Volume 18 Number 1      January 1993

                                        The Academy of Management


Articles                           Deconstructing Organizations                                                             
Martin Kilduff


Tailored Meanings: On the Meaning and Impact of Organizational Dress

Anat Rafaeli and Michael G. Pratt


Personal Networks Of Woman and Minorities in Management: A Conceptual Framework

                                       Herminia Ibarra


Emotional Labor in Service Roles: The Influence Of Identity                                                                                          

Blake E. Ashforth  and Ronald H. Hunmphrey


The Architecture of Simplicity

                                                      Danny Miller

C Academy of Management Review


1993. Vol. 18. No. 1. 116- 138.








Ecole des Hautes Etudes Commerciales and McGill University


This article argues that over time the world views, goals, strategies, cultures and processes of successful organizations will become more pure or “simple”: they will come to focus more narrowly on a single theme, activity or issue at the expense of all others. This is explained by managerial, cultural, structural, and process factors within the organization. It is also attributed to both the complementary way in which these factors configure and the paradox that although simplicity may trigger ultimate failure, it can bring about initial success. The article offers some illustrative propositions concerning the nature, causes, moderating factors, and consequences of simplicity, and it makes suggestions for conducting further research.



Researchers have claimed that successful organizations fail because they have lost their edge: Star marketers lose touch with their customers, brilliant innovators become stale, and high quality producers start to turn out shoddy merchandise (Deal & Kennedy, 1982; Peters & Waterman, 1982; Weitzel & Jonsson, 1989). This article offers a contrary thesis: It argues that most outstanding organizations lapse into decline precisely because they have developed too sharp an edge. They amplify and extend a single strength or function while neglecting most others. Ultimately, a rich and complex organization becomes excessively simple-it turns into a mono­lithic, narrowly focused version of its former self, converting a formula for success into a path toward failure.

Even though this problem has been neglected by the literature on organizational decline (Cameron, Whetten, & Kim, 1987: Hambrick & D’Aveni, 1988; Murray & Jick, 1985; Weitzel & Jonsson, 1989; Whetten, 1980), a good deal of rich anecdotal evidence suggests that increasing simplic­ity victimizes many once-outstanding organizations. Meyer and Starbuck (1991), Miller (1990), and Starbuck, Greve, and Hedberg (1978) have pre­sented numerous examples of formerly thriving companies that came to focus—almost to the exclusion of everything else—on the one goal, as­pect of strategy, department, or even skill that they credited for their success. Indeed, the business literature is full of such accounts (see Hal­berstam (1986] on Ford Motor Co., Wright [1979] on General Motors, Lyon [1984] on Dome Petroleum, and Colvin [1982] on ITT).

Control Data, Polaroid, and Wang Labs, for example, which had suc­ceeded by out-innovating their competitors, turned this policy into an obsession. They began to concentrate only on technological innovation— no matter what the costs or the needs of the customer. Marketing, pro­duction, and financial considerations fell by the wayside as the depart­ments handling these activities became less influential and the power of the R&D elite mounted. Subsidiary goals of service and market penetra­tion were driven out by an increasing obsession with scientific progress. And the organizational culture became more homogeneous as dissenting managers left. Even corporate systems and routines came to monitor and control mainly technical matters while ignoring issues of quality or prof­itability (Miller, 1990).


The Notion of Simplicity


The Concise Oxford Dictionary defines simple as “not compound, consisting of one element, all of one kind, involving only one operation or power.” In the context of this paper, simplicity is an overwhelming pre­occupation with a single goal, strategic activity, department, or world view—one that increasingly precludes consideration of any others. As simplicity increases in a company, secondary issues are forgotten, and the parties responsible for them lose influence. The organization becomes more monolithic, with its members and subunits having fewer and in­creasingly similar preoccupations and its systems becoming more spe­cialized.

There are both objective and subjective varieties of organizational simplicity, and most of these are interdependent. The objective varieties include dominance of a single goal or subunit, information systems and routines that reflect only a narrow range of skills and concerns, and a lion’s share of resources going to one central tactic or activity. But sim­plicity may also be reflected subjectively, by the narrowing, increasingly homogeneous managerial “lenses” or world views that often underlie the more objective forms of simplicity.


Simplicity Versus Inertia. Momentum, and Convergence


Simplicity is related to, but very different from, the much-discussed notions of inertia, momentum, and convergence. Inertia is resistance to change, or, at least, resistance to changes that run counter to a funda­mental existing orientation. Hannan and Freeman (1984) claimed that inertial programs and routines increase an organization’s reliability of functioning but limit its capacity to change. However, inertial structures, processes, or systems need not be simple. Inertia limits only interperiod variety, but simplicity implies little variety at a point in time.

             Simplicity is also different from Miller and Friesen’s (1980a, 1984) notion of momentum—the tendency to extrapolate previous directions of evolution in strategy and structure, and from Tushman and Romanelli’s (1985) concept of convergence—the idea that organizations incrementally build upon or refine an existing orientation. In fact, momentum or con ; vergence can result in more elaborate and complex orientations as well as more simple ones.

We shall see that simplicity can be influenced by many of the same things that cause inertia. It also can increase inertia by so narrowing organizational attention that the need for reorientation goes unrecog­nized. Moreover, inertial routines and processes can, in turn, contribute to simplicity. Nonetheless, the two concepts are quite distinct.


Simplicity and Performance


Escalating simplicity can have dire consequences for the perfor­mance of an organization. This problem, perhaps, is best illustrated by the law of requisite variety. According to Buckley (1968: 495), this law states that “the variety within a system must be at least as great as the environmental variety against which it is attempting to regulate itself. Put more succinctly, only variety can regulate variety.” Weick (1979: 189) reinforced this point:


It a simple process is applied to complicated data, then only a small portion of that data will be registered, attended to, and made unequivocal. Most of the input will remain untouched and will remain a puzzle to people concerning what is up and why they are unable to manage it.


In their words, if an organization were too simple.. to manage the complexity of its environment, its very survival might be threatened. Par­adoxically, however, simplicity can initially bring great rewards when it marshalls the strengths of an organization to accomplish what it does best. But success is not only a product but a cause of simplicity.

The thesis of this article is that in the long run, success will cause many organizations to become more “simple.”  Their rich strategic char­acter will devolve into bland and truncated caricature. Cultures, systems, processes, and world views will become too monolithic to allow organi­zations to embrace and adapt to the complex currents of their settings.’ And, ultimately, these developments will result in companies’ reflecting the winds of change not with the responsiveness of sandy terrain but with the inertia of a field of boulders.

Our global expectations can be summarized by two propositions:


Proposition 1: In successful organizations, primary goals, values, and strategies will be pursued more ag­gressively, whereas secondary ones will be increasingly neglected.


Proposition 2: All varieties of simplicity will increase as a function of ‘the duration and degree of success in achieving the goals of the dominant coalition.


Proposition 2 helps to establish the scope of our analysis. It suggests that the more successful a firm is in achieving the goals most valued by its dominant managers, and the longer the interval of success, the more likely and the more pronounced the organization’s move toward simplic­ity. Firms whose performances are poor or whose intervals of success are interrupted by occasional shocks of disappointment are more apt to pre­serve a healthy level of doubt, debate, and diversity.

The balance of this article will present three classes of reasons for this encroaching and dangerous simplicity. First, individual managerial, cultural, structural, and process factors provoke simplicity. Such factors include organizational learning; the “natural selection” of values, heroes, and skills; and confining programs and routines. Second, these factors tend to interact, generating increasingly pure and simple corporate con­figurations—constellations that become ever more aligned with a single dominant theme and less tolerant of deviation or variation. Third, a trou­blesome paradox exists: The sources of dangerous simplicity may under­lie initial success and, thus, may be doubly difficult to combat. Indeed, it is very hard to distinguish between the concentration and passionate h dedication so necessary for success and competitive advantage and the simplistic fixations and extremes that lead to faijure. The article closes by operationalizing the notion of simplicity and summarizing the proposi­tions concerning its evolution, causes, facilitating conditions, and conse­quences.




The “Lenses” of Experience


Organizational simplicity is caused, in part, by managerial learning and the initial successes it brings. At first, new managers grope to find out more about their environments. But over time, positive reinforcements induce them to search more narrowly. Ultimately, experienced managers form quite definite opinions about what works and why (Levitt & March, 1988). Such thinking is especially true of the executives of thriving orga­nizations who credit—and therefore cling to—past behaviors, policies, programs, and strategies (Bettman & Weitz, 1983; Milliken & Lant. 1991; Staw, McKechnie, & Puffer, 1983).

Experience, in other words, shapes the lenslike cognitive structures through which managers see the world. These structures take the form of established sets of values, assumptions, and beliefs. According to Nys­trom and Starbuck (1984: 55):


What people can see, predict, understand depends on their cognitive structures—by which we mean logically integrated and mutually reinforcing systems of beliefs and values. Cog­nitive structures manifest themselves in perceptual frame­works, expectations, world views, plans, goals,..  myths, rituals, symbols.. . and jargon.


These lenses dictate what managers will perceive, what they will ignore, and how they will interpret their perceptions. Gradually and unconsciously the lenses harden and focus more narrowly around those notions that have been—or appear to have been—most thoroughly vin­dicated (Staw, Sandelands, & Dutton, 1981). The same kinds of stimuli then get targeted for attention, and the same mental frameworks are used to understand them (Starbuck & Hedberg, 1977). Thereafter, managers are likely to be more consistent and more selective in what they attend to (Levinthal & March, 1981; Starbuck & Milliken, 1988).


Single-Loop Learning Versus Double-Loop Learning


Serious and persistent problems may force managers to broaden their horizons. But in successful organizations, managerial learning will be biased toward honing existing categories and attributing success, often superstitiously, to a single pet policy (Levitt & March, 1988; March, 1991). Again, the result is a simpler, more consistent view of the world.

Argyris and Schon (1978) distinguished between typical single-loop learning and the much rarer double-loop learning. Single-loop learning occurs as organizations compare their performances to a set of preestab­lished standards and try to make appropriate adjustments. Most success­ful firms accomplish this procedure very well. Companies that differen­tiate themselves by producing products of superb quality, for example, adeptly hone the precision of their manufacturing processes. They pursue a single objective with increasing virtuosity.

Double-loop learning, on the other hand, requires that firms period­ically reassess standards themselves to ensure that these are relevant. Because this can be a costly and painful process, it is normally elicited only by major difficulties (March, 1991; Miller & Friesen, 1980a), which are rare among good performers. Healthy quality leaders, for example, do not question the broader appropriateness of their technical parameters and are reluctant to adopt potentially more relevant standards such as cus­tomer reactions (Miller, l990b). Because the old standards are so closely tied to managerial goals, expectations, and world views, they are seldom reappraised. And when managers reevaluate their standards, it is not to challenge but to refine and extend them. Therefore, standards become more precisely focused, as do the strategies used to attain them.


Ambiguity. Rationalization, and Defenses


Skeptics might argue that disappointments will inevitably cause most managers to broaden or to abandon their views. Even in successful organizations, there are occasional small failures that should induce managers to reexamine their premises and to gather more broadly rele­vant information (Weick, 1979). Yet executives are unlikely to learn from these shocks because problems can be rationalized away as aberrations, as temporary, or as beyond management’s control. And, as we shall see, the routines, information systems, and homogeneous cultures that develop in successful organizations further channel conditions and values and make double-loop learning very difficult.

Unfortunately, managers are more likely to broaden their knowledge if upsets are multiple, earthshaking, or linked unambiguously to one specific behavior (Mintzberg, 1989: 35—37). Such instances are rare, espe­cially in thriving organizations. As Staw (1977: 468) has claimed, the at­tributions administrators make in deciding which of their behaviors is effective are often inaccurate, “given the incomplete and confounded data” with which they must work. Weick (1979: 175) argued that “the en­vironment is full of equivocal cues that are easy to misinterpret—multi. pie significations—they fit numerous classifications and might be indi­cators of any one of several states.” Thus, most organizational disappoint­ments may easily be dealt with through “retrospective rationalizations.”

Managers may, for example, attribute success to their own actions and absolve themselves of responsibility for any failures (Bettman & Weitz, 1983; Miller & Ross, 1975). They interpret weak clues as indicating that their favorite strategies have led to success (Salancik & Meindl, 1984; Staw, McKechnie, & Puffer, 1983; Staw & Ross, 1978). Marketers attribute victory to good marketing; engineers credit good design; and both blame failure on “the economy” or “the guys down the hail.”

Staw (1976) demonstrated just how hard it is for administrators to “backtrack” or make decisions that can be construed as an admission of previous error. Indeed, many managers may actually escalate their com­mitment to a particular course of action in the face of disappointments in order to vindicate their convictions.

Managers’ defense mechanisms also may prevent them from broad­ening their beliefs and policies. Most executives are so committed to the strategies and cultures they have nurtured that it is painful for them to admit that these are obsolete. So they might deny that there is a problem, believe that the problem isn’t serious, or think that others are trying to steal power by attacking their contributions (Kets de Vries & Miller, 1984).


Overconfidence and Intolerance


Success not only narrows executives’ perceptions, it also changes their attitudes. It can give them too much confidence in a single way of conducting business or in one dominant element of strategy. Successful leaders have triumphed over their competitors and are worshiped and idolized by subordinates; they have had many of their opinions enshrined in policies and vindicated by the turn of events. So many leaders become cocky and overconfident and are emboldened to take reckless, impulsive actions. Successful diversification strategies, for instance, may prompt managers to pursue more ambitious and wide-ranging acquisitions (Kets de Vries & Miller, 1987). Pioneering firms that do well often accelerate their pace of innovation (Bylinsky, 1973). And brilliant salespeople become con­vinced that they can market anything through aggressive selling tech­niques (Miller, 1990b). Attention to most other aspects of strategy wanes. Success also can make leaders intolerant of opposing points of view. In some cases, bosses become bullies who resent criticism and isolate dissenters. They nurture corps of sycophants and foster a monolithic band of like-minded decision makers (Zaleznik & Kets de Vries, 1975). In es­sence, all capacity for frank discussion and organizational learning is lost.

The arguments of this section give rise to the following proposition:

Proposition 3: In successful organizations, managerial world views will become more homogeneous and will focus on ever fewer objectives, issues, and cues from the environment.




If managerial factors make successful organizations simpler, then so do many aspects of corporate culture. A culture is a constellation of basic views and assumptions, expressed as beliefs and values, that is shared by the key members of an organization. A culture gives an organization its identity, both to its members and to outsiders (Deal & Kennedy, 1982; Frost, Moore, Louis, Lundberg, & Martin, 1985).


Natural Selection of Values


According to the population ecology view, the environment helps to extinguish dysfunctional forms of organizations while it encourages the proliferation of more viable ones (Aldrich, 1979; Hannan & Freeman, 1977). Such ecological forces also may be at work inside organizations, “select­ing in” or reinforcing the values and behaviors associated with success and “selecting out” or extinguishing those values and behaviors that are deemed to be peripheral, unsuccessful, or unimportant (Burgelman, 1991; Campbell, 1970; Singh, 1990). Over time, the culture of the organization comes to focus more narrowly and passionately on one or two pervasive and dominant goals. Such strong cultures can make work meaningful, can galvanize employees to take action, and can generate tremendous enthusiasm. But they may also mire managers in a single way of seeing and doing things. They can bring about oppressive conformity, blindness, and intolerance. Indeed, Irving Janis (1972: 197) argued that the more cohesive the group and the more it distinguishes and insulates itself from other groups, the more conformist and intolerant its members. Ironically, .~ the dedicated, rousing corporate cultures so eloquently lauded by Deal and Kennedy (1982) are precisely the ones that may be the most suscep­tible to a hazardous “groupthink” narrowness.


Departmental Dominance


If particular goals and values can come to dominate organizations. then so can the departments that best embody these goals and values. Successful quality leaders get taken over by production-engineering Ca­bals, image differentiators by packs of marketers. Again, a selection process helps to explain how this happens (Burgelman, 1991). Innovator firms, for example, because they so greatly value innovation, usually accord some initial primacy to their R&D group. Researchers are avidly recruited, handsomely rewarded, and quickly promoted. This swells their ranks, attracts the best candidates, and boosts the power, resources, and impact of the R&D department. The opposite is true for the manufacturing staff who become disenfranchised, are starved for resources, and have little chance of moving up the hierarchy. The best production people soon leave for companies where they will be appreciated; the worst ones stay because they have little choice. This further erodes the production de­partment’s reputation. In this vicious circle, favored R&D types become more numerous, more talented, and more influential—and the opposite happens to manufacturing managers (Miller, 1990b).

The increasing dominance of a single department creates a mono­lithic culture, in which all influential managers embrace the same highly constrained values and strategies. The effect of such a culture is to dra­matically narrow managers’ viewpoints, abilities, and options. Gareth Morgan (1986: 242) referred to such organizations as developing an “ego­centric self-image”:


                      They draw boundaries around a definition of themselves, and attempt to advance the

                      self-interest of this narrow domain. In the process, they truncate their understanding

                      of the wider context in which they operate, and surrender their future to the

                      way the context evolves.


Specialized Knowledge


Over time, a thriving organization hones and focuses its set of ideas and the technology that is associated with it. Kim Clark (1988) has argued that an organization’s technologies stabilize and become more special­ized. Its employees develop a narrow knowledge base; they know how to do various tasks, but they forget why work is done in a specific way. In a sense, the search for comprehension is replaced by the quest for ref me­ment (March, 1991). The result is that most organizations unreflectively embrace a narrowing set of skills and employ people whose knowledge is confined to a single technology (Meyer & Starbuck, 1991; Miller, 1990b).

In a more general vein, successful organizations come to concentrate only on certain skills—those required to implement their current strate­gies and those corresponding to the knowledge of only the most esteemed managers and departments (Tushman, Newman, & Romanelli, 1986). Sec­ondary skills are lost because the practitioners of such skills fail to garner much power or respect (Milliken & Lant, 1991). And strategy becomes increasingly constrained by this narrowing skill set.

The above arguments suggest the following proposition:


Proposition 4: In successful organizations, values will become more homogeneous, reducing

subunit differentiation; a single department or elite will become more dominant; and the

skill set of the organization will nar­row. These changes will contribute to the formation

of monolithic cultures and strategies.




Structure defines an organization’s reporting relationships. power, and job responsibilities. And by doing so, it profoundly channels both managers’ perceptions and the way in which they make decisions.


Programs and Routines


Strategies tend to remain stable in successful organizations because there is little incentive to alter them (Miller & Friesen, 1980; Tushman, Newman, & Romanelli, 1986). This stability makes strategies more sus­ceptible to explicit codification and routinization, which, in turn, makes them narrower and still more change resistent. The more an organiza­tion’s goals and tasks are factored into routines and programs. the more these restrict and homogenize the range of things its managers think about (Cyert & March, 1963; March & Simon, 1958; Nelson & Winter, 1982). Routines also create unconscious premises for action. For example. man­agers begin to react automatically to signals from standard operating procedures. machine deterioration, or work-in-process, as they increas­ingly ignore stimuli not explicitly captured by the routines (Perrow, 1986).

In fact, Starbuck and Hedberg (1977: 250) argued that an organiza­tion’s adaptation to its markets becomes stylized and frozen by routines:


The organization sets up behavior programs that promote ha­bitual responses to expected cues. . . . Because situations ap­pear equivalent as long as they can be handled by the same programs, programs remain in use long after the situations they fit have faded away… Continued success incubates potential failure, by increasing an organization’s dependence on its programs.


Programs and routines restrict perceptions and activities and rein­force existing policies by recognizing only recurring and anticipated problems. suggesting only conventional courses of action, and imple­menting only traditional solutions. Routines can promote innovation— but only of a sort that is in line with current ideologies and strategies (Nelson & Winter, 1982). Routines also may be perpetuated by values of efficiency (Hannan & Freeman, 1984). so that expedient procedures be­come codified and endure, while the rest die out (Starbuck, 1985).


The Concentration of Power


In thriving organizations, power will usually gravitate toward those dominant departments and individuals that have been identified with past successes; it will recede from others (Pfeffer, 1981). This trend favors