How Should You Organize Manufacturing?

Leer en español

Ler em português

Among the characteristics of a company that shape corporate and therefore manufacturing strategy are its dominant orientation (market or product), pattern of diversification (product, market, or process), attitude toward growth (acceptance of low growth rate), and choice between competitive strategies (high profit margins versus high output volumes). Once the basic attitudes or priorities are established, the manufacturing arm of a company must arrange its structure and management so as to reinforce these corporate aims. Examining the extremes of “product-focused” and “process-focused” organizations, the authors illustrate the development of a “manufacturing mission” whereby the organization of manufacturing supports management’s needs.

Manufacturing organizations tend to attract the attention of general managers the way airlines do: one only notices them when they’re late, when ticket prices rise, or when there’s a crash. When they are operating smoothly, they are almost invisible. But manufacturing is getting increasing attention from business managers who, only a few years ago, were preoccupied with marketing or financial matters.

The fact is that in most companies the great bulk of the assets used—the capital invested, the people employed, and management time—are in the operations side of the business. This is true of both manufacturing and service organizations, in both the private and public sectors of our economy. These resources have to be deployed, coordinated, and managed in such a way that they strengthen the institution’s purpose; if not, they will almost certainly cripple it.

The problems and pressures facing manufacturing companies ultimately find their way to the factory floor, where managers have to deal with them through some sort of organizational structure. Unfortunately, this structure often is itself part of the problem. Moreover, problems in a corporation’s manufacturing organization frequently surface at about the same time as problems in the rest of the company, and they surface in a variety of ways. For example:

• A fast-growing, high-technology company had quadrupled in size in a ten-year period. Its manufacturing organization was essentially the same at the end of that period as before, dominated by a powerful vice president for manufacturing and a strong central staff, despite the fact that its product line had broadened considerably, that the company was beginning to make many more of the components it formerly purchased, and that the number of plants had both increased and spread into four countries. A sluggishness and sense of lost direction began to afflict the manufacturing organization, as overhead and logistics costs soared.

• A conglomerate had put together a group of four major divisions that made sense in terms of their financial and marketing synergy. But these divisions’ manufacturing organizations had little in common, little internal direction, and no overall coordination. The parent company was confronted with a series of major capital appropriation requests and had little understanding of either their absolute merits or the priorities that should be attached to them.

• A fast-growing company in a new industry had for a number of years operated in a seller’s market, where competition was based on quality and service rather than price. Its manufacturing organization was highly decentralized and adept at new product introduction and fast product mix changes. In the 1970s severe industry overcapacity and price competition caused corporate sales to level off and profit to decline for the first time in its history. Manufacturing efficiency and dependability clearly had to be improved, but there was fear of “upsetting the corporate culture” and “crippling the golden goose.”

Why did these companies’ manufacturing arms get into trouble? And to what extent were these problems the outgrowth of poorly designed organizational structures? In attempting an answer to these questions, we will begin with a review of the concepts of “manufacturing mission” and “manufacturing focus” that were first defined and explored in a series of articles by Wickham Skinner beginning in 1969.1 These concepts, and the conclusions that flow logically from them, have since been polished, elaborated, and tested by him and a number of his colleagues in conjunction with various manufacturing companies over the past several years.

After this review we will evaluate the advantages and disadvantages of different approaches to organizing a company’s manufacturing function and then apply our concepts to recommending the type of organizational design that is most appropriate for a given company. Finally, we will discuss the various kinds of growth that companies can experience and how these expectations should affect the organization of the manufacturing function.

Basic Elements of Strategy

The concept of manufacturing strategy is a natural extension of the concept of corporate strategy, although the latter need not be as rational and explicit as management theorists usually require.2 As we use the term, a corporate strategy simply implies a consistency, over time, in the company’s preferences for and biases against certain management choices as shown in Exhibit I. We use the term company to refer to a business unit that has a relatively homogeneous product line, considerable autonomy, and enough of a history to establish the kind of track record we refer to here. Such a “company” could, of course, be a relatively independent division within a larger enterprise. The following four “attitudes” shape those aspects of a company’s corporate strategy that are relevant to manufacturing.

Exhibit I Corporate Attitudes That Imply Strategic Preferences

Dominant orientation—Some companies are clearly market oriented. They consider their primary expertise to be the ability to understand and respond effectively to the needs of a particular market or consumer group. In exploiting this market knowledge, they use a variety of products, materials, and technologies. Gillette and Head Ski are examples of such companies. Other companies are clearly oriented to materials or products; they are so-called steel companies, rubber companies, or oil companies (or, more recently, energy companies). They develop multiple uses for their product or material and follow these uses into a variety of markets. Corning Glass, Firestone, DuPont, and Conoco come to mind. Still other companies are technology-oriented—most electronics companies fall into this class—and they follow the lead of their technology into various materials and markets.

A common characteristic of a company with such a dominant orientation is that it seldom ventures outside that orientation, is uncomfortable when doing so, often does not appreciate the differences and complexities associated with operating the new business, and then often fails because it hesitates to commit the resources necessary to succeed. A recent example of a company that ventured, with considerable trauma, outside its dominant orientation was Texas Instruments’ entry into consumer marketing of electronic calculators and digital watches.

Pattern of diversification—Diversification can be accomplished in several ways: (1) product diversification within a given market, (2) market diversification (geographic or consumer group) using a given product line, (3) process or vertical diversification (increasing the span of the process so as to gain more control over vendors and/or customers) with a given mix of products and markets, and (4) unrelated (horizontal) diversification, as exemplified by conglomerates. Decisions about diversification are closely interrelated with a company’s dominant orientation, of course, but they also reflect its preference for concentrating on a relatively narrow set of activities or, alternatively, its willingness to enter into a wide variety of activities, products, and/or markets—and which ones it will enter.

Corporate attitude toward growth—Does growth represent an input to or an output of the company’s planning process? Every company continually confronts a variety of growth opportunities. Its decisions about which to accept and which to reject signal, in a profound way, the kind of company it prefers to be. Some companies, in their concentration on a particular market, geographic area, or material, essentially accept the growth permitted by that market or area or material consumption. A company’s acceptance of a low rate of growth reflects a decision, conscious or unconscious, to retain a set of priorities in which a given orientation and pattern of diversification are more highly valued than growth.

Other companies, however, are so structured and managed that a certain rate of growth is required in order for the organization to function properly. If its current set of products and markets will not permit this desired rate of growth, it will seek new ones to “fill the gap.” Again, this decision will closely reflect its attitudes regarding dominant orientation and diversification. One obvious indication of a company’s relative emphasis on growth is how growth is treated in its planning, budgeting, and performance evaluation cycle, and particularly the importance that is placed on annual growth rate, compared with such other measures as return on sales or return on assets. It is necessary to differentiate between a company’s stated goals—words on paper—and what actually moves it to action.

Choice of competitive priorities—In its simplest form this choice is between seeking high profit margins or high output volumes. Some companies consistently prefer high margin products, even when this limits them to relatively low market shares. Others feel more comfortable with a high-volume business, despite the fact that this commits them to severe cost-reduction pressure and often implies low margins. An interesting article describes David Packard’s attempts to redirect Hewlett-Packard away from the latter approach, where it was nose-to-nose with Texas Instruments, and back toward the former approach.3

This concept can be expanded and enriched, however, since companies can compete in ways other than simply through the prices of their products. Some compete on the basis of superior quality—either by providing higher quality in a standard product (for example, Mercedes-Benz) or by providing a product that has features or performance characteristics unavailable in competing products. We intend here to differentiate between an actual quality differential and a perceived difference, which is much more a function of selling and advertising strategy.

Other companies compete by promising utter dependability; their product may be priced higher and may not have some of the competitive products’ features or workmanship. It will, however, work as specified, is delivered on time, and any failures are immediately corrected. IBM has been cited as an example of a company that competes on this basis; in a sense, so do AT&T and Sears, Roebuck.

Still others compete on the basis of product flexibility, their ability to handle difficult, nonstandard orders and to lead in new product introduction. This is a competitive strategy that smaller companies in many industries often adopt. And, finally, others compete through volume flexibility, being able to accelerate or decelerate production quickly. Successful companies in cyclical industries like housing or furniture often exhibit this trait.

In summary, within most industries different companies emphasize one of these five competitive dimensions—price, quality, dependability, product flexibility, and volume flexibility. It is both difficult and potentially dangerous for a company to try to compete by offering superior performance along several competitive dimensions. Instead, a company must attach definite priorities to each that describe how it chooses to position itself relative to its competitors.

Practically every decision a senior manager makes will have a different impact on each of these dimensions, and the organization will thus have to make trade-offs between them. Unless these trade-offs are made consistently over time, the company will slowly lose its competitive distinctiveness.

Without such consistency, it does not matter how much effort a company puts into formulating and expounding on its “strategy”—it essentially does not have one. One test of whether a company has a strategy is that it is clear not only about what it wants to do but also about what it does not want to do—what proposals it will consistently say no to.

Toward a Manufacturing Mission

Once such attitudes and competitive priorities are identified, the task for manufacturing is to arrange its structure and management so as to mesh with and reinforce this strategy. Manufacturing should be capable of helping the company do what it wants to do without wasting resources in lesser pursuits. This is what we call the company’s “manufacturing mission.”

It is surprising that general managers sometimes tend to lose sight of this concept, since the need for priorities permeates all other arenas of management. For example, marketing managers segment markets and focus product design, promotional, and pricing effects around the needs of particular segments, often at the expense of the needs of other segments. And management information systems must be designed to emphasize particular kinds of information at the expense of others.

While it is possible to chalk up to inexperience the belief of many general managers that manufacturing should be capable of doing everything well, it is harder to explain why many manufacturing managers themselves either try to be good at everything at once or focus on the wrong thing. They know that all-purpose tools generally are used only when a specific tool is not available. Perhaps they fall into this trap because of pride, or too little time, or because they are reluctant to say no to their superiors.

All these factors enter into the following scenario. A manufacturing manager has nicely aligned his organization according to corporate priorities when suddenly he is subjected to pressure from marketing because of “customer complaints” about product quality or delivery times. Under duress, and without sufficient time to examine the trade-offs involved, he attempts to shore up performance along these dimensions. Then he is confronted with pressure from finance to reduce costs or investment or both. Again, in the attempt to respond to the “corporate will,” or at least to oil the squeaky wheel, he reacts. Step by step, priorities and focus disappear, each lagging dimension being brought into line by some function’s self-interest.

Falling into such a trap can be devastating, however, because a manufacturing mission that is inconsistent with corporate strategy is just as dangerous as not having any manufacturing mission at all. The more top management delegates key manufacturing decisions to “manufacturing specialists” (usually engineers), the more likely it is that manufacturing’s priorities will be different from corporate priorities. They will reflect engineering priorities, or operating simplicity (often the goal of someone who has worked his way up from the bottom of the organization)—not the needs of the business.

Using Structural Decisions

Translating a set of manufacturing priorities into an appropriate collection of plant, people, and policies requires resources, time, and management perseverance. As we mentioned earlier, the great bulk of most companies’ assets (capital, human, and managerial) is found in manufacturing. Moreover, these assets tend to be massive, highly interrelated, and long lived—in comparison with marketing and most financial assets. As a result, it is difficult to redirect them, and “fine-tuning” is almost impossible. Once a change is made, its impact is felt throughout the system and cannot be undone easily.

Such manufacturing inertia is made worse by many manufacturing managers’ reluctance to change. And it is further compounded by many top managers’ lack of understanding of the kind of changes that are needed, as well as by their unwillingness to commit the resources to effect such changes.

The decisions that implement a set of manufacturing priorities are structural; for a given company or business they are made infrequently and at various intervals. They fall into two broad categories: facilities decisions and infrastructure decisions.

Facilities decisions involve the following considerations:

1. The total amount of manufacturing and logistics capacity to provide for each product line over time.

2. How this capacity is broken up into operating units (plants, warehouses, and so on), their size and form (a few large plants versus many small ones), their location, and the degree or manner of their specialization (for example, according to product, process, and so on).

3. The kind of equipment and production technology used in these plants.

4. The span of the process—that is, the direction of vertical integration (toward control either of markets or of suppliers), its extent (as reflected roughly by value added as a percentage of sales), and the degree of balance among the capacities of the production stages.

Infrastructure decisions involve the following considerations:

1. Policies that control the loading of the factory or factories—raw material purchasing, inventory, and logistics policies.

2. Policies that control the movement of goods through the factory or factories—process design, work-force policies and practices, production scheduling, quality control, logistics policies, inventory control.

3. The manufacturing organizational design that coordinates and directs all of the foregoing.

These two sets of decisions are closely intertwined, of course. A plant’s total annual capacity (a facilities decision) depends on whether the production rate is kept as constant as possible over time or, alternatively, changed frequently in an attempt to “chase demand” (an infrastructure decision). Similarly, work-force policies interact with location and process choices, and purchasing policies interact with vertical integration choices. Decisions regarding organizational design also will be highly dependent on vertical integration decisions, as well as on the company’s decisions regarding how various plants are located, specialized, and interrelated.

Each of these structural decisions places before the manager a variety of choices, and each choice puts somewhat different weights on the five competitive dimensions. For example, an assembly line is highly interdependent and inflexible but generally promises lower costs and higher predictability than a loosely coupled line or batch-flow operation or a job shop. Similarly, a company that attempts to adjust production rates so as to chase demand will generally have higher costs and lower quality than a company that tries to maintain more level production and absorb demand fluctuations through inventories.

If consistent priorities are to be maintained, as a company’s strategy and manufacturing mission change, then change usually becomes necessary in all of these structural categories. Again and again the root of a manufacturing crisis is that a company’s manufacturing policies and people—workers, supervisors, and managers—become incompatible with its plant and equipment, or both become incompatible with its competitive needs.

Even more subtly, plant may be consistent with policies, but the manufacturing organization that attempts to coordinate them all no longer does its job effectively. For, in a sense, the organization is the glue that keeps manufacturing priorities in place and welds the manufacturing function into a competitive weapon. It also must embody the corporate attitudes and biases already discussed.

In addition, the way manufacturing chooses to organize itself has direct implications for the relative emphasis placed on the five competitive dimensions. Certain types of organizational structures are characterized by high flexibility; others encourage efficiency and tight control, and still others promote dependable promises.

Approaching the Design

How are the appropriate corporate priorities to be maintained in a manufacturing organization that is characterized by a broad mix of products, specifications, process technologies, production volumes, skill levels, and customer demand patterns? To answer this question, we must begin by differentiating between the administrative burden on the managements of individual plants and that on the central manufacturing staff. Each alternative approach for organizing a total manufacturing system will place different demands on each of these groups. In a rough sense, the same amount of “control” must be exercised over the system, no matter how responsibilities are divided between the two.

At one extreme, one could lump all production for all products into a single plant. This makes the job of the central staff relatively easy (in some respects it becomes almost nonexistent), but the job of the plant management becomes horrendous. At the other extreme, one could simplify the job of each plant (or operating unit within a given plant), so that each concentrates on a more restricted set of activities (products, processes, volume levels, and so on), in which case the coordinating job of the central organization becomes much more difficult.

Although many companies adopt the first approach, by either design or default, in our experience it becomes increasingly unworkable as more and more complexity is put under one roof. At some point a single large plant, or a contiguous plant complex, breaks down as more products, processes, skill levels, and market demands are added to it. Skinner has argued against this approach and for the other extreme in an article in which he advocates dividing up the total manufacturing job into a number of focused units, each of which is responsible for a limited set of activities and objectives:

“Each [manufacturing unit should have] its own facilities in which it can concentrate on its particular manufacturing task, using its own work-force management approaches, production control, organization structure, and so forth. Quality and volume levels are not mixed; worker training and incentives have a clear focus; and engineering of processes, equipment, and materials handling are specialized as needed. Each [unit] gains experience readily by focusing and concentrating every element of its work on those limited essential objectives which constitute its manufacturing task.”4

If we adopt this sensible (but radical) approach, we are left with the problem of organizing the central manufacturing staff in such a way that it can effectively manage the resulting diversity of units and tasks. It must somehow maintain the total organization’s sense of priorities and manufacturing mission, even though individual units may have quite different tasks and focuses. It carries out this responsibility both directly, by establishing and monitoring the structural policies we mentioned earlier (for example, process design, capacity planning, work-force management, inventory control, logistics, purchasing, and the like), and indirectly, by measuring, evaluating, and rewarding individual plants and managers, and through the recruitment and systematic development of those managers.

These basic duties can be performed in a variety of ways, however, and each will communicate a slightly different sense of mission. To illustrate this, let us consider two polar examples—a “product-focused organization” and a “process-focused organization.” To clarify this discussion, look at the two highly simplified organizations shown in Exhibit II and think about what the tasks of the corporate manufacturing staff and plant managers would be in each.

Exhibit II Product-Focused and Process-Focused Organization

The corporate staff clearly must play a much more active role in making the second organization work. Logistics movements have to be carefully coordinated, and a change in any of the plants (or the market) can have repercussions throughout the system.

Only at the last stage (Process C) can the plant manager be measured on a profitability basis, and even that measure depends greatly on negotiated transfer prices and the smooth functioning of the rest of the system. He will not have much opportunity to exercise independent decision making, since most variables under his control (capacity, output, specifications, and so on) will affect everybody else. Thus he will probably be regarded as a “cost center” and be measured in large part on his ability to work smoothly within this highly interdependent system.

The distinction between such product-focused and process-focused manufacturing organizations should not be confused with the distinction between traditional functional and divisional corporate organizations. In fact, it is entirely possible that two divisions within a divisionally organized company would choose to organize their manufacturing groups differently. The important distinction has less to do with the organization chart than with the role and responsibilities of the central manufacturing staff and how far authority is pushed down the organization. In a sense, the distinction is more between centralized control and decentralized control.

With this brief overview, let us turn to more realistic product and process organizations.

Product-Focused Organization

Basically, the product-focused organization resembles a traditional plant-with-staff organization, which then replicates itself at higher levels to handle groups of plants and then groups of products and product lines. Authority in the product-focused organization is highly decentralized, which contributes to the flexibility of this type of organization in new product introduction. Each product group is essentially an independent small company, and thus it can react quickly to product development considerations.

A product focus tends to be better suited to less complex, less capital-intensive process technologies, where the capital investment required is generally not high, where economies of scale do not demand large common production facilities, and where flexibility and innovation are more important than careful planning and tight control. A product focused organization is a “clean” one, with responsibilities well delineated, and profit or return on investment the primary measures. Such an organization tends to appeal most to companies that have a high need and tolerance for diversity, and whose dominant orientation is to a market or consumer group, as opposed to a technology or a material.

The responsibility for decisions on capital, technology, and product development are thrust down from the corporate level to lower levels of management. Plant managers become very important people. This places special burdens on the organization. Product focus demands talented, entrepreneurially minded junior managers and thus much concern for recruiting and managerial development. Junior managers must be tracked carefully through the system, and this implies devoting considerable resources to the company’s evaluation and reward system.

And, because staff functions are isolated in individual product lines, the corporate staff must coordinate general policies, goals, and personnel across all the product lines. The corporate level central staff is well removed from day-to-day operations, but it is instrumental in communications and coordination across groups regarding such issues as personnel policies, manpower availability, special services (from computer assistance to training programs), capital appropriation requests, and purchasing.

Process-focused Organization

Within a process-focused organization, individual plants are typically dedicated to a variety of different products. Sometimes a product is produced entirely by a single plant in such an organization, but more often the plant is only one of several that add value to the product.

Responsibilities throughout the plant and also throughout the upper management hierarchy are delineated, not by product line, but by segment of the full manufacturing process. Plants tend to be cost centers, not profit centers, and measurement is based on historical or technologically derived standards. An organization with this division of responsibility can properly be called process-focused.

Process focus tends to be better suited to companies with complex (and divisible) processes and with large capital requirements, companies we earlier called material- or technology-oriented companies. Questions of capacity, balance, logistics, and technological change and its impact on the process are critical for such companies and absorb much of top management’s energies. A process focus is not conducive to the rapid introduction of new products, since it does not assign authority along product lines. Nor is it flexible in altering the output levels of existing products, because of the “pipeline momentum” in the system. But it can facilitate low-cost production if there are cost advantages deriving from the scale, continuity, and technology of the process.

A process-focused organization demands tremendous attention to coordinating functional responsibilities to ensure smooth changes in the product mix. And, because control is exercised centrally, young managers must endure a long and generally a more technical apprenticeship with less decision-making responsibility. This places a burden on upper level management to keep junior managers motivated and learning.

Despite the strong centralization of control in a process-focused organization, it may not be more efficient (in terms of total manufacturing costs) than a well-managed product-focused organization. The central overhead and logistics costs required by a process focus can sometimes offset any variable cost reductions because of tight control and economies of scale. A product focus, however, is inherently easier to manage because of its small scale and single mindedness. This usually results in shorter cycle times, less inventories, lower logistics costs, and, of course, lower overhead.5

The plants in a process organization can be expected to undertake one task that the central staff in a product organization cannot adequately perform, however. Since these plants are technologically based, they tend to be staffed with people who are highly expert and up to date in that technology. They will be aware of technological alternatives and trends, current research, and the operating experience of different technologies at other plants. Operating people in such a plant are more likely to transfer to a similar plant of a competitor’s than they are to move to one of the other plants in their own company.

In a product organization, each product-plant complex will involve a number of technologies, and there may not be a sufficient mass of technical expertise to keep abreast of the changing state of the art in that technology. This becomes, then, more a responsibility of the corporate staff or, possibly, of a separate research group in the corporation, which may not even be under the aegis of the manufacturing organization. For this reason, businesses that use highly complex and evolving technologies are often forced to gravitate toward process organizations.

A process organization tends to manage purchasing somewhat better than a product organization does. If purchasing becomes too fragmented because of decentralization, the company as a whole tends to lose economies of scale as well as “clout” with suppliers. Conversely, centralized purchasing tends to be more bureaucratic and less responsive to local or market needs. The result is usually a combination of both, where through some decision rule the product organizations are given responsibility for certain purchases and a central purchasing department handles the procurement and distribution of the remainder.

Exhibit III gives a summary of the important differences between product-focused and process focused organizations.

Exhibit III Differences Between Product-Focused and Process-Focused Manufacturing Organizations

Product or Process Focus?

The polar extremes of manufacturing organization—product and process focus—place fundamentally different demands and opportunities on a company, and the choice of manufacturing organization should essentially be a choice between them. That is, manufacturing confronts a very definite either/or choice of organization—either product-focused or process-focused. Just as individual plants must have a clear focus, so must a central manufacturing organization.

Because the demands of a process-focused organization are so different from those of a product-focused organization—as to policies and practices, measurement and control systems, managerial attitudes, kinds of people, and career paths—it is extremely difficult for a mixed manufacturing organization, with a single central staff, to achieve the kind of policy consistency and organizational stability that can both compete effectively in a given market and cope with growth and change.

A mixed or composite production focus will only invite confusion and a weakening of the corporation’s ability to maintain consistency among its manufacturing policies, and between them and its various corporate attitudes. If different manufacturing groups within the same company have different focuses, they should be separated as much as possible—each with its own central staff.

To illustrate, we can examine some mixed organizational focuses and the difficulties they might encounter.

• A process-focused factory producing for two distinct product groups would have the organization chart shown in Exhibit IV. Here the corporation is trying to serve two different markets and product lines from the same factory, whose process technology appears to meet the needs of both (it may, in fact, consist of a series of linked process stages operating under tight central control). This kind of organization invites the now classic problems of Skinner’s unfocused factory. The manufacturing mission required by each market may be vastly different, and a plant that tries to carry out both at the same time is likely to do neither well.

Exhibit IV Process-Focused Factory Serving Two Different Product Markets

Similarly, an organization that uses the manufacturing facilities of one of its product groups to supply a major portion of the needs of another product group market would be risking the same kind of confusion—that is, a nominally product-focused organization with an organization chart like the one in Exhibit V.

Exhibit V One Product Group Serving Another Product Group’s Market

• A process-focused factory supplying parts or materials to two distinct product groups would have the organization chart shown in Exhibit VI. In this instance a corporate staff oversees two independent product groups, which serve two distinct markets, and a process-focused plant that supplies both product groups. The usual argument for an independent supplier plant is that economies of scale are possible from combining the requirements of both product groups. No matter what the reason, the supplier plant is coordinated by the same staff that oversees the product groups. One vice president of manufacturing directs a corporate manufacturing staff with one materials manager, one chief of individual engineering, one head of purchasing, one personnel director—all supervising the activities of two product-focused organizations and a process-focused organization.

Exhibit VI Two Product Groups and a Supplier Plant

Another variant of this difficulty is for the captive supplier plant for one product group to supply a major portion of the requirements of another product group’s plant. Or a plant belonging to a product-focused division might act as a supplier to one of the plants within a process-focused division.

How else can a company organize around such situations? The important notion is that a plant that attaches certain priorities to different competitive dimensions is likely to prefer suppliers who have the same priorities. This suggests that a company should erect managerial dividing lines between its product- and process-focused manufacturing segments. In particular, transfer of products between product- and process-focused plant groups should not be coordinated by a central staff group but handled through arm’s-length bargaining, as if, in effect, they had independent “subsidiary” relationships within the parent company.

Such an in-house supplier would then be treated like any other supplier, able to resist demands that violate the integrity of its manufacturing mission just as the customer plant is free to select suppliers that are more attuned to its own mission. The organization chart might look something like that shown in Exhibit VII.

Exhibit VII Product-Focused Division Supplying to Process-Focused Division

Such an arrangement may appear to be needlessly complex and add to the manufacturing’s administrative overhead without clear financial benefits. However, combining two dissimilar activities does not reduce complexity; it simply camouflages it and is likely to destroy the focus and distinctiveness of both. Our position is not that both product and process focus cannot exist within the same company but simply that separating them as much as possible will result in less confusion and less danger that different segments of manufacturing will be working at cross purposes.

Test for Organizational Focus

Many companies, consciously or unconsciously, have moved toward precisely this kind of wide separation. In some cases it is explicit, with two or more different staff groups operating relatively autonomously; in others, although a single central staff appears on the organization chart, subgroups within this staff operate independently. One way for a company to test the degree of organizational focus in its manufacturing arm, and whether adequate insulation between product- and process-focused plant groups exists, is to contemplate how it would fragment itself if forced to (by the Antitrust Division of the Department of Justice for example). A segmented and focused organization should be able to divide itself up cleanly and naturally, with no substantial organizational changes.

Consider the large auto companies. From the point of view of the marketplace, they are organized by product groups (Oldsmobile, Lincoln, Mercury, Chevrolet, and so on), but this organization is essentially cosmetic. In reality, the auto companies are classic examples of large process-focused organizations. Any effort by the Department of Justice to sever these companies by product group is foolish because it cuts across the grain of their manufacturing organization. If the companies had to divest themselves, it could only be by process segment. But the point is that divestiture could be accomplished readily, and this is the acid test of an effective and focused manufacturing organization.

The Impact of Growth

Up to this point we have been arguing that a company’s manufacturing function must structure and organize itself so as to conform to the company’s priorities for certain competitive dimensions. Moreover, the choice of manufacturing organizational structure—which provides most of the key linkages between the manufacturing group and the company’s other people and functions—must also fit with the basic attitudes, the preferences, and the traditions that shape and drive the remainder of the company.

But companies change and grow over time. Unless a manufacturing organization is designed so that it can grow with the company, it will become increasingly unstable and inappropriate to the company’s needs. Therefore, simplicity and focus are not sufficient criteria; the organizational design must somehow also incorporate the possibility of growth.

In fact, growth is an enemy of focus and can subvert a healthy manufacturing operation—not all at once, but bit by bit. For example, growth can move a company up against a different set of competitors at the same time it is acquiring new resources and thus force a change in its competitive strategy.

The strategy change may be aggressive and deliberate or unconscious and barely perceived. In either case, however, success for the company may now require different skills from those already mastered—a different manufacturing mission and focus to complement a new corporate strategy.

Even without a change of strategy, growth can diminish a manufacturing organization’s ability to maintain its original focus. Especially if growth is rapid, top-level managers will be pressed continually to decide on capital acquisitions and deployment, and to relinquish some authority over operational issues in existing plants. Slowly, focus disintegrates.

To cope with growth, we believe that first one must identify and understand the type of growth being experienced and the demands it will place on the organization. Growth has four important dimensions:

1. A broadening of the products or product lines being offered.

2. An extended span of the production process for existing products to increase value added (commonly referred to as vertical integration).

3. An increased product acceptance within an existing market area.

4. Expansion of the geographic sales territory serviced by the company.

These types of growth are very different, but it is important to distinguish among them so that the organization design can reflect the kind of growth experienced, not simply the fact of growth. This means keeping the organization as stable and focused as possible as growth proceeds.

If growth is predominantly a broadening of product lines, a product-focused organization is probably best suited to the demands for flexibility that such a broadening requires. With such organizations, other aspects of manufacturing, particularly the production of the traditional product lines, need change only little as growth proceeds.

Alternatively, if growth is chiefly toward increasing the span of the process (that is, vertical integration), a process-focused organization can probably best introduce and manage the added segments of the full production process. In this fashion, the separate pieces of the process can be coordinated effectively and confusion can be reduced in the traditional process segments.

Then again, if growth is realized through increased product acceptance, the product becomes more and more a commodity and, as acceptance grows, the company is usually pressed to compete on price. Such pressure generally implies changes in the production process itself: more specialization of equipment and tasks, an increasing ratio of capital to labor expenses, a more standard and rigid flow of the product through the process. The management of such changes in the process is probably best accomplished by an organization that is focused on the process, willing to forsake the flexibilities of a more decentralized product focus.

Growth realized through geographic expansion is more problematic. Sometimes such growth can be met with existing facilities. But frequently, as with many multinational companies, expansion in foreign countries is best met with an entirely separate manufacturing organization that itself can be organized along either a product or a process focus.

Recognizing Common Pitfalls

As we examined a number of manufacturing organizations that had “lost their way”—become unfocused or whose focus was no longer congruent with corporate needs—it became apparent that in most cases the culprit was growth. Problems due to growth often surface with the apparent breakdown of the relationship between the central manufacturing staff and division or plant management. For example, many companies that have had a strong central manufacturing organization find that as their sales and product offerings grow in size and complexity, the central staff simply cannot continue to perform the same functions as well as before. A tenuous mandate for changing the manufacturing organization surfaces.

Sometimes, product divisions are broken out. But the natural inclination is to strengthen the central staff functions instead, which usually diminishes the decision-making capabilities of plant managers.

As the central staff becomes stronger, it begins to siphon authority and people from the plant organization. Thus the strong tend to get stronger and the weak weaker. At some point this vicious cycle breaks down under the strain of increasing complexity, and then a simple executive order cannot accomplish the profound changes—in people, policies, and attitudes—that are necessary to reverse the process and cause decentralization.

We do not mean to imply that decentralizing manufacturing management is always the best path to follow as an organization grows. It may be preferable in some cases to split it apart geographically, with two strong central staffs coordinating the efforts of two independent plant organizations.

However, it is sometimes dangerous to delegate too much responsibility for capacity-expansion decisions to a product-oriented manufacturing manager. To keep his own task as simple as possible, he may tend to “expand in place”—continually expanding current plants or building nearby satellite plants. Over time he may create a set of huge, tightly interconnected plants that exhibit many of the same characteristics as a process organization: tight central control, inflexibility, and constraints on further incremental expansion.

Such a situation could occur in spite of the fact that the corporation as a whole continues to emphasize market flexibility, decentralized responsibility, and technological opportunism. The new managers trained in such a complex will have to be different in personality and skills from those in other parts of the company, and a different motivation and compensation system is required. Such a situation can be remedied either by dismembering and reorganizing this product organization or by decoupling it from the rest of the company so that it has more of an independent, subsidiary status, as described earlier.

Product focus can also encroach on an avowed process focus. For example, a company offering several complex products whose manufacture takes these products through very definite process stages, in which the avowed focus is process-oriented, and with separate divisions for stages of the process all subject to strong central direction, must resist the temptation to alter manufacturing so that it can “get closer to the market.” If the various product lines were allowed to make uncoordinated requests for product design changes or new product introductions, the tightly coupled process pipeline could then crumble. Encroaching product focus would subvert it.

Concluding Remarks

Manufacturing functions best when its facilities, technology, and policies are consistent with recognized priorities of corporate strategy. Only then can manufacturing gain efficiency without wasting resources by “improving” operations that do not count.

The manufacturing organization itself must be similarly consistent with corporate priorities. Such organizational focus is aided by simplicity of design. This simplicity in turn requires either a product- or a process-focused form of organization. The proper choice between these two organizational types can smooth a company’s growth by lending stability to its operations.

Manufacturing—Weapon or Millstone?

A company’s manufacturing function typically is either a competitive weapon or a corporate millstone. It is seldom neutral. The connection between manufacturing and corporate success is rarely seen as more than the achievement of high efficiency and low costs. In fact, the connection is much more critical and much more sensitive. Few top managers are aware that what appear to be routine manufacturing decisions frequently come to limit the corporation’s strategic options, binding it with facilities, equipment, personnel, and basic controls and policies to a noncompetitive posture which may take years to turn around.

1. See, for example, Wickham Skinner, “Manufacturing—Missing Link in Corporate Strategy,” HBR May–June 1969, p. 136, and “The Focused Factory,” HBR May–June 1974, p. 113.

2. Two representative texts are: Kenneth R. Andrews, The Concept of Corporate Strategy (Homewood, Ill.: Dow Jones-Irwin, 1971), and H. Igor Ansoff, Corporate Strategy (New York: McGraw-Hill, 1965).

3. “Hewlett-Packard: Where Slower Growth Is Smarter Management,” Business Week, June 9, 1975, p. 50.

4. See Wickham Skinner, “The Focused Factory,” HBR May–June 1974, p. 121.

5. E.F. Schumacher has eloquently argued a similar point in a somewhat different context in his provocative book Small Is Beautiful (New York: Harper & Row, 1975).

A version of this article appeared in the January 1978 issue of Harvard Business Review.