Three Perspectives on
Organizations
From: Module 2
MANAGING FOR THE FUTURE:
ORGANIZATIONAL PROCESSES AND BEHAVIOR
(Southwest Publishing: 3rd edition, 2004)
Deborah Ancona
Tom Kochan
John Van Maanen
Eleanor Westney
Perspective 1
THE ORGANIZATION AS STRATEGIC DESIGN
From Max Weber’s discussion of “machine bureaucracy” at the turn of the century to the customer-oriented designs of the early twenty-first century, the dominant perspective on organizations has viewed them as strategic designs: that is, as systems deliberately constructed to achieve certain strategic goals. This perspective asserts that by understanding basic principles of organization design, by aligning the organization’s design with its strategy, and by making sure that both strategy and design fit the environment in which the organization is operating, managers can make their organizations successful. The strategic design perspective emphasizes the efficiency and effectiveness of the organization. Efficiency involves accomplishing strategic goals with the least possible expenditure of resources; effectiveness involves ensuring that goals are accomplished to the standard necessary for the organization to succeed.
As the term “strategic design” implies, this perspective on organization is built on the assumption that the organization has a strategy for creating that provides the test for generating and assessing the organization’s design. This value proposition or distinctive competitive advantage establishes what activities the organization must carry out to achieve success in its strategies: for example, providing speedy and reliable servicing of products to customers, getting products from R&D to the market quickly, continuously driving down production costs through continuous improvement. But the activities at which the organization excels can (and should) influence the strategy. For example, an organization with an extensive marketing organization might be better off using that organization to get fast and detailed information on customer needs in order to adopt a strategy of providing premium-priced products and after-sales service, rather than adopting a low-cost strategy that involves cutting back on its full-time marketing personnel.
People working with this perspective often use metaphors of the organization as a mechanism or system, of “engineering” or “re-engineering” the organization, of organization-building and “organizational architecture”, of the organization as a complex organism that can be “diagnosed” like a medical patient. Like engineering, architecture, or medicine, management is seen as a matter of understanding and applying basic principles and processes, and adapting them to the context in which one is operating.
KEY ELEMENTS OF ORGANIZATION DESIGN
One way to think about the “strategic design” perspective is that it involves simultaneously drawing boundaries around clusters of tasks or activities (to define jobs, departments, processes) -- strategic grouping (differentiation) -- and then creating links across those boundaries -- strategic linking (integration). In addition, it involves aligning other elements of the organization (such as rewards and incentives) so that each part of the organization has the resources and the incentives to do the tasks it is assigned by the grouping and linking. The basic assumption of the strategic design perspective is that an organization is most effective when its strategy fits the conditions of its environment and when the organizational components are aligned with the strategy and with each other.
The basic element of organization design is often seen as the task: the smallest unit of the activities that need to be performed if the organization is to realize its strategic goals. Tasks vary in complexity, from the relatively simple, like inserting circuit boards into the CPU of a personal computer on an assembly line, to more complex, such as reprogramming an industrial robot on an assembly line, to the extremely complex, like setting up a new business division to develop, produce, and sell industrial robots. Tasks also vary in the level of routinization: that is, the extent to which the activity can be specified and programmed. Usually simple tasks are more routine, but even complex tasks can also be routinized: for example, the analysis of how software is designed and the breaking of what had been seen as a complex “art” or task into discrete programming steps has led to the creation of “software factories” with high levels of efficiency in producing certain kinds of software programs.
In addition, tasks vary in the nature of their interdependence. Some tasks are highly independent of other tasks, and can be performed quite separately from others, with very little linking across them. Most tasks that are incorporated into an organization however, involve some level of interdependency. Task interdependence at its simplest varies from low to high, but it can also be seen in terms of the kind of interdependence. James Thompson (1967) developed a highly influential typology of task interdependence, identifying three different types: sequential interdependence, when one task is completed and then handed off for the next stage; pooled interdependence, when interdependent tasks are undertaken at the same time, and the final results that put together or pooled; and reciprocal interdependence, when tasks are conducted in repeated interaction with each other (see Figure 2A-1). To turn once again to software engineering for illustrations, sequential interdependence can be seen when the development of a software program is divided into distinct stages or phases, with “milestones” to mark the completion of one stage and the hand-off to the next. Pooled interdependence occurs when different groups work on modules of a program that are then put together to form the final program. And reciprocal interdependence characterizes complex programs where the different tasks involved in developing the program are carried on in dense interaction with each other, because the solutions to problems in one element of the program affect the solutions that can be implemented in others. Pooled interdependence is the easiest to manage: once the task assignments are defined, each unit can proceed without detailed information exchange with other units. Sequential interdependence is harder to manage than pooled, because information flows are quite dense at the point of transfer. And it is in turn easier to manage than reciprocal interdependence, which involves sustained and interactive linkages across the boundaries of the units.
Figure 2A-1 about here
As these examples show, many complex tasks can be taken apart into simpler activities. Nearly a century ago Frederick Taylor developed what came to be called “Scientific Management”, using time and motion studies to analyze the most efficient set of movements needed to perform industrial tasks, such as shoveling coal in a steel mill. Then he re-aggregated the movements into a Standard Operating Procedure that defined the job. Nearly one hundred years later, this same approach of the disaggregation and careful analysis of tasks underpins Total Quality Management and Business Process Re-engineering (TQM analyzes each step in a process to identify and correct the sources of problems; BPR analyzes the steps in delivering a product or service to the customer to eliminate unnecessary activities and then recombines tasks into more effective and efficient jobs and subunits).
Organizational design choices begin with strategic grouping: that is, the differentiation of clusters of activities, positions, and individuals into work units. Once activities are divided into “boxes”, however, the units must be linked according to the nature and level of interdependence in their tasks, to ensure that information and other needed resources flow effectively and efficiently between the activities or groups separated by the grouping boundaries (linking). In other words, creating boundaries in organizations (differentiation) (integration)creates the need for linking across boundaries. Finally, the design must use a variety of alignment mechanisms (incentive systems, information systems, etc., discussed below) to ensure that people have the resources and the incentives to carry out the tasks assigned to them by the grouping structures and the linking mechanisms.
Strategic grouping, linking, and alignment are relevant at every level of the organization, from the design of teams or departments to the overall design of the organization.
STRATEGIC GROUPING
Grouping decisions dictate the basic framework within which all other organizational design decisions are made. Grouping gathers together some tasks, functions, or disciplines, and separates them from others. It is (or ought to be) a direct outgrowth of the strategy of the organization as a whole and the associated strategy of the particular organizational unit. A fundamental assumption of organization design is that coordination and communication are easier and denser within a unit than across units. This means that sharing information and building and adding to a common knowledge base are easier within than across units. Therefore the most important areas of interdependence should be under a unified reporting structure. The strategy should guide the design by identifying the most strategically important parameters of coordination, interdependence, and knowledge sharing
Strategic grouping focuses on questions of how to cluster tasks and activities. Should people performing the same kinds of tasks in similar ways be clustered together, or should people performing complementary tasks be grouped together? How many subgroups should be created, and in how many layers? Grouping can be seen as “drawing the boxes” of the organization design. There are three basic forms or “ideal types” of grouping, each with a distinct set of strengths and weaknesses.
BASIC STRUCTURES
Grouping by Expertise/Function
Grouping by function (or more broadly, by expertise) brings together individuals who share similar functions, disciplines, skills, and work processes. At the level of the organization, grouping by activity gives rise to functional organizations, the oldest form of business enterprise and one which is still often the form first adopted in new organizations. In the functional organization, all the activities concerned with a particular function are grouped into separate divisions, as shown in Figure 2A-2.
Figure 2A-2 about here
Grouping by function has three major strengths. The first is that it allows the development of deep functional expertise and a high degree of specialization of knowledge within each function. These organizations can be extremely innovative in specific technologies or functions. The second strength is what economists call “economies of scope”: that is, the functional organization makes it relatively easy to transfer resources across activities within functions. If sales of one product decline, for example, the manufacturing division can switch a production line or a factory to a product that is in greater demand; engineers can be switched across projects; marketing divisions can reallocate their sales and support efforts. Finally, a functional organization allows each group to create separate incentive and control systems suited to its needs and to reinforcing its strengths.
However, these advantages come at the cost of integration across functions. This model often assumes a sequential interdependence across the functions, from upstream (R&D) to downstream (marketing). Functional organizations are often not very responsive to changes in markets or customers. Moreover, as their level of specialization increases, individuals tend to develop narrower perspectives, and have difficulty in solving problems that require joint efforts with other groups. It can be difficult in this kind of organization to assess costs clearly, especially on a product line basis, and to assign accountability for the overall performance of the organization. Finally, because career ladders are primarily within functions and the number of functions is limited, there is a tendency for the number of levels of management in each function to expand over time. Large functional organizations often have very tall hierarchies that can inhibit speedy and effective information flows. Moreover, because there are few opportunities for managers to gain experience outside their functional areas as they move up the career ladder, the functional organization does not develop a large supply of general managers who can see the organization as a whole and make decisions that serve the organization’s strategy instead of following a specifically functional logic.
A functional organization is frequently adopted by new organizations and maintained over time by organizations that have a single major business, or several businesses that share the same technologies and have very similar markets.
Grouping by Output/Product
This structure organizes on the basis of the service or product provided. The people within the group perform a variety of different tasks and activities, but they are all contributors to the same final output (a product or set of closely related products or services). For example, firms with a range of product lines and markets usually find the functional organization too inflexible and instead adopt a product line or multidivisional structure. In this design, pioneered in the U.S. by DuPont and General Motors (Chandler, 1962) and in Japan by Matsushita, the functions are distributed across the business or product line they support, as shown in Figure 2A-3.
Figure 2A-3 about here
The Product Division structure has two major advantages. First is transparency of performance: it makes the costs and profits of each business much clearer than does the functional structure. Second is the clear strategic focus it provides for the managers of the product division. These strengths made the multidivisional structure the dominant form of business enterprise for most large American corporations in the 1980s.
On the other hand, it is not without disadvantages. Making each business unit accountable for its own “P-and-L” (profit and loss statement) can make it difficult for units to share resources and can lead to duplication of activities (each unit wants its own accounting staff, its own training staff, and so on). It can also make new business creation difficult: in some cases business units focus exclusively on expanding their existing business, instead of finding new opportunities; in others, units compete to “own” new business. And because functional specialists are spread across different groups, they can lose their professional focus and become less attuned to breakthrough innovations in their own fields (and this holds true across functions, from R&D to accounting). Distributing activities across business units can also lead to missed learning opportunities in core functions (for example, factories in one business unit may be unaware of potentially useful innovations in another business unit’s factories). Many companies spent the 1990s trying to break down their “functional silos” by setting up Business Divisions only to discover that they had created “business unit silos” that were just as internally competitive and resistant to developing potential cross-boundary synergies as the old functional organization had been.
Grouping by Market (Geography or Customer)
Market-oriented companies often adopted this structure, which gathers together people who perform different activities and tasks and produce different outputs but who serve the same customers or market segments. The most common dimension is geography. Large multinational companies operating around the world, for example, historically favored a geographic organization, grouping by geographic region and country (see Figure 2A-4).
Figure 2A-4 about here
Even domestic companies have often grouped by geography: by sales territories, for example. But grouping by the type of customers rather than their location is also an variant of this form: publishers, for example, are often organized on the basis of what customer group the division serves: Textbooks (often subdivided by educational level -- Primary, Secondary, College), Business and Professional Books, Mass Market, etc.
The strength of this structure is its capacity for developing deep customer knowledge and close customer relationships, and therefore it is often found in service industries. It allows the organization to tailor its products and services to differentiated customer needs. The weaknesses are similar to those of the product division structure: the duplication of activities and resources, the erosion of deep technical expertise, missed opportunities for synergies and learning.
There is no universally ideal choice of grouping pattern. Each strategic grouping option comes with its own set of strengths and weaknesses. And every organization design must address all three elements of activity/function, business/product, and geography/customer. The critical grouping question is which dimension will be primary, and how the others will be nested at the next levels: function within business within geography, for example, or function within geography within business.
HYBRID STRUCTURES
Some companies have found that grouping on any single dimension -- function, business, geography -- is inadequate. Corporate strategies frequently require attention to multiple priorities simultaneously -- product and function, for example, or customers and technical expertise. Many organizations have turned to hybrid or multi-dimensional grouping structures in an attempt to break out of the constraints imposed by a single mode of strategic grouping. Two of the most common hybrids today are matrix organizations and the somewhat awkwardly named “front end/back end” structure.