The purpose of a plan is to specify a desired output—a standard—at some future time. And the purpose of control is to assess whether or not that standard has been achieved. Thus, planning and control go together like the proverbial horse and carriage: There can be no control without prior planning, and plans lose their influence without follow-up controls. To- gether plans and controls regulate outputs and, indirectly, behavior as well.
Plans may specify (standardize) the quantity, quality, cost, and tim- ing of outputs, as well as their specific characteristics (such as size and color). Budgets are plans that specify the costs of outputs for given periods of time; schedules are plans that establish time frames for outputs; objectives are plans that detail output quantities for given periods of time; operating plans are those that establish a variety of standards, generally the quantities and costs of outputs. Typically, planning systems, as well as the reporting systems that feed back the control information, are designed in the tech- nostructure, by analysts with titles such as Planner, Budget Analyst, Con- troller, MIS Analyst, Production Scheduler, and Quality Control Analyst.
We can distinguish two fundamentally different kinds of planning and control systems, one that focuses on the regulation of overall perfor- mance and the other that seeks to regulate specific actions. Since the for- mer is concerned primarily with after-the-fact monitoring of results, we shall call it performance control. The latter, oriented to specifying activities that will take place, is labeled action planning. In other words, as shown in Figure 4-1, the organization can regulate outputs in two ways. It can use performance control to measure the results of a whole series of actions, and use this information to make changes: “The profit rate should increase from 7 percent to 10 percent,” or, “The drilling of holes should be in- creased from fifty to sixty per day.” Alternatively, it can use action plan- ning to determine in advance what specific decisions or actions are re- quired: “Blue widgets should be sold to customers X, Y, and Z,” or, “The hole should be drilled 1.108 centimeters wide.” As we shall see, whereas performance control is a pure means of standardizing outputs, action plan- ning—because it specifies particular actions—resembles in some ways the design parameter of formalization of behavior.
Figure 4-1. The relationships between decisions and ac- tion planning and performance control
1. Performance control
The purpose of performance control is to regulate the overall results of a ‘ given unit. Objectives, budgets, operating plans, and various other kinds of general standards are established for the unit, and its performance is later measured in terms of these standards and the results fed back up the hierarchy by the MIS. This suggests two important points: First, perfor- mance control systems map onto the bases for grouping in the organiza- tion. The planning system establishes output standards for each unit, and the control system assesses whether or not these have been met. Second, performance control is concerned with overall results for given periods of time, not with specific decisions or actions at specific points in time. For example, a performance plan may call for the production of 70,000 widgets in June, or the reduction of costs by 3 percent in July; it does not call for the shift from blue widgets to green ones or the achievement of cost reduction by the purchase of a more efficient machine. Thus, performance control influences decision making and action taking only indirectly, by establish- ing general targets that the decision maker must keep in the back of his mind as he makes specific decisions in the front.
Where is performance control used in the organization? To some extent, everywhere. Because cost control is always crucial and because costs—at least economic ones—are easily measured, virtually every organi- zational unit is given a budget—that is, a performance plan to standardize its expenditures. And where the unit’s production is easily measured, its performance plan will typically specify this as well. The plant is expected to produce 400,000 widgets this month; marketing is expected to sell 375,000 of them.
But performance control systems are most relied upon where the interdependencies between units are primarily of a pooled nature—name- ly, where the units are grouped on the basis of market. Here, the major concern is that the unit perform adequately, that it make an appropriate contribution to the central organization without squandering its resources. In other words, because there is little interdependence between units, coordination requires the regulation of performance, not actions. And this is facilitated in the market-based structure by the fact that each unit has its own distinct outputs. Thus, its overall behavior is regulated by perfor- mance controls; otherwise, it is left alone to do its own action planning.
Indeed, such performance controls are typically crucial for market- based units. Because they are self-contained, they are generally given con- siderable freedom to act. Typically, as noted in the last chapter, a great many such units report to a single manager. Without a performance con- trol system, the manager may be unable to catch serious problems, until it is much too late. A wayward Sears store or Bank of America branch could, for example, get lost for years, too small a part of the organization to be otherwise noticed. And from the perspective of the market unit itself, the performance control system serves to preclude direct supervision and so to grant it the freedom it needs to determine its own decisions and actions. Thus, the conglomerate corporation sets up each of its market units (its “divisions”) as a profit or investment center, and holds it responsible for its own financial performance.1
Performance control systems can serve two purposes: to measure and to motivate. On the one hand, they can be used simply to signal when the performance of a unit has deteriorated. Higher-level management can then step in and take corrective action. On the other hand, they can be used to elicit higher performance. The performance standards are the carrots that management places before the unit manager to motivate him to achieve better results. Whenever he manages a nibble, the carrot is moved a little farther out and the manager runs faster. Systems such as management by objectives (MBO) have been developed to give unit managers a say in the establishment of these standards, so that they will be committed to them and therefore, the theory goes, strive harder to achieve them.
But this motivational aspect introduces a variety of problems. For one thing, given the right to participate in the setting of performance stan- dards, the unit manager has a strong incentive to set the standards low enough to ensure that they can easily be met. And he also has an incentive to distort the feedback information sent up the MIS to make it look as though his unit has met a standard that it, in fact, missed. Second is the problem of choosing the planning period. There is, as noted, no direct link between the performance standards and specific decisions taken; it is only hoped that the manager will bear the standards in mind when he makes decisions. Long planning periods loosen the connection, and short ones defeat a prime purpose of the system—to give the manager freedom of action. The “flash reports” on the tenth of every month used by some corporations certainly keep the manager hopping after short-term results. But do they let him think beyond thirty days? The third problem of moti- vation arises with standards that cannot be realized for reasons beyond the manager’s control—say, the bankruptcy of a major customer. Should the organization insist on honoring the agreement to the letter, and penalize the manager, or should it overrule the performance control system, in which case the system loses a good deal of its motivational punch?
2. Action planning
As we have seen, performance control is a key design parameter in market- based structures. But what happens in functional structures? Functional work flows sequentially or reciprocally across them. This means that dis- tinct organizational goals cannot easily be identified with any one unit. So aside from budgets and the like to control expenditures, performance con- trol systems cannot really cope with the interdependencies of functional units. Other means must be found.
As we saw in Chapter 3, direct supervision effected through the superstructure and standardization of work processes effected through behavior formalization emerge as key mechanisms to coordinate the work in functional structures. These are preferred because they are the tightest available coordinating mechanisms. But sometimes they cannot contain all the interdependencies. And so the organization must turn to planning and control systems to standardize outputs, specifically, to action planning. Simon provides a dramatic example of what can happen when action plan- ning fails to coordinate the remaining work-flow interdependences:
In the first portion of the Waterloo campaign, Napoleon’s army was divided in two parts. The right wing, commanded by the Emperor himself, faced Blucher at Ligny; the left wing, under Marshal Ney, faced Wellington at Quatre Bras. Both Ney and the Emperor prepared to attack, and both had prepared excellent plans for their respective operations. Unfortunately, both plans contemplated the use of Erlon’s corps to deliver the final blow on the flank of the enemy. Because they failed to communicate these plans, and because orders were unclear on the day of the battle, Erlon’s corps spent the day marching back and forth between the two fields without engaging in the action on either. Somewhat less brilliant tactical plans, coordinated, would have had greater success. (1957:193)
Two points should be noted about action planning. First, unlike per- formance control, action planning does not necessarily respect unit auton- omy, nor does it necessarily map onto the system of grouping. Action plans specify decisions that call for specific actions—to market new prod- ucts, build new factories, sell old machines. Some of the proposed actions may be taken within single units, but others can cut across unit boundaries.
Second, by its imposition of specific decisions, action planning turns out to be a less than pure form of standardizing outputs; more exactly, it falls between that and standardizing work processes. This point can be expressed in terms of a continuum of increasingly tight regulation, as follows:
- Performance control imposes general performance standards over a period of time, with no reference to specific actions.
- Action planning imposes specific decisions and actions to be carried out at specific points in time.
- Behavior formalization imposes the means by which decisions and actions are to be carried out.
So whereas performance control says, “Increase sales by 10 percent this year [in any way you care to],” action planning says, “Do it by intro- ducing blue widgets.” It, too, specifies outputs, but in a way that con- stitutes the specification of means. At the limit, action planning becomes behavior formalization—namely, the specification of the work flow: “.. . the plan may control, down to minute details, a whole complex pat- tern of behavior. The completed plan of the battleship will specify the design of the ship down to the last rivet. The task of the construction crew is minutely specified by this design” (Simon, 1957:231).
Action planning emerges as the means by which the nonroutine decisions and actions of an entire organization, typically structured on a functional basis, can be designed as an integrated system. All this is accomplished in advance, on the drawing board so to speak. Behavior formalization designs the organization as an integrated system too, but only for its routine activities. Action planning is its counterpart for the nonroutine activities, for the changes. It specifies who will do what, when, and where, so that the change will take place as desired.
3. The hierarchy of action planning and performance control systems
How do these two planning and control systems relate to the superstruc- ture and to each other? Figure 4-2 shows performance control and action planning as two separate hierarchical systems, with certain “crossovers” between them. Performance control is shown as a system in which overall objectives at the top give rise to subobjectives, budgets, and other output standards, which in turn are elaborated into ever more detailed subobjec- tives, budgets, and standards until they emerge at the bottom of the struc- ture as operating plans. The final outcome is, of course, organizational actions, but the connection between the plans and the actions is shown as a series of dotted lines to indicate that it is only indirect.
The arrows in the diagram are two-sided, to indicate that the perfor- mance control system may be not only top-down—where objectives decided at the strategic apex are elaborated into ever more detailed performance standards as they pass down the hierarchy—but also bottom-up, where the units at the bottom establish their own performance standards, and these are then aggregated up the hierarchy by unit, until they emerge at the strategic apex as composite standards—in effect, objectives for the whole organization. In actual practice, however, we would expect the performance control system to function most commonly, not in a purely top-down or bottom-up manner, but in a combination of the two. Some performance standards are elaborated down the hierarchy and others are aggregated up it; at each level, managers seek to impose standards on their employees, who propose less stringent ones instead. Through this kind of bargaining, there emerges a set of performance standards at all levels, composite and detailed.
Fiaure 4-2. Hiierrarrchvy of planning and control svstems
The action planning system is essentially top-down. In theory, it begins with strategic planning, wherein the organization systematically assesses its strengths and weaknesses in terms of trends in the environ- ment, and then formulates an explicit, integrated set of strategies it intends to follow in the future. These strategies are then developed into “pro- grams,” that is, specific projects—such as introducing a new product line, building a new factory, reorganizing the structure. These programs are, in turn, elaborated and scheduled, and eventually emerge as a set of specific operating specifications—to call on a customer, pour concrete, print an organigram—which evoke specific actions directly.
As shown in Figure 4-2, these two systems can be linked. At the top line a), there is a crossover from performance objectives to strategic plans. According to the conceptual literature, the whole action planning process must begin with the specification of the overall objectives of the organiza- tion: it is believed that only with a knowledge of what the organization wants—operationalized in quantitative terms—can strategic plans be gen- erated. The crossover from subobjectives or budgets to strategic plans (line b) is similar. Where there is unit autonomy, as in market-based structures, the strategic apex may develop overall objectives and then negotiate sub- objectives and budgets with each of the units. These then become the objectives that initiate the action planning process in each unit.
A crossover also takes place from subobjectives and budgets to pro- grams directly, shown by line c. This is more common in a functional structure, where a budget given to a department evokes specific programs rather than overall strategies. Thus, when the research department is told that its budget will be increased by $300,000 next year, it proceeds with plans to build the new laboratory it has been wanting.
The last crossover (line d) runs from programs to budgets and eventually to operating plans. This reflects the fact that the unit must assess the effect of all its proposed actions—the products to be marketed, machines to be bought, and so on—on its flow of funds (its budgets), the subobjectives it can reach, the manpower it must hire, and so on. In other words, the effect of specific actions on overall results must be assessed, hence the crossover from action planning to performance control.
Another crossover—perhaps the most important one, but not shown because of the nature of our diagram—is the overall feedback from perfor- mance control to action planning. As the organization assesses its perfor- mance, it initiates new action plans to correct the problems that appear.
4. Planning and control systems by part of the organization
Various forms of both action planning and performance control can be found at all levels of the hierarchy. In the case of the former, we have strategic planning and capital budgeting at the strategic apex and upper levels of the middle line, programming and PERT or CPM scheduling techniques at the middle levels, and production scheduling at the level of the operating core. In the case of performance control, we have already seen that objectives, budgets, and standards can be set for units and posi- tions at any level, from the strategic apex to the operating core. At the top is the setting of overall organizational objectives; high up in the middle line are commonly found the financial reporting systems that treat major mar- ket units as profit or investment centers; elsewhere in the middle line are the standard costing systems to control aggregated performance and MBO systems to motivate line managers; and near the bottom, we find the operating plans and quality control systems.
However, our discussion also made clear that there are important differences by part of the organization. For example, although perfor-mance control can be used for individual positions—as when salespeople are given quotas, or machine operators quality control standards—we would expect it to be more commonly applied to units (and, of course, to the managers who supervise those units). Not so for action planning. We would expect action planning to apply to individual operators, as when a machinist is given specifications for the products he is to make.
Higher up in the hierarchy, we would expect the situation to be reversed. The more global the responsibilities of a unit, the greater the propensity to control its overall performance rather than its specific ac- tions. For market-based units, as noted earlier, the performance control system is a critical device for control, whereas action planning is not. And since, as noted in the last chapter, the market basis for grouping is more common at higher than at lower levels in the structure, we find another reason why performance control would be favored over action planning in the upper reaches of the middle line. Of course, action planning systems may also be used at these levels where the basis for grouping is functional. As for the strategic apex, should it be subject to outside control (say, by a single owner), it may also have to respond to a performance control sys- tem. And if the basis for grouping the highest-level units is functional, then action planning may very well start right in the strategic apex.
Even though the technostructure is largely responsible for the design of all these planning and control systems, that does not mean that its own work is regulated by them. In fact, owing to the difficulty of standardizing the outputs of analytic work—activity that is usually carried out on a pro- tect or ad hoc basis—we would expect little use of performance controls in the technostructure. As for action planning, again the technocratic units do a good deal of it but seem to be only marginally affected by it themselves. We would expect the use of planning and control systems to vary
considerably in the support staff. Only those units that act as relatively autonomous entities and that have easily measured outputs—such as the cafeteria in the plant or the bookstore in the university—can be controlled primarily by performance standards. Some staff units with important inter- dependencies with other parts of the organization—such as the research department in the corporation—may be subject to action planning, at least to the extent that the line departments they serve are so subjected. And others, such as legal council, may experience little in the way of any plan- ning and control system.
Source: Mintzberg Henry (1992), Structure in Fives: Designing Effective Organizations, Pearson; 1st edition.