One of my first teaching experiences was with an evening class of MBA students, at least half of whom were older than I was and all of whom had a lot more real-world experience.1 The evening’s topic was organization design, specifically, the choice to centralize or decentralize—a choice with profound implications for the structure of incentives, communication patterns, and decision rights. We discussed how decentralized designs promote innovation, motivation, and autonomous action, while centralized designs support control, efficiency, and coordination. My central (rather predictable) message was about the need to “design for fit”—to match organizational structure to strategy, an idea that seems on the surface to be very sensible.
After the case discussion ended, a student employed at McDonnell Douglas (now Boeing) raised his hand and remarked: “I disagree with everything you have said. I have watched our company move back and forth between centralization and decentralization repeatedly, and it has nothing to do with what you just described.” A classmate chimed in, “I agree completely with [the first student]. Our company has also bounced back and forth between centralization and decentralization. I am quite sure management has no idea what they are doing.”
This rather sobering start to my teaching career presented a puzzle that remained with me for years. If fit is the central object of organization design—fit with the environment or fit with strategy—why do firms change their design so often? Are strategies or environments really changing at this pace? Are managers just flailing about to elevate performance?
Perhaps we’re asking the wrong question. If you assume that sustained value creation is about having a theory that repeatedly points to new combinations of assets and activities and that these new and evolving combinations require a complex and divergent set of incentives and behaviors, such as both widespread attention to innovation and an obsession with cost reduction, then it quickly becomes apparent that your organization can never in fact fit perfectly to a strategy. Quite simply, no internal organizational design (or set of contractual relationships) can configure or target the entire array of complementary behaviors that pursuit of a corporate theory demands. The path to sustained value creation requires pushing the organization in any number of directions. Designing a strategy that attempts to pursue all these directions or accelerates attention to all of these behaviors simultaneously will thus necessarily be less effective than an approach in which designs for pursuing corporate theory are temporally sequenced to emphasize at any point in time a more focused set. Note that in this view, changes in design or emphasis are not driven by dynamics in the environment that call for shifting the target, but rather by the complexity of the target and the limits of design in optimally shaping both individual and organizational attention.
In hindsight, it’s ironic that the case I used in that class to illustrate the best-fit theory of organizational structure was Hewlett-Packard. To be sure, it powerfully illustrated the importance of fit—both the consequences of poor fit and the improvement that accompanies correction. But it also turned out to be a case study in the impermanence of organizational design and the need for ongoing, leader-directed dynamics.
There Is No One Best Fit
When I taught that evening’s class, HP had completed its first big organizational restructuring. The company had had a long history of extreme decentralization and considerable autonomy. At the beginning of the 1980s, the firm was composed of forty-five small divisions, each crafting, manufacturing, and to a large degree marketing its own specific products. This organizational design had fueled HP’s early success as a manufacturer of testing and measurement tools and positioned it as one of the world’s most innovative firms.
But by the early 1980s, HP had emerged as a major player in computing, increasingly competing with powerhouse IBM. Unlike its historic businesses in stand-alone test and measurement instruments, whose customers preferred the products dropped on their desks without so much as an instruction manual in a classic geek-to-geek transaction, computing demanded integrated solutions—and the autonomous divisions were now producing incompatible components, supporting redundant development efforts, and creating confused customers. For the first time, HP was confronted with problems and challenges for which its current organizational design provided poor solutions.
To remedy this, beginning in 1983, HP took significant steps toward centralizing manufacturing, marketing, and engineering. The shift quickly worked wonders. The company agreed on common standards and platforms. It eliminated redundancy, developed integrated solutions, and generated happier customers. Financial performance also improved. Moreover, management was convinced that with this new structure, HP had achieved a design that balanced efficiency and coordination with autonomy and innovation. This was where the case ended—a powerful illustration of how design for fit in the wake of a shifting strategy elevates performance.
However, as it turned out, the HP case was just beginning. About the time I was delivering my ill-fated lecture, Hewlett-Packard was actually beginning to reverse its course—decentralizing again. By the late 1980s, innovation at HP had tanked. New products were late. Decision making was mired in administrative processes; even simple decisions were being pushed well up the management ladder. Financial performance had plummeted. In response, HP decentralized its structure, granting divisions considerable autonomy, even moving division heads away from Palo Alto headquarters. Innovation and overall performance improved. With this shift, HP management and securities analysts covering the company were now convinced that HP had achieved proper balance between the need for autonomy’s innovation capabilities and centralization’s coordination and efficiency.
In 1995, the company again switched course. Its performance had softened and executives perceived a need to coordinate on the development of solutions and integrate their disparate technological expertise. HP once more began a shift toward centralization. As coordination improved, so did performance. The firm’s executives and the reports of securities analysts again expressed conviction that HP had achieved a good balance—now a “faster, more competitive company, with an improved product and services offering [and] greater ability to deliver solution.”2
It should come as no surprise, however, that another structural shift quickly followed, as HP’s performance again declined in 1998. Predictably, HP shifted to decentralization, and performance improved for a period. But when HP’s first externally appointed CEO, Carly Fiorina, arrived in 1999, she was struck by the inefficiency and lack of coordination across divisions. She immediately centralized HP to a degree unprecedented in its history, and the results were positive. But sure enough, by 2004, Fiorina faced enormous pressure from both the board and market analysts to decentralize. Most credit her subsequent abrupt dismissal to an unwillingness to listen to the board, particularly an unwillingness to loosen the reins at HP and decentralize. New CEO Mark Hurd, of course, did precisely that, and financial performance accelerated again.
Interestingly, through all of this often wrenching and expensive structural change—this back-and-forth in design—HP had emerged as the world’s largest IT firm. Its stock price significantly outperformed all broad market indices over the overall time frame. This strongly suggests that, far from being a bad thing, repeated restructuring may actually be what big companies chasing sustained competitive advantage need to do. Organizational design, in other words, is not a quest for a holy grail of best fit—a stable solution that configures the internal behaviors, investments, decisions, communication, and knowledge flows that will together generate the value you envision. It is fundamentally a dynamic endeavor, even though the strategy and surrounding environment may be stable, even static. HP, for instance, was consistently focused through this whole history on two critical performance dimensions: innovation and efficiency, or what some have labeled exploration and exploitation. When HP decentralized, it was successful in innovation, when it centralized it drove exploitation. Despite its best efforts, and ongoing rhetoric, it was never able to discover a structure that balanced these two permanent concerns. Instead, it achieved success in both by dynamically changing design as the relative importance of the two dimensions changed.
The Complexity Challenge
Even if HP had not been pursuing two somewhat conflicting goals, it would almost certainly still have had to design and redesign as it grew. That’s because of another fundamental problem with the assumption that organizational design is stable: there is no way that a single, well-crafted, comprehensive design, anchored on activities that can be measured, can actually generate a highly complex pattern of behaviors all at once. Managers, no matter how skillful, simply cannot engineer the set of incentives and structures and the accompanying social environment that will generate the behaviors, investments, and outcomes that a theory for sustaining repeated value creation (as opposed to the quest for an individual competitive advantage) must imply. Any attempts to craft such a comprehensive, definitive organizational model for competing in a complex world of multiple incentives and behaviors inevitably run into one or more of three problems:
Brain overload
Our brains are just not wired to respond effectively to complex designs that seek to push us in a multitude of directions simultaneously. We are not cognitively programmed to multitask. In fact, one study suggests that those who claim extraordinary multitasking capacity are actually particularly inept.3 A study published in 2010 by two Paris-based researchers provides an explanation. Participants were assigned first one, then two, and finally three complicated letter-matching tasks, while brain activity was monitored with FMRI technology. When subjects were assigned only one task, both right and left brain hemispheres were focused on performing the singular task. With the introduction of a second task, the left brain focused on one activity, and the right focused on the other, each hemisphere working independently to pursue its respective goal and reward. When a third task was added, participants consistently forgot one of the three tasks. They also made three times as many errors as did those balancing only two tasks. The conclusion was that when we attempt to pursue three goals at the same time, we simply discard one and still perform rather poorly on the remaining two.4
It is not surprising, therefore, that organizations struggle when pushed to focus on multiple objectives and performance dimensions. While organizations may divide and conquer by assigning different problems, goals, and objectives to different groups, the approach has clear limits. Many of the problems, goals, and objectives that a corporate theory reveals have application across all groups and individuals and thus, cannot simply be allocated “two per group.” Given simple limits to cognitive attention, keeping substantial organizational attention on multiple goals requires sequencing focus and emphasis over time. At a purely cognitive level, asking for concerted attention to an abundance of goals simultaneously is simply wasted effort.5
Divided motivations
Even if we set aside individual cognitive limits, a fundamental motivation problem typically plagues efforts to simultaneously direct attention to multiple goals and objectives. The only exception is when the measures and goals revealed by your corporate theory are highly correlated—that is, when attention to one goal improves another. Here the behaviors required to generate measure A are substantially the same as those required to generate measure B. Motivating one motivates the other. But this scenario is highly unlikely, and therein lies the challenge. For instance, when the behaviors required to generate autonomous action and innovation differ from the behaviors required to produce efficiency and coordination, how does the employee determine an effective response? Telling employees to maximize current profits, quality, market share, service delivery, future growth in profits, and anything else deemed important leaves them confused about how to prioritize.6
When performance metrics differ in accuracy and in the degree to which employees can control them, the challenge in motivation becomes even greater. Activities whose performance dimensions are poorly measured or difficult to control are neglected in favor of those whose metrics produce more measurable results and are more easily controlled. Lack of attention to the tasks with more difficult metrics degrades performance. Steven Kerr’s paper, “On the Folly of Rewarding A While Hoping for B” first captured the essence of this broad problem. Economists later formally designated and further developed this as the multitasking problem.7
Whatever the label, the implications of the problem are clear. Measuring all dimensions of performance and linking rewards to improvement on each does not generate the behavior desired. Attention is selective. Shifting attention to those difficult-to-measure performance dimensions that are often those most central to performance requires dulling incentives for other performance dimensions. For instance, if quantity produced is easily controlled and measured, while quality is difficult to measure and observe, incentives focused on quantity compromise attention to quality. There is simply no combination of incentive composition that generates balanced attention to all critical dimensions. For many firms, the best they can do is essentially provide no incentives at all—a rather poor solution.
Inconsistent choices
Attempts at comprehensive design solutions are further complicated by the fact that effective organizational designs are bundles of complementary choices. Design is not an exercise in ordering à la carte. Instead, it requires choosing from set menus of design elements that mutually reinforce one another and drive desired behaviors and outcomes. There is thus an inherent discreteness to design. Crafting a design that effectively generates one behavior involves the selection of elements inconsistent with generating another. Attempting to configure an organization that drives all necessary behaviors would involve the selection of highly inconsistent design elements.
Consider, for instance, decentralization and centralization as two distinct designs, each generating a distinct trajectory of behavior. As the HP example that opened this chapter shows, decentralization promotes decision making, communication, and knowledge flows, which fuel exploration and innovation, while decentralization reduces redundancy and facilitates coordination, which fuels efficiency. Each requires differing choices regarding design, incentives, and measurement. Decentralized structures distribute decision rights and performance measures downward and offer powerful incentives. Centralized structures elevate decision rights and performance measure to high levels in the organization and commonly provide lower-level incentives.
Sometimes, of course, corporate theories are strongly aligned with an internally consistent organizational design. Google’s corporate theory, for example, calls for a decentralized organizational design. The theory is about the development of cutting-edge, breakthrough technology. Google believes in throwing new technology out into the market to see how it sticks, worrying only afterward about how to exploit it for profit. The emphasis is on hiring the best, brightest, and most creative employees and then providing them with resources to pursue novel opportunities, frequently of their own selection. Google’s organization, therefore, is decentralized, granting considerable autonomy to individuals and groups to pursue projects. Incentives tend to mirror those offered in small entrepreneurial firms (that is, they are high-level) and the culture seeks to replicate the same informality and latitude. There are few efforts to integrate the disparate pursuits of these distinct units.
But it’s rare that the full set of behaviors and investments foreseen by a corporate theory align so neatly with the outcomes promoted by one of the discrete design options available to managers. As we saw with HP, many firms simultaneously need the close integration, coordination, and capacity to streamline that centralization delivers, yet also the more autonomous adaptation and innovation that decentralization supports. In fact, while the mix may vary, sustained value creation for nearly all organizations demands both—the consistent generation of innovative new products and services and consistent improvement in production and distribution efficiency. And, of course, the design problem is even more complex, as there is a considerable array of performance dimensions that don’t align neatly with either centralization or decentralization.
The Logic of Dynamic Design
How can firms confront these challenges to crafting a comprehensive design that delivers attention to the multiple goals, objectives, or investments revealed by a corporate theory? The key is to view design as a dynamic endeavor rather than static engineering.
For all the rhetoric about the importance of sharing the organization’s broader vision, an individual’s job at any given moment is likely focused on a rather limited set of organizational goals. There are good reasons for this. As Dan Levinthal and Sendil Ethiraj note, having a focused set of goals and attention provides individuals with the clarity required to motivate action.8 A single goal or a narrow set of goals matched with the appropriate organizational design motivates focused attention. Such focus accelerates progress, even if that progress is imbalanced relative to the full set performance dimensions or desired behaviors and investments that a corporate theory may reveal. Of course, the patterns of complementarity among performance dimensions and desired behaviors revealed by your theory ensure that extended attention to one elevates the returns to focusing on another in the future.
In other words, if your firm has become extremely lean and efficient in its production and distribution of product, the optimal path to value creation may be generating new innovative products to take advantage of the system. If on the other hand, the organization is tremendously innovative and has generated a pipeline of innovative products, the optimal path to value creation may be to streamline production and distribution. Dynamic design involves selecting the optimal path for the moment on the understanding that you will switch to another in the future.
In this sense, optimal organization design is a bit like sailing into the wind. Attempts to configure main sail, jib, and rudder in order to sail a direct course generate no progress (or negative progress). However, configuring the ship to sail 40 degrees off wind can generate tremendous speed in a useful direction. Yet as the ship travels further in this path, the benefit of a directional shift increases, and the next move is to come about and lay in a new course 40 degrees off wind in the other direction. While each course correction momentarily compromises forward momentum, skillful sailors master these reconfigurations of sail, jib, and rudder to minimize lost time. By periodic tacking, the ship achieves its destination far faster than it could were it to maintain a constant trajectory for an extended period of time.
In much the same way, skilled organizational architects deploy organizational design elements—structure, measures, and incentives—into a coherent form that generates momentum along a valuable trajectory. Moreover, it is precisely the success of the current trajectory, not its failure, that precipitates the benefits of choosing another form and path. The benefits of dynamically shifting design and focus arise in part because the benefits of the current focus do not immediately dissipate upon shifting to another. Thus, an organization’s capacity for innovation spurred by decentralization does not immediately disappear with a shift toward centralization and a focus on efficiency. It merely dissipates with inertia. Thus, in pursuing your corporate theory—in attempting to generate this complementary bundle of activities, investments, or behaviors, dynamically choosing the pattern and cadence of change is central. This capacity to dynamically shape attention, focus, and structure is precisely what we observe in effective organizations. The trick is discovering when to do what, not how to do it all at once.
In managing dynamic design, the leader’s first task is to determine the optimal path at any given moment—to constantly monitor the outcomes of the present form and determine at what point switching to another will maximize performance for the firm. Once a decision to change has been taken, there are various ways in which it can be achieved, any and all of which the skilled leader will use. For some firms, dynamic design is best achieved through periodic structural change. For others, however, dynamic design can simply mean a set of sequenced initiatives. Let’s briefly consider each.
Structural change
Organizations are massively inert. The communication patterns, work routines, and decision-making processes that comprise organizations resist change. Therefore, to shift the focus and attention of an organization requires more than a gentle nudge. It requires an aggressive shove. Significant structural change provides precisely this shove, as the HP saga illustrates. By switching between centralization and decentralization over twenty-five years, HP shifted its focus between innovation and efficiency over time, arguably generating more of each than it could have achieved were it to have selected best fit at the outset and maintained that structure through the duration. HP’s story is not unique. “Dynamic Design at Ford,” details Ford’s efforts at globalization—a similar story of vacillating between two distinct structures to achieve both locally tailored designs and global production efficiency.
Two broad performance dimensions are central to global success in automobiles: (1) geographically tailored offerings, responsive to local tastes and preferences and (2) global efficiency (scale economies) in design, procurement, and assembly, frequently achieved by common models, parts, and platforms. These two dimensions are complements in generating performance. Tailored designs increase volume, which provides the scale necessary for efficient manufacturing. Scale provides the efficiency needed to competitively price and sell locally tailored vehicles. However, while local design and global scale are complements to performance, they are substitutes in their production. In other words, organizational designs that drive locally tailored automobile models are inconsistent with organizational designs that generate efficient manufacturing. Attempts to design for both generate weak attention to both. Consequently, Ford found it efficient to shift its structure over the years to pursue one and then the other.
For many decades, the Ford Motor Company was globally decentralized. Each region enjoyed considerable autonomy to design, manufacture, and purchase as it pleased. The result was well-made automobiles tailored to local preferences, but at a very high cost position, given the redundant design and global incompatibility generated by decentralized regional units. To remedy this, in 1994 Ford globally centralized purchasing, engineering, and manufacturing. Costs declined dramatically as Ford exploited economies of scale, common design platforms, and global purchasing power. Profitability jumped. However, with this centralization, regional managers lost considerable autonomy over product design. Over time, the result was predictable—the cars became poorly adapted to local tastes and circumstances. Soon, there were significant sales declines in international markets: in Europe, the Ford brand dropped from second to fourth place in market share; in Brazil, it lost 4 percentage points in market share.a Predictably, in 2000, Ford dramatically decentralized, giving local regional managers autonomy even greater than they had had before 1994.
Thus, decades of divisional autonomy at Ford generated highly innovative and locally responsive designs. But they also yielded poor cross-divisional coordination. Six years of centralization yielded vital coordination, but designs that didn’t cater to local preferences. The important point, however, is that Ford’s efficiency was unambiguously better off in 2000 for having switched six years previously. It now had common platforms, greater commonality of parts, and perhaps most importantly, reshaped communication patterns and design routines across its global operations. Perhaps Ford should have shifted back sooner or found ways to build platforms that enabled greater local autonomy in design, but it seems hard to dispute that by temporarily centralizing, Ford achieved levels of global efficiency that it could not have achieved had it remained decentralized.
a. Kathleen Kerwin and Keith Naughton, “Remaking Ford,” BusinessWeek, October 10, 1999.
Of course, patterns of structural change are more complex than simple shifts between centralization and decentralization. Corporate theories reveal multiple dimensions along which organizational design can improve performance. At consulting and accounting firms, for example, success requires leveraging existing relationships, sharing industry-specific knowledge, and transferring best practices. Each one of these activities calls for a distinctive organizational design. A geographic structure facilitates access to relationships held by senior partners. An industry structure facilitates knowledge exchange regarding industry opportunities and trends. A functional or practice structure facilitates best practice sharing by consulting area. Not surprisingly, such firms often cycle through the three. Each structure powerfully pushes on a critical performance dimension. Therefore, for these firms, sustained high performance demands dynamic change, and they discover that sequencing designs generates higher overall performance than sticking with one.
Initiative and goal sequencing
Shifting organizational structure is but one way to redirect focus and effort. Other leaders build and deploy sequences of initiatives that direct attention to different goals, problems, and performance dimensions revealed by a corporate theory. GE’s Jack Welch was a master at sequencing initiatives, each a way to direct attention to a new class of problems or to focus attention on a new performance dimension.
The initiatives cumulatively delivered major changes to GE’s focus during Welch’s twenty years at the helm. For a while, the focus was on delayering, layoffs, and putting together the right business portfolio through an M&A program, with the aim of positioning the remaining businesses as number one or number two in their industries. GE then became obsessed with Work-Out, a process focused on employee engagement, quick decision making, and solving internal problems and obstacles to performance. Attention then shifted outward, and the organization focused on identifying external best practices and bringing them into the company. GE next turned its attention to services, challenging its business units to increase the portion of their overall sales focused on service from 60 to 80 percent. Finally, Welch trained GE’s focus on Six Sigma, with its emphasis on improved quality and reduced cost.
Some firms, often those particularly skillful in acquisitions, are serial sequencers. They have developed scripts—patterns of initiatives that are imposed over time on an acquisition. Danaher, for instance, a firm that has averaged 25 percent compound annual shareholder returns since its inception in 1995, has a “tool box” of dozens of initiatives, processes, and training modules that are deployed as circumstances warrant. These tools and training modules focus on value selling, customer segmentation, product life cycle management, ideation, lean software design, supply chain management, Six Sigma, measurement analysis, and literally dozens of others. They target a range of goals, behaviors, areas of the firm, and performance dimensions. While all of these tools may be deployed at some point in a business’s history, there is a common initial sequence that is typically followed post acquisition. Thus, the leader’s task in dynamic design is to identify and select the proper sequence of programs, initiatives, or structures.
Whatever form the organizational changes take, major transformations are often accompanied by a change in leadership. Consider 3M, which focused for decades on innovation. Its structure, culture, and policies all targeted new products. Its philosophy was: hire great scientists, provide them with ample resources, and get out of their way. Employees were invited to devote 15 percent of their time to innovative projects that interested them personally. But in the late 1990s, 3M’s stock flatlined. There was a general sense that while the company was great at exploring new product terrain, it was not so great at squeezing profits out of the terrain it occupied. For several years, costs had grown at twice the pace of sales.
In response, the board looked for a new CEO who was skilled in execution, and hired an outsider, GE executive Jim McNerny. GE at the time was in its Six Sigma phase, and McNerney brought this passion for it to 3M. For the next four and a half years, McNerney used Six Sigma, coupled with complementary cost cutting and sourcing initiatives, to shift 3M’s focus toward exploiting existing product positions by trimming cost and waste. In the process, 3M shifted from being what many described as a playground for scientists to a more centralized and disciplined organization. The result was a sharp spike in profitability and share price.
But in what should be a now-familiar pattern, by the time McNerney moved on to Boeing in 2005, the new emphasis on exploitation and efficiency, while rendering 3M more profitable, had squeezed much of the innovative life out of the company. The board shifted back to an internal CEO steeped in the ways of 3M’s success with innovation. Some might conclude that the McNerney years were a misguided departure from leveraging 3M’s historic capability, but this interpretation ignores the fact that 3M enters this new phase having solved long-neglected problems and equipped with new routines, new skills, and better operations. However, at this time, the path to improved performance demands an alternative and historically more familiar emphasis. (This story might be instructive for two other firms famed for their historical focus on innovation, but in very different ways; see “A Change for Apple and Google?”)
The leadership changes rung in at 3M reflect a common pattern. When a PhD student at Washington University, James Yen, now a professor at Peking University’ business school, conducted a fascinating study examining CEO successions at all publicly traded companies from 1992 to 2011. He divided the CEOs into two camps based on their functional background and experience. One group he labeled output CEOs; these included those with backgrounds in strategic planning and consulting, entrepreneurship, sales, and R&D. The other group he labeled throughput CEOs, who had backgrounds in manufacturing, finance, accounting, process engineering, human resources, and law.
A Change for Apple and Google?
One could make a compelling case that Apple would now greatly benefit from a healthy dose of decentralization to cultivate and nurture new ideas and projects. A company so remarkably skillful in design execution might benefit from the breadth of new innovative ideas that decentralization generates. By contrast, Google might benefit from a sustained dose of centralization that would integrate its disparate technologies and applications into more coherent, user-friendly experiences. Admittedly, these revised organizational approaches would certainly have limited functional life spans. Permanently decentralizing the Apple organization would crush its remarkable capacity to develop integrated and effective user experiences. A permanently centralized Google would lose its capacity to broadly innovate.
What Yen found is that firms switch between both types of CEO over time. He specifically found that the probability that a firm will switch to a new CEO type conforms to a predictable pattern. The longer a firm has had a CEO of one type, the more likely the next CEO will be of the other. While this simple classification of CEOs into throughput and output hides a wealth of nuance and complexity, the results further support the principle of dynamic design. Firms generate high performance as they pursue complementary goals, behaviors, and investments that corporate theories reveal. However, organizing to pursue all of these at once is infeasible. Throughput and output initiatives are clearly complements to performance. Somewhat ironically, therefore, it may be the success of their initiatives, rather than their failure, that costs CEOs their jobs: their success in doing what they know how to do well triggers the return to doing something different.
This principle is at work not only in the selection of CEOs. Rather, it has application at all levels of the organization. The task in designing organizations is to dynamically construct the value a theory reveals. Theories reveal bundles of goals, behavior, and investments vital to value creation. In order to sequence the organization’s focus over time, leadership change may prove a valuable lever.
LESSONS LEARNED
Several key lessons emerge in this chapter:
• There is no design that will generate all desired behaviors. Firms need to constantly monitor their current position relative to their corporate theory and the array of behaviors, goals, and investments it reveals. The need for change is frequently and ideally not a symptom of failure, but rather a symptom of success. It is the success of the old design that invites the new—that elevates the benefit of a new approach that invites a new set of complementary investments or accelerates a new set of complementary behaviors.
This logically leads to the second lesson:
• Organizational design is a problem of dynamic optimization. The design question is not as simple as, What is the best organization design for my corporate theory? Rather, the critical question is, What is the appropriate organization design now? Different designs solve different problems; they invite different behaviors and investments. The leader’s task is to identify the problem that most needs solving today and design for it, recognizing that tomorrow’s problem and the design it needs will be different than today’s.
This leads to the third lesson:
• Timing is everything. Timing is the most important tool an organizational architect possesses. Like the sailor attempting to sail into the wind, your challenge is to not only put together the necessary change initiatives that will generate real velocity in the direction you select, but to impose them at the correct time. Bad timing dooms what may otherwise be great design.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.