or a group of plants who is only partially responsible for a calculated "profit," and who has some personnel, accounting, and other staff reporting to him.
Of these seven types, the functional CEO, autonomous division GM, and the operations GM are probably the most common in existence today. The multidivisional CEO and group GM are probably the rarest (accounting for only about 1 percent of the total). For example, there are probably fewer than 1500 group GMs in the U.S. There are undoubtedly even more types of GM jobs that are even rarer and less visible today (such as, for example, "Sector" GMs) or are slightly different from the seven types described here.
Of the GMs in this study, two were operations GMs (Allen and Long); four were product/market GMs (Gaines, Martin, Jackson, and Donahue); three were semiautonomous division GMs (Poullin, Thompson, and Sparksman); five were autonomous division GMs (Anderson, Cohen, Franklin, Firono, and Papolis); one was a functional CEO (Richardson); and none was a group GM or mutidivisional CEO GM. The absence of the latter two types was purposive, although in retrospect it was a mistake. (When designing the study I was warned not to include GMs who were managers of GMs because those jobs may be "different." Implicitly, the assumption was that there may be two types of GM jobs.)
Some people might argue that only the first four types of GM jobs are really GM jobs (or perhaps just #1, #2, and #4, or even just #1 and #4). In doing so, they implicitly apply a relationship-oriented criterion in defining the job. For example, the most common restrictive definition assumes that all GM jobs, like the traditional one, should have no lateral relationships of any consequence inside the firm and always have all relevant functions reporting to them. But such a definition misses, I think, the core of the job, which is responsibility-oriented. All of the seven types of GM jobs identified here have some multifunctional responsibility for a business or businesses; that is the real essence of a GM job.
One of the giant corporations in this study had six of the seven types of GM jobs within its structure. Figure 2.4 shows in a rough way how these six types of GM jobs were different yet related in this one setting. The corporate CEO job in that corporation was unique in that it had long-run responsibilities that were much more demanding than its short-run responsibilities. The operations GM job stands out for the opposite reason; it had short-run responsibilities that were much more demanding than its long-run ones. The product/market GM job was most different because it had the greatest lateral relationship demands. And the group GM job stands out because, when the incumbents had autonomous division GMs reporting to them, they were in the unusual position of having both a boss and a subordinate who had more autonomy (in a lateral sense) than they did (which did not make the group GMs very comfortable).
The overall pattern in Figure 2.4 is that as one moves down
Excerpted from General Managers by John P. Kotter Copyright © 1986 by John P. Kotter. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.