CHAPTER SEVENTEEN

The Building Blocks of Organizations …

IN DESIGNING THE BUILDING blocks of organization four questions face the organizer.

1. What should the units of organization be?
2. What components should join together, and what components should be kept apart?
3. What size and shape pertain to different components?
4. What is the appropriate placement and relationship of different units?

From the earliest beginnings of work on organization, well over a century ago, these were the tasks the organizer had to confront before designing a structure.

We therefore now have a considerable amount of experience. There are no prescriptions for the design of the building blocks or for the design of the structure itself. But one can clearly indicate what the right approaches are and what approaches are unlikely to work.

The traditional approach to the identification of the basic units of organization has been to analyze all the activities needed for performance in the enterprise. This then results in a list of typical functions of a manufacturing business or of a retail business.

This approach to the typical functions sees organization as mechanical, as an assemblage of functions. But organization has to be “organic.” Organizations will indeed use typical activities—though not necessarily all of them. But how the structure is to be built depends on what results are needed. Organizing has to start out with the desired results.

The Key Activities

What we need to know are not all the activities that might conceivably have to be housed in the organization structure. What we need to know are the load-bearing parts of the structure, the key activities.

Organization design, therefore, starts with these questions:

In what area is excellence required to obtain the company’s objectives?

In what areas would lack of performance endanger the results, if not the survival, of the enterprise?

Here are some examples of the kind of conclusions these questions lead to.

Sears, Roebuck in the United States and Marks & Spencer in England are in many ways remarkably similar, if only because the founders and builders of Marks & Spencer consciously modeled their company on Sears, Roebuck. But there is a pronounced difference in the organizational placement and organizational role of the “laboratory” in these two companies. Sears, which defines its business as being “the buyer of the right goods for the American family,” uses its laboratory to test the merchandise it buys. Accordingly, the laboratory, while large, competent, and respected, is organizationally quite subordinate; for what goods Sears will buy has already been decided by the marketplace. The buyer in Sears makes the decision; the laboratory determines only standards and checks up on the goods the manufacturer delivers. Marks & Spencer, on the other hand, defined its business as “creating and developing upper-class goods for the working-class family.” The goods Marks & Spencer sells did not, as a rule, exist until Marks & Spencer took an expensive “upper-class” article and completely redesigned it so that it retained or improved its quality but could be made at a fraction of its former price. And this has largely been the task of the laboratory. As a result, the laboratory is central to Marks & Spencer’s organization structure. The laboratory rather than the buyer decides what new products are desirable, develops the new merchandise, designs it, tests it, and then gets it produced. Only then does the buyer take over. As a result, the head of the Marks & Spencer laboratory is a senior member of management and, in many ways, the chief business planner.

Any company that shows outstanding success will be found to have made the key activities—and especially those in which excellence is needed to attain business performance and business objectives—the central, load-carrying elements in its organization structure.

But equally important are the questions “In what areas could malfunction seriously damage us? In what areas do we have major vulnerability?” They are questions, however, that are much less often asked.

The New York brokerage community, by and large, did not ask it. If it had, it would have realized that malfunction of the “back office,” where customer orders, customer accounts and securities are handled, could seriously endanger the business. Failure to organize the back office as a key activity was the single most important cause for the severe crisis that overtook Wall Street in 1969 and 1970 and destroyed a good many of the best-known and apparently most successful firms. The one Wall Street firm, however, that had asked those questions, Merrill Lynch, and had organized the back office as a load-bearing key activity in its structure, emerged from the crisis as the giant of the brokerage business.

Finally, the question should be asked “What are the values that are truly important to us in this company?” It might be product or process safety. It might be product quality. It might be the ability of the company dealers to give proper service to the customer and so on. Whatever the values are, they have to be organizationally anchored. There has to be an organizational component responsible for them—and it has to be a key component.

These three questions will identify the key activities. And they in turn will be the load-bearing, the structural elements of organization. The rest, no matter how important, no matter how much money they represent, no matter how many people they employ, are secondary. Obviously, they will have to be analyzed, organized, and placed within the structure. But the first concern must be those activities that are essential to the success of a business strategy and to the attainment of business objectives. They have to be identified, defined, organized, and centrally placed.

An analysis of key activities is needed in the business that has been going for some time, and especially in the business that has been going well. In such a business the analysis will invariably reveal that important activities are either not provided for or are left hanging in midair to be performed in a haphazard fashion. It will almost invariably bring out activities that, once important, have lost most of their meaning but continue to be organized as major activities. It will demonstrate that historically meaningful groupings no longer make sense but have, instead, become obstacles to proper performance. And it will certainly lead to the discovery of unnecessary activities that should be eliminated.

The new business needs such thinking. But the greatest need for key-activity analysis is found in the business that has been growing fast. Rapid growth is both a disorganized and a disorganizing process. The enterprise that starts out, so to speak, in a lowly but functional two-room cottage, puts in, as it grows, a new wing here, an attic there, a partition elsewhere, until it is housed in a twenty-six-room monstrosity in which all but the oldest inhabitants need a St. Bernard to bring them back from the water cooler. To reorganize mechanically in such a situation—the usual approach—will make things worse. To copy the “GM organization” in such a situation will put on a tremendous superstructure of “staff” and “coordinators,” without remedying the basic structural defects. Only a key-activities analysis starting out from objectives and strategy can provide the organization structure the business really needs.

A business should always analyze its organization structure when its strategy changes. Whatever the reason—a change in market or in technology, diversification or new objectives—a change in strategy requires a new analysis of the key activities and an adaptation of the structure to them. Conversely, reorganization that is undertaken without change in strategy is either superfluous or indicates poor organization to begin with.

The Contributions Analysis

From the earliest days of concern with organization, a hundred years ago, the most controversial question has been “What activities belong together and what activities belong apart?” A number of answers have been given over the years.

Perhaps the earliest one was the German division of a business into two major areas: the “technical,” embracing research, engineering, and production, and the “commercial,” embracing sales and finance. Somewhat later came “line” and “staff,” which tried to distinguish “operating” and nonoperating “advisory” activities. Finally there was Henri Fayol’s analysis of functions, defined (too narrowly) as “bundles of related skills,” which still underlies the typical organization of most businesses.

All of these have merit. But a more searching analysis is needed which groups activities by the kind of contribution they make.

There are, by and large, four major groups of activities, if distinguished by their contribution.

There are, first, result-producing activities—that is, activities which produce measurable results which can be related, directly or indirectly, to the results and performance of the entire enterprise. Some of these activities are directly revenue-producing. Others contribute measurable results.

There are, second, support activities which, while needed, and even essential, do not by themselves produce results but have results only through the use made of their “output” by other components within the business.

There are, third, activities which have no direct or indirect relationship to the results of the business, activities which are truly ancillary. They are hygiene and housekeeping activities.

Finally, and different in character from any of these, is the top-management activity.

Among the result-producing activities, there are some that directly bring in revenues (or, in service institutions, directly produce “patient care” or “learning”). Here belong innovating activities, selling and all the work needed to do a systematic and organized selling job, such as forecasting, market research, sales training, and sales management. Here also belongs the treasury function, that is, the supply and management of money in the business.

In a commercial bank, all lending operations, the fiduciary activities of managing other people’s money, and, of course, the money-making operation of the bank itself, that is, the management of its own liquid funds, are revenue-producing activities. In a department store buying and selling are always revenue-producing operations (and at Marks & Spencer innovation has also been a revenue-producing activity). In a life insurance company selling is obviously revenue-producing. But so is the actuarial activity insofar as it develops new types of policy; and finally, investment is an important—in many insurance companies the most important—revenue-producing activity.

The second group of result-producing activities are those which do not generate revenue but can still be directly related to the results of the entire business, or of a major revenue-producing segment. I call them result-contributing rather than result-producing.

Manufacturing is typical of these activities. But training of people belongs here too, as does their original recruitment and employment, that is, the activities concerned with the supply of qualified and trained people to the enterprise. Purchasing and physical distribution are result-contributing but not revenue-producing activities. “Engineering,” as the term is normally understood in most manufacturing businesses, is a result-contributing but not a revenue-producing activity. In a commercial bank “operations,” that is, the handling of data and papers, belong here; in a life insurance company, claims settlement. Labor negotiations and many other similar “relations” activities are result-contributing though not revenue-producing.

The third group of result-producing activities are information activities. They do produce a “finished product” that is needed by everyone in the system. Information performance can also be defined and measured, or at least appraised. Yet information, by itself, does not produce any revenue. It is “supply” to revenue and cost centers alike.

First among the “support” activities which do not by themselves produce a “product” but are “input” to others stand the “conscience” activities. These activities set standards, create vision, and demand excellence in all the key areas where a business needs to strive for excellence.

Conscience activities tend to be slighted in most organizations. But every company—and every service institution—needs to provide itself and its managers with vision, with values, with standards, and with some provision for auditing performance against these standards.

There are indeed in all larger businesses people who are supposed to do this job, usually the executives who head up major “service staffs.”But their first duty is not to be the organization’s conscience but to be servant of, and advisor to, operating managers. As a result they rarely get around to doing the conscience job systematically. Instead they run departments.

Another support function is advice and teaching, i.e., the traditional service staffs. The contribution is not in what the activity itself does or can do, but the impact it has on the ability of others to perform and to do. The “product” is an increased performance capacity of the rest of the organization.

A good many of the “relations” activities are also “support”—as is the legal staff, or the patent department.

The last group of activities defined by their contribution are the hygiene and housekeeping activities, ranging from the medical department to the people who clean the floor, from the plant cafeteria to the management of pension and retirement funds, from finding a plant site to taking care of all the manifold record-keeping requirements imposed on business by government. These functions contribute nothing directly to the results and performance of the business. Their malfunction, however, can damage the business. They serve legal requirements, the morale of the work force, or public responsibilities. Of all activities they are the most diverse. And of all activities they tend to get the shortest shrift in most organizations.

This is a very rough classification, and far from scientific. Some activities may be put into one category in one business, into another one in another, and in a third company will be left fuzzy and without clear classification at all.

In some manufacturing companies, manufacturing is a cost center. It contributes results but does not generate revenue. But there are some true manufacturing businesses, i.e., businesses whose revenue is generated by manufacturing sans research, sans engineering, sans selling. There are businesses where licensing, selling, and buying of patents are major revenue producers.

Purchasing, while normally a support activity, can also be defined as part of a result-contributory activity: “materials management,” which includes manufacturing and physical distribution, all three managed together to minimize costs of goods and money needs, and maximize quality, delivery, and customer satisfaction.

Why classify then? The answer is that activities that differ in contribution have to be treated differently. Contribution determines ranking and placement.

Key activities should never be subordinated to nonkey activities.

Revenue-producing activities should never be subordinated to nonrevenue-producing activities.

And support activities should never be mixed with revenue-producing and result-contributory activities. They should be kept apart.

The “Conscience” Activities

Activities that are the conscience of an organization must never be subordinated to anything else. They also should never be placed with any other activity; they should be clearly separate.

The conscience function of giving vision, of setting standards, and of auditing performance against these standards is basically a top-management function. But it has to work with the entire management group. Every business, even a small business, needs this function. In a small business, it need not be set up as a separate function but can be discharged as part of the top-management job. In any business of more than medium size, however, the function has usually to be set up and staffed separately.

However, there should be very few people actually doing the conscience job. It is a job for a single individual rather than a staff. It is a job for someone whose performance has earned the respect of the management group. It is not a job for a “specialist.” It is best discharged by a senior member of the management group with proven performance record who has manifested concern, perception, and interest in the area for which that senior member is supposed to act as the conscience.

Only those few areas that are vital and central to a companys success and survival should become areas of conscience. Objectives and strategy determine what conscience activities are needed. Managing people is always a conscience area, and so is marketing. The impact of a business on its environment, its social responsibilities and basic relations with the outside community, are also basic conscience areas. Innovation (whether technological or social innovation) is likely to be a conscience area for any large business.

Beyond these, however, there is no formula.

Conscience work is incompatible with operating and with giving advice.

The only activity that should “report” to a person in charge of a conscience activity is auditing and the actual performance of managers. For it is not enough to have vision and to set standards. Performance of the organization against these standards needs to be appraised regularly.

“Conscience,” many people will argue, is a very strong, in addition to being a rather strange, term. But it is the right term. The task of the conscience activities is not to help the organization do better what it is already doing. The task is to remind the organization all the time of what it should be doing and isn’t doing. The task is to be uncomfortable, to hold up the ideal against the everyday reality, to defend the unpopular and to fight the expedient.

This requires, however, self-discipline on the part of the conscience executive and the organization must accept the conscience executive’s competence and integrity.

The tenure of the few conscience executives should be limited as a rule. No matter how greatly they may be respected, and no matter how successful they have been, most conscience executives will eventually wear out either their integrity or their welcome. This is a good place for a senior person to end a distinguished career. A younger person in the job should be moved out after a few years—preferably back into a “doing” job.

Making Service Staffs Effective

There are similarly stringent rules with respect to advisory and teaching activities, that is, with respect to service staffs.

There should be very few of them. They should be set up only in key activity areas. It is counterproductive to have a service staff in every function. The secret of effective service work is concentration rather than busyness.

Advisory and teaching staffs should never try to do a little bit of everything. They should zero in on a very small number of crucial areas. Rather than serve everybody they should select “targets of opportunity,” i.e., areas within the organization where the managers are receptive and do not have to be “sold,” and where achievement will lead to wide-spread imitation throughout the company.

The staffs and their activities should be kept lean.

The supply of people of the right temperament for this kind of work is not very large. To do a decent job in an advisory and teaching capacity requires someone who genuinely wants others to get the credit. It requires someone who starts out with the aim of enabling others to do what they want to do, provided only that it is neither immoral nor insane. It requires further a person who has the patience to let others learn rather than go and do the work himself or herself. And finally, it requires the kinds of people who will not abuse the position in headquarters close to the seat of power to politick, to manipulate, and to play favorites. People who possess these personality traits are rare. Yet people without them in services work can do only mischief.

One basic rule for advisory and teaching staffs is that they abandon an old activity before they take on a new one. Otherwise they will soon start to “build empires” or to produce “canned goods,” that is, programs and memoranda, rather than develop the knowledge and performance capacity of the operating staff whose job it is to produce. They will also otherwise be forced to use second-raters rather than people of outstanding competence. Only if they are being required to abandon an old activity before taking on a new one will they be able to put on the job the really first-rate people in the group.

Advisory and teaching activities should never “operate.” A common weakness of personnel staffs is that they operate. They run the labor negotiations, they do a lot of housekeeping chores such as managing the cafeteria, or they train. As a result the advisory and teaching work does not get done. The “daily crisis” in “operations” takes precedence over the work of advice and teaching which can always be postponed. Mixing advice and operations means building large staffs rather than building performance.

Other service staffs are just as guilty of mixing “doing” work in with advice and teaching—and thereby shifting either the one or the other.

A company may have need for acoustics engineering while no division, however, has enough work in the area to justify its own acoustics engineers. It seems logical therefore to put a few good acoustics engineers into “manufacturing services” or “engineering services.” The acoustics people, however, are not service staff but result-producers who go to work wherever, in a given division, an acoustics job is to be done. They are not expected to give advice or to teach, but to do. If placed into a services component the unit will rarely produce. The good people in it will become frustrated and are unlikely to stay long. Acoustics engineering is “operating” work, no matter where it is done. And if no division needs enough of it to support a full effort, it has to be set up as “joint operating” work by a number of divisions, either within the division that needs it the most but serving the other divisions as well, or as a separate group that has a number of “customers,” that is, a number of divisions for whom it does acoustics work.

If “joint-operating work” is needed—and it often is—there might be a separate central operating pool under one manager for all such work, regardless of technical area. The management problems in all joint operating work are the same: relations, assignments, priorities, and standards.

Advisory and teaching activities are service institutions. They should be required to impose on themselves the self-discipline of setting objectives, of setting targets, of determining priorities and measuring their results against them. They should not have a monopoly. If consulting or teaching work has to be done other than in their areas of concentration somebody from the outside should be brought in to do it. And insofar as possible, their “customers,” the managers of the various units, should have the choice between using the internal advisory and teaching staff, going on the outside, or not using any staff at all.

Advisory and teaching work should not be a career. It is work to which managers or career professionals should be exposed in the course of their growth. But it is not work which someone should, normally, do for long. As a career it corrupts. It breeds contempt for “those dumb operating people,” that is, for honest work. It puts a premium on being “bright,” rather than on being right. It is also frustrating work because one does not have results of one’s own but results only at secondhand.

But it is excellent training, excellent development, and a severe test of a person’s character and ability to be effective without having the authority of command. It is an experience everyone should have had who rises to the top of an organization. But it is an exposure no one should suffer for more than a limited period of time.

There is constant discussion these days whether this kind of work requires a high degree of specialized knowledge or whether a good “generalist” can rapidly acquire enough of a “smattering of ignorance” to be effective in advisory and teaching work. The debate rages forever in any consulting practice. There is no answer to the question—indeed, it is probably the wrong question. In some areas, clearly, professional and highly specialized competence is a requirement. If a company, for instance, needs advice and teaching in advanced areas of polymer chemistry or in structuring highly complex and risky international capital transactions, somebody with a background in marketing or in purchasing need not apply, no matter how good and perceptive a teacher he or she is. But in many areas of advice and teaching the generalist who is willing to learn, who will think through the relationship with the “client,” and who takes responsibility for his or her contribution, is likely to do a better job than the highly specialized expert who refuses to try to be understood and is contemptuous of the “laity.” Indeed, in most successful advisory and teaching activities, the expert is the “inside” man or woman who furnishes the tools for the consultants, but who is not active, let alone effective, in advising and teaching work.

The Two Faces of Information

Information activities present a special organizational problem. In the terms the chemist uses, they are “bi-valent”; they have two faces, two dimensions, and require two different “bonds.” Unlike most other result-producing activities, they are not concerned with the entire process itself. This means that they have to be both centralized and decentralized.

Information-producing activities, whether accounting work or operations research, resemble the nervous system of a biological organism which is also both centralized and decentralized down to the smallest and most remote cell in the body.

Information activities therefore have two organizational homes rather than one.

The traditional organization chart expresses this in the two different lines that connect an information activity to “bosses”: a solid line to the head of the unit for which it is the information provider, and a dotted line to the central information group, for instance, the company controller. One conclusion from this is that information work should be kept separate from other kinds of work.

American business has typically violated this rule by putting accounting, i.e., a traditional information activity, into one component with the treasurer, i.e., the result-producing operating work of supplying capital and managing money in the business. The justification has been that both “deal with money.” But, of course, accounting does not deal with money; it deals with figures. The consequence of the traditional approach has been the slighting of financial management. As long as money was—or seemed—cheap this could be excused; to neglect money management did not cost very much. But the era of cheap money came to an end around 1970; since then too slight financial management has become an expensive mistake.

The tough question with respect to information activities is which of them belong together and which should be kept apart. There is much talk today about “integrated total information systems.” This of course implies that all—or at least most—information activities should be in one component. Insofar as this means that new and different information activities, e.g., operations research or a computer system, should not be subordinated to traditional accounting, the point is well taken. But should they be coordinated? Or should they be separate?

There is, so far, no clear answer and no satisfactory way to organize information work—though it is clearly a key activity. Nobody has yet seen a total information system. No one may ever see one. But as we develop information capacity we will have to grapple with the organizational problem and will have to find answers, or at least approaches.

Hygiene and Housekeeping

The last group of activities according to their contribution are hygiene and housekeeping activities. They should be kept separate from other work, or else they will not get done. The problem is not that these activities are particularly difficult. Some are. Many others are not. The problem is that they are not even indirectly related to results. Therefore, they tend to be looked down upon by the rest of the organization. They are “donkey work” because they are neither result-producing nor professional work.

One reason for the tremendous increase in health-care costs in the U.S. is managerial neglect of the “hotel services” by the people who dominate the hospital, such as doctors and nurses. They all know that the hotel services are essential and that patients do not get well unless they are reasonably comfortable, are being fed, have their beds changed and their rooms cleaned. But these are not professional activities for a doctor, for a nurse, or an X-ray technician. They are not willing to yield an inch to make it possible for the people in charge of the hotel services to do their jobs. They are not willing to have these activities represented on the upper levels of hospital management. As a result, no “respectable” manager in a hospital wants to have anything to do with these activities. They are left unmanaged. And this means they are done badly and expensively.

There rarely is such a problem with the medical department within a company—if only because our value system respects and places the doctor high in our social hierarchy. But even so important a function as the selection of a plant site, or the construction of a plant are often considered “extraneous” by the people within a business. Activities where less seems to be at stake, whether the parking lot, the cafeteria, or maintenance in general, tend to be slighted and neglected.

This extends even to activities in which a great amount of money is at stake. Very few companies in the United States for instance have done even an adequate job of managing the pension funds of their employees, despite the enormous amount of money involved and the serious impact on the company’s future. It is an activity which does not, it seems, have any relationship to results and therefore it is an activity which tends to be relegated to somebody else.

One way out is to turn hygiene and housekeeping activities over to the work community to run. They are activities “for” the employees and they are therefore best managed by the employees. Or, such activities may be farmed out to somebody whose business it is to run a pension fund or to manage a cafeteria.

But insofar as a company’s management has to do these things itself—and picking a plant site and building a factory is something a company has to do for itself, or at least has to participate in actively—hygiene and housekeeping activities ought to be kept separate from all others. They require different people, different values, different measurements—and should require little supervision by business management itself.

One example are the autonomous real estate management companies which large businesses have created to handle everything concerned with the procurement of real estate, the construction of a building or a factory, and the management and maintenance of buildings. Another example is the General Services Administration of the U.S. Government, which handles all housekeeping tasks for all government agencies. For the senior members of the departments of government, managing the automotive fleets for their units is a chore for which they have neither interest nor respect. Yet there obviously is a good deal of money at stake—and cars need organized, systematic purchasing and organized, systematic maintenance. For the General Services Administration the administration of the government automotive fleet is its business and can be organized as such.

There is one overall rule. Activities that make the same kind of contribution can be joined together in one component and under one management, whatever their technical specialization. Activities that do not make the same kind of contribution do not, as a rule, belong together.

It is entirely feasible—indeed, it often is the best way—to put all advising and teaching activities, in personnel, in manufacturing, in marketing, or in purchasing, in one “services” group under one manager. Similarly, in any but large companies, one person might well be the company’s conscience in major conscience areas. Contribution rather than skill determines function.