Leading organizational transformations

By Steven F. Dichter, Chris Gagnon, and Ashok Alexander

The leader's role is to turn separate initiatives into a balanced, integrated program of change.

Many senior managers today are aggressively trying to transform their companies, seeking radically to improve performance by changing behavior and capabilities throughout the organization. Unfortunately, most leadership groups lack a proven way of thinking about the challenge.

Ask your management team what a good business plan looks like, and you will probably find close agreement. But ask them—especially in the middle of a major change effort—what a good change plan should include, and opinions will vary all over the map. A CFO will insist on creating new financial measures; an operations VP, on installing a quality program; an HR specialist, on revising compensation and training; a marketing executive, on getting everyone to be more customer focused. And all these managers will have handfuls of articles to wave—and mantras of buzzwords to invoke—to defend their choices.

The chaos of opinion created by hype and buzzwords is doubly unfortunate. Most obviously, if left unresolved, it can easily turn a desire for bold, systemic change into a rag-tag collection of discrete, ad hoc initiatives. Less obviously, but perhaps more troublingly, it can also prevent the kind of meaningful discussion that keeps a management group pulling together toward a common end. The CEO of a company facing transformational change must be, by definition, the driver and facilitator of just this sort of top-level "conversation." Without it, no change program will stay focused, integrated, and in balance. And without balance, integration, and focus, no disjointed set of initiatives will lead to significant performance-enhancing change.

Today, however, generating and capturing such quantum leaps in performance lie at the heart of many CEOs' jobs. "To meet our performance goals—or to stay ahead of the competition—we need to reinvent ourselves," they acknowledge. "Virtually everything about the way we do business must change." But if leaders are unable to translate these beliefs into a coherent basis for conversation and learning with their leadership group, then the chances of developing an effective, tangible, and manageable program of change are much reduced.

For that, the right kind of conversation is essential. Which, in turn, means having in place a shared framework for structuring activities and responsibilities, a road map for laying out their proper sequence, and a background set of guiding principles about the "natural laws" that govern organizational transformations. All three of these—framework, road map, and guiding principles—are necessary for a successful conversation, because all three have a critical role to play in giving CEOs the practical means to shepherd through a balanced, integrated change program.

Axes of change

Our experience indicates that no single type of change initiative is sufficient to bring about acceptable levels of performance improvement. Though companies spend a lot of time, money, and energy on a broad-scale quality program, or a training program, or a program to refocus their organization's culture, measurable downstream benefits—in, say, customer satisfaction or on-time delivery or cost reduction—fall well short of expectations. The inevitable result: frustration, an exhausted and increasingly cynical organization, and a deteriorating competitive position.

Examples of the failure of single-initiative "magic pills" abound. Recent work indicates that nearly two out of three companies launching quality programs to increase worker involvement are dissatisfied with their progress. Other equally well-intentioned initiatives face similar difficulties.

One industrial firm began its aggressive efforts in the mid-1980s by cascading, top down, a well-crafted vision of change throughout the company. Each mill and factory took the corporate vision and developed its own companion vision. Senior executives traveled the country describing their objectives and signaling their personal commitment. An ambitious array of corporate training programs was developed, emphasizing participative management and situational leadership skills, team development, and group problem solving. Managers were called in to head office every quarter to describe how they were implementing the program in their area.

After three years, however, it became clear that only sporadic progress was being made. Top-down assertions of the need for change were not enough. No agreed-on process existed for translating broad objectives into specific, focused performance goals at functional, plant, or machine-operation level. Nor did managers have the skills to define these goals in a way that would engage their people in finding new ways to improve performance—not once, but continually. Though the new training programs were useful, they had no vital or clear-cut connection to the primary levers of performance improvement.

Of the dozens of leading US companies that have embarked on transformational change efforts in recent years, we have looked in detail at the experience of more than thirty. Although each company's program is unique, the successful programs have developed points of view on all three types of initiative. Where any one is absent, the ill-matched collection of initiatives under way is falling short. Poor results are invariably the result of focusing efforts along only one or two—rather than all three—of the key axes of change:

  • Top-down direction setting to create focus throughout an organization and develop the conditions for performance improvement.

  • Broad-based, bottom-up performance improvement to get people at all levels to take a fresh approach to solving problems and improving performance.

  • Cross-functional core process redesign to link activities, functions, and information in new ways to achieve breakthrough improvements in cost, quality, and timeliness.

Together, these three axes make up what we think of as a "transformation triangle"—a balanced, integrated framework for combining separate initiatives into a coherent overall program.

Each axis is necessary. If top-down initiatives are lacking or faulty, managers will be left to guess where to aim new skills or activities. If bottom-up involvement is absent, motivation will falter, momentum will flag, opportunities for improvement will be overlooked, and the new skills and behavior will not be built. If horizontal core processes are ignored, function-specific efforts will never add up to the critical mass of change required. A quality program here, a new training program there, a set of internal strategy taskforces, and an executive team-building exercise may not add up to anything other than a jumble of parts that can sap, rather than build, energy. Real transformations in performance come only when efforts along all three axes are coordinated and engaged.

Top down

To develop the necessary preconditions for performance improvement, successful transformations start with clear, consistent, and ongoing direction-setting initiatives. In almost all the efforts we examined, the leadership team made a concerted effort to clarify priorities, create energy, and signal commitment to change in performance and behavior through a variety of approaches: everything from new themes and visions (General Electric's "Boundaryless Organization" or Motorola's "Six Sigma") to new measures and objectives.

Federal Express, for example, has twelve closely-watched numbers it publishes every day. These numbers reflect the corporation's customer service goals, and management has put a lot of thought into how to express them. On-time performance, for instance, is not expressed as a percentage. If it were 98.5 percent one day and 98.4 percent the next day, no one would know how to interpret the difference. So the company publishes the number as an absolute tally of late deliveries. If 1,100 packages were delivered late yesterday, everyone can understand that there were 1,100 customers who were inconvenienced or annoyed.

No single initiative offers a "magic bullet" to unfreeze and redirect an organization. What distinguishes success here is consistency among initiatives, as well as their continuing refinement and development.

Most companies start their transformation efforts with very broad objectives—say, "to lead the industry in customer satisfaction." But this is like saying "head west" as a direction for getting from New York to California. Successful efforts push over time for increasing clarity and specificity in top-down direction as change pushes toward tangibility at the front line.

At one railroad, for example, the vision and goals started broadly ("be the quality leader in the transportation industry") and became more specific ("achieve the three Rs of precision execution: right car, right train, right time") as customer needs and operational requirements came into sharper focus. This clarity helped align other change efforts to make it evident how they contributed to the overall goal. That way, a headquarters taskforce could redesign train scheduling while front-line teams attacked execution problems with individual trains.

Bottom up

Although top-down efforts create the focus and the necessary preconditions for transformational change, they alone are not sufficient to achieve it. One of the biggest challenges to overcome is the widely held management view that "all we have to do is tell employees what we want, provide some training and rewards, and change will happen." This approach may work when the desired results lie well within the existing capabilities of an organization—for instance, developing a product extension. But it falls far short when the change requires fundamentally new ways of doing business—like moving from a product to a customer orientation. In these cases, embedded skills, systems, and attitudes are usually so at odds with the new requirements that a much more intensive process is needed to retool the organization to effect lasting change.

What's needed, therefore, is to get large numbers of people throughout an organization (in operations, support units, and business management teams alike) aggressively and creatively working to improve performance. This, in turn, depends on the availability—or the creation—of disciplined processes for identifying opportunities and developing plans to close clearly identified performance gaps. Many such problem-solving processes exist, most of which are rooted in the Quality movement and share common principles: set goals, determine gaps, understand root causes, brainstorm and try out solutions, monitor results, and make adjustments.

To be truly effective, however, these approaches must be tailored to the specific challenges, skills, and change readiness of a given part of the organization. This requires, among other things, designing a methodology for setting appropriate goals and performance objectives, developing analytical templates to guide problem solving, and determining specific information needs that, of course, will vary by level and unit. For most parts of an organization, this effort will start simply and become more advanced over time.

  • Front-line operations will tend to focus on improving the cost, quality, or timeliness of products and services. At one railroad, for example, front-line teams in each terminal analyzed their operational delays and helped move on-time performance from 20 to 79 percent.

  • Staff functions will tend to work on aligning their activities to increase the value of products or services through joint efforts with front-line operations. At one insurance company, finance and human resource teams redesigned planning and compensation systems to be consistent with desired new agent behavior.

  • Management groups will tend to concentrate on identifying the most attractive performance improvement opportunities and on designing the processes to exploit them. Over a two-year period, a steel company's management team started with relatively simple efforts to improve safety and housekeeping and moved on to design advanced processes to address yields, labor productivity, and throughput time.

The net effect of launching such team-based problem-solving efforts is much like getting a flywheel spinning. Initially, tremendous inertia exists, and the first cycle can be lengthy and difficult, requiring substantial energy from outside the group to get it started. But if the process continues to be supported and rewarded by management, momentum gradually builds, improvements are achieved, the problem-solving cycle runs a more regular course, and the promise of "continuous improvement" becomes a real possibility.

Tapping the brains and energy of thousands of people is powerful in itself, but there is a second reason for using bottom-up problem solving. In many cases, you already know what needs to be done, but you don't believe that people can change their behavior just because they are told—with good reason—to do so. What does it actually take to create new behavior?

Think, for a moment, about the mechanics of a golf swing. You know you have to set up square to the target. You know you have to take the club head back slowly. You know, because a golf pro has told you at one time or another, each of the fifteen things you have to do to hit a golf ball well.

But knowing is not enough. You have to experience it. You have to be able to try it in a risk-free environment, get the feel of it. In other words, you have to go through the process of finding the right answer yourself.

For these reasons, bottom-up initiatives go far beyond the familiar "pilot testing and implementation." This is largely a function of their ...

  • Scope. In most cases, intensive problem-solving efforts ultimately have to spread across an entire enterprise. With pilots, by contrast, the normal pattern is to try them in one or two isolated locations, watch them for a year or so, and then re-evaluate the effort.

  • Objectives. Bottom-up efforts go beyond simply implementing a new solution. They have wider objectives: rapid and sustained performance improvements, development of new skills, increased change readiness, and deeper insights into how an organization must adapt to sustain the improvements.

  • Process. These efforts depend on effective "problem solving for process"—that is, developing creative ways to involve people in improving performance and redesigning their work. Again, this goes well beyond the top-down implementation of a solution defined by others.

  • Iteration. Bottom-up activities are not one-off initiatives. They call for successive rounds of effort to improve performance and build skills.

Core process

There are, however, limits to what can be achieved through both top-down and broad-based efforts that fundamentally operate within existing organizational boundaries. Some opportunities for breakthrough improvements in performance can be addressed only through a cross-functional core process redesign perspective, in which people, activities, and information are linked in new ways. The goal of CPR is to produce simultaneous, dramatic improvements (25 to 75 percent or more) in cost, quality, and time by shifting the focus of work and decision making from hierarchical channels to new horizontal flows across functions, locations, and organizational boundaries.

All companies, whether they recognize it or not, have a few (three to five) core processes that deliver the majority of an enterprise's value to its customers. In consumer electronics, for example, one core process might be product development, which links functional activities in R&D, manufacturing, marketing, and distribution to provide customers with a steady stream of innovative products.

In a sense, of course, companies have always had cross-functional initiatives. But CPR takes them much further with its intense performance orientation, its focus on the few processes that drive value and competitive differentiation, and its support for the changes in organization structure, management, and communication systems needed to "institutionalize" new levels of performance.


Top-down, bottom-up, and core process activities are not ground breaking in and of themselves. What is important is that they are:

  • Focused. Organizations can perform well with less than perfect strategies, but not with unclear objectives. Especially during periods of change, it is easy to let attention drift away from tangible performance goals toward a more general concern for effecting the necessary shifts in organizational culture. But this puts things the wrong way around. The best way to change culture is to work on improving performance at the same time.

    When, for example, management and union are at loggerheads, direct attacks on each other's entrenched position are seldom a constructive way to more forward. If, however, both sides can agree on new, shared performance goals—better on-time performance, say, or improved customer service—possible areas of cooperation begin to open up. Similarly, GE's "Workout" program may provide secondary cultural benefits in terms of how people work together. But the driving force and primary aim is to get them collaborating to solve a specific performance-related problem.

  • Integrated. All three axes are worked on simultaneously and in a way that is mutually reinforcing. Emphasis will fall in different places depending on the problem and the goal. Where the issue is to do with strategic focus or direction setting, the main action will probably be top down; where it is front-line involvement, bottom up; and where it is multiple inefficient handoffs across functions, redesign of cross-functional core processes. But wherever the main action is, transforming performance levels requires integrated effort along all three axes.

  • Balanced. Explicit attention must also be given to the relative emphasis paid to each axis. Putting too much weight on top-down efforts risks creating cynicism and confusion; excessive emphasis on bottom-up efforts means people may focus on issues that will not make any difference competitively; and a bias toward cross-functional processes could produce a solution so complex in design that implementing it is beyond an organization's capabilities.

  • Team-based. Teams are critical for all three axes. With top-down activities, it is essential to build a leadership team to integrate initiatives and lead the process; in bottom-up initiatives, there will ultimately be hundreds of performance-improvement teams working in every part of an organization; and in cross-functional efforts, process management teams have to come together across functions.

Road map

Transformational change is by its nature iterative. Although its phases overlap and interact, sequence does matter. Furthermore, you can also learn from the experiences of path-breaking companies that have preceded you. So a general road map may help managers plot their course or identify missed turnings that may be slowing progress.

1. Direction setting

The goal of phase 1, which can last anything from two to twelve months, is to kindle an urgent need for change within an organization and to articulate a new sense of direction. The best efforts involve thought leaders throughout a company (both to build support and to tap the experience of multiple constituencies) in an objective assessment of the competitive environment, the organization's current capabilities, and the outlook for its future. Whether the outcome is called a new vision, a mission, a strategic framework, or something else, what is crucial is that it states the rationale for change and defines broad performance and organizational objectives.

Key activities in this phase include:

  • Analyzing both marketplace and organization to highlight the need for change, the barriers to be overcome, and the potential payoff.

  • Structuring workshops and other forums to help first the leadership group and then the rest of the organization reach a common vision and begin to identify the actions required to make the vision a reality.

  • Examining the experience of other companies undergoing change to help build courage and conviction—and to develop insights about how the organization could evolve.

2. Process design

The goal of phase 2 is to translate the change vision into a much more specific set of performance objectives and to design processes that involve all three axes of change in order to engage the organization in achieving these goals. Senior managers almost always underestimate the importance of this structured planning phase. But it is essential if the change program is not to degenerate into a hodge-podge of well-intentioned individual initiatives.

Key activities in this phase include:

  • Creating a change organization. Few companies have succeeded without creating a special group or groups to drive their change effort. Many variations exist, "steering committees," "change czars," and "core teams" among them. The common requirement is that the group must include well-respected line and staff executives each with the depth of personal commitment to risk their careers in order to ensure a successful outcome. Their role is to help line managers focus their change efforts and to provide a forum for objective discussions of progress and lessons learned. They may also help provide centrally sponsored support for process design, facilitation, and the like.

  • Quantifying specific performance objectives. Broad statements ("lead the industry in customer satisfaction") need to be translated into the kind of measurable performance-improvement target ("reduce product development cycles from twelve months to two") that individuals or cross-functional teams can tackle. This task of translation is a central responsibility of the change leadership team.

  • Mapping objectives to organization units. It is often useful to create a "map" of an entire organization that links performance objectives to the groups of people (departments, teams, positions, or cross-functional process groups) that can most directly affect them. This map can also be used to help decide which projects to launch first.

  • Designing performance-improvement approaches. Working from its organizational "map" and its assessment of the problems and opportunities each unit faces, a change team can design the proper mix of problem-solving techniques to address such questions as: Which units need a tailored analytic exercise? Which need training? What new information must be supplied? Which core processes must be overhauled?

3. Performance improvement

In phase 3, the "transformation triangle" comes into full play. Efforts along all three axes, planned in phase 2, are now launched to begin improving performance.

  • Top down. Ongoing initiatives to build awareness and capability in support of the change process might include intensive communication efforts, training sessions, and management forums for sharing expertise and best practices in implementing the new change vision.

  • Bottom up. Starting with localized pilots to debug the methodology designed in phase 2 and then rolling out the program through the rest of the organization, teams set about evaluating current versus desired performance and developing plans to close the gap.

  • Core process redesign. Cross-functional teams tackle one or more of the core processes that senior managers identify as needing fundamental redesign.

Proper sequencing of bottom-up and core process initiatives requires careful planning. In some cases, it is best to start with the former, even if core process redesign is contemplated, because front-line teams may yield insights about current processes. In others, the structural shortcomings of the current work process are so apparent that a fundamental redesign makes more sense than trying to improve the current system.

The timing of phase 3 is usually a function of the size and complexity of the organization and the amount of resources supporting the process. In general, a unit of 20 to 50 people can go through one cycle of problem solving or work design in two to four months. However, multiple cycles may be needed as units attack increasingly complex or ambitious targets. Core process redesign efforts may take anywhere from six months to two years for full implementation, though substantial results can be realized in the first year.

Significant time is required both from the change team and from senior management to harness the "chaos" of the many initiatives which inevitably get under way. Special care is needed to make sure efforts are matched to the performance needs of each unit rather than sweeping across the system in broad waves of "one size fits all" activity.

4. Realignment

Performance-improvement efforts inevitably bring to light the size and shape of organizational barriers. They also help clarify how an organization must evolve to institutionalize or "lock in" the new capabilities that have begun to develop. Here the leadership role is to identify the needed changes systematically and take the required actions to institutionalize them.

  • Structure. Line and staff roles will be redefined as managers become more focused on adding value (and less on controlling), and clearer on the role they must play to keep change going. Elements of many traditional staff functions (such as planning, hiring, training, and purchasing) will often be absorbed into line-based teams. Autonomous, self-managing teams will be created, as individuals master multiple skills and roles. Layers will be removed as self-management takes hold.

  • Systems. Information systems will be redesigned as the information available to the front line changes dramatically and as teams better define their own information needs. These new systems will give a much clearer view of how a team's performance contributes to the whole, and will have a strong focus on customer-oriented value. Compensation systems are usually restructured at this point too. Broad-based profit-sharing approaches often give way to much more focused incentives, such as pay-for-knowledge and gainsharing.

  • Staff. A much clearer definition of the types of people and skills needed will be developed through the performance improvement efforts. Hiring and promotion requirements, as well as the balance between technical and managerial skills, will be redefined.

One interesting note: companies that wait until the end of the process to restructure these three Ss of organization are often better able to align them properly because they can calibrate against the new organizational "reality" created in phases 1, 2, and 3.

"Natural laws"

As noted earlier, effective management "conversation" about performance improvement achieved through transformational efforts reveals that the specific techniques employed matter less than does adherence to a set of underlying principles. Based on the experience of the companies we studied, we have come to define these principles or "natural laws" as:

Performance is the objective. This principle is easily forgotten, especially with today's emphasis on quality, service, empowerment, and new skills and culture. These are important objectives, of course, but they are insufficient unless management demands a rock-solid linkage between them and hard measures of performance—increased revenues, reduced costs, or better use of assets. For example, leaders must clearly understand how quality brings about increased market share in certain market segments, or how empowerment results in faster operating decisions and reduced costs.

These clear linkages are required for two reasons. First, senior managers need to quantify potential benefits so they can make informed decisions regarding levels of investment in the transformation process and its priority in their overall management agenda. If they do not, they are unlikely to maintain the focus and commitment necessary to see the transformation effort through. Second, the front line needs clear direction on where to focus and what to measure. Broad exhortations about quality or service may prompt some units to figure out the performance levers that matter to the organization as a whole, but the process is likely to be slow and uncertain. Much more effective is the simple, pointed message—say, that quality on line 4 in the factory means reduced machine downtime and improved yields.

Strategy and structure still matter. Organizations can easily grow enamored with the promise of continuous improvement, and forget that the transformation process cannot overcome fundamental strategic and structural disadvantages by itself. A company with the wrong technology, outmoded assets, an uncompetitive value proposition, or a flawed organizational structure will not—indeed, cannot—be saved by an organizational transformation effort alone. A winning strategy and a viable economic and organizational structure must underpin any transformation effort.

Teams are the key building blocks. Successful change necessitates generating quantities of extra energy throughout an organization to keep the enterprise performing while it develops the skills and capabilities it needs for the future. In practice, this means identifying team-building opportunities and creating the conditions in which teams can become genuinely high-performing units.2

Process must embody values. You cannot create a performance-oriented and empowered workforce with a change effort that perpetuates rigid hierarchies or that focuses on procedures, rather than results. Success entails designing a process that embodies the desired values. Leaders must be clear about the values they espouse, as well as about how those values link to performance. And they must ensure that their own behavior, no less than the performance-improvement effort they are managing, is consistent with those values. This could mean, for example, working to ensure that the transformation process is continually monitored for its effectiveness and corrected as necessary ("continuous improvement"), or that clear guidelines for delegated decision making are provided for down-the-line participants ("empowerment").

Learning-oriented, evolutionary effort. Organizational transformations are inherently complex, multidimensional processes. Leaders are often tempted to define a master plan, declare the planning phase complete, and delegate implementation to others. Successful initiatives are managed quite differently. Leaders recognize that the effort can never be fully planned in advance. The leadership group must learn as they go and allow for the effort to proceed in an evolutionary (and continuously improving) manner.

These efforts may start out with broad objectives and a modest process, such as benchmarking or developing a vision. But successful efforts make leaps forward in the clarity of objectives every three to six months, as experience is gained and lessons are learned. It is fine to begin with goals like "we will lead our industry in customer satisfaction." But it is not fine to stop there. Goals must, over time, become more specific: "In 1993, we will gain three points in market share by shortening delivery times from 40 to 20 days."

Focus is essential. The complexity of transformational change can easily overwhelm an organization, dissipating energy before the effort achieves its objectives. Ineffective efforts exhort the organization to "fix everything at once." Far better to choose just a few objectives at any one time (improve customer response, reduce order lead times) and devote all energy to them until measurable progress is achieved.


You are facing an exciting but challenging journey. We hope that the framework, road map, and natural laws can help you sort out the hype and buzzwords and create a better plan for transforming your organization. They can also help you make a coherent program out of a jumble of discrete initiatives. What they offer is a set of guidelines for focus, balance, and integration. Not a bad place to start.

About the author(s)

Steve Dichter is a partner in McKinsey's New York office; Chris Gagnon and Ashok Alexander are consultants in the New Jersey and New Delhi offices, respectively.

Authors' note: We wish to acknowledge the contribution of Doug Smith to this article.

More from the McKinsey Quarterly
Article - McKinsey Quarterly

The case for digital reinvention

Article - McKinsey Quarterly

The dark side of transparency

Report - McKinsey Global Institute

Harnessing automation for a future that works

Interview - McKinsey Quarterly

ING’s agile transformation