The Art of Making Change Initiatives Stick

Too many managers have experienced this scenario: The chief executive announces a bold new corporate initiative aimed at generating dramatic performance improvements. The initiative calls for sweeping changes in the company’s processes, systems and culture. The launch proceeds with great fanfare and a substantial investment of the company’s resources. Several years later, however, managers look back and wonder what went wrong.

In some cases, the initiative never produces positive economic benefits; in others, hard-working employees squeeze out some short-term improvements. Management heralds the results. Yet the organization soon slips back into its old ways of doing things. Performance regresses, process improvements do not stick and dysfunctional behaviors return. The initiative fails to produce sustainable changes in processes, behavior or performance. In short, it does not alter the essential nature of the organization. Employees dismiss it as just another “flavor of the month,” and senior management loses credibility. Future initiatives are met with widespread skepticism.

That, in much-simplified and condensed form, is still what happens all too often despite decades of study of change-management disciplines, the writings of leading scholars and the legions of consultants and their change methodologies. Studies note that more than two-thirds of change initiatives fail.1 It is no surprise, then, that change-management programs are decreasing in popularity.2

Our research focused on understanding how managers can design and implement corporate initiatives in ways that produce lasting change in the architecture and fabric of the organization. (See “About the Research,”) We began this line of study by trying to apply models of programmatic change to our examination of those initiatives. The dominant view suggests that change processes unfold in three phases, typically described as “unfreeze-change-freeze.”3 First, leaders create a sense of urgency and seriously challenge existing ways of doing things. Then new processes and systems are introduced. Last, those changes are institutionalized. Our research findings generally conform to the conventional model of a programmatic change process, with one critical caveat: The seeds of effective institutionalization — the process of embedding procedural and behavioral changes in the organization’s fabric — must be planted long before the rollout of the initiative gets under way.

Four Critical Processes

Four critical processes — chartering, learning, mobilizing and realigning — enable organizations to avoid the “flavor of the month” trap and lay the foundation for the successful institutionalization of a strategic change initiative. (See “Defining the Four Antecedent Processes.”) All four elements rely on understanding the associated mix of task-related, emotional and behavioral factors to a much greater extent than is fashionable in today’s metrics-driven environment.

One point of definition: “Strategic initiatives” does not mean projects focused on one functional area or a set of loosely related activities across multiple units of the company. Nor does it mean projects launched and run locally by division managers or led largely by external consultants. Strategic initiatives are not analytical studies; they entail the broad implementation of new processes and systems. We define them as corporate programs aimed at creating new business processes or transforming existing ones to accomplish major goals, such as enhancing productivity or improving customer service. The efforts are meant to provide overarching, unifying themes that apply to many parts of the company, as seen in the four big initiatives that reshaped General Electric Co. during the 1990s.4 They are driven and monitored by an individual or team at the head office and cut across multiple divisions, businesses and functional areas.

Apparelizm Corp. is the pseudonym for a Fortune 500 retailer with more than 1,000 store locations. It has the leading share in its segment, has grown sales and earnings per share by more than 20% per year over the past decade and continues to open more than 100 new stores a year. At the time of our research, the company had begun to undertake a sweeping organizational transformation effort involving several major strategic initiatives. We began to evaluate whether Apparelizm could make its changes stick and, if so, how. Although many of the specifics of the case are unique, its broad lessons are entirely relevant to many other organizations today.

A Case of Strategic Change

One particular strategic initiative at Apparelizm yielded an opportunity for a unique comparative analysis. The retailer had launched what it called its Customer Service Enhancement initiative, or CSE initiative, in the early 1990s. After some initial success, management found that the stores soon fell back into their old patterns. By the mid-1990s, few traces of the initiative remained beyond some physical changes in store layout.

A few years ago, Apparelizm relaunched the CSE program. In spite of its long-term growth track, the company was being outpaced. Its financial performance was sagging as competition in the retail sector intensified, while its chief rival’s numbers were healthy. Apparelizm’s same-store sales (known as “comps”) had been falling steadily, and sales per square foot of retail space had fallen off over several successive quarters. Worse news was that several measures of customer service and satisfaction had begun to show significant declines. An independent market research report brought these issues into sharp relief when it exposed many of Apparelizm’s anemic numbers. One of the two co-leaders of Apparelizm’s new program had examined why the 1990s CSE effort had crumbled, and he was convinced that there were ways to avoid the problems that had doomed the earlier effort.

While the goals and objectives for the relaunch were similar to those set for the 1990s program, the results have been markedly different. There has been a fundamental change in the system by which Apparelizm’s stores are operated and customers are served. The company has adopted new approaches to inbound logistics, inventory management, merchandise replenishment and customer service, as well as to the hiring, training and supervision of store associates. Employees now spend much more of their time interacting personally with customers during peak shopping hours rather than restocking the racks or performing other operational tasks. They also tailor their interactions to deal differently with different types of customers, such as seasoned shoppers versus first-timers or people buying clothes for work rather than for leisure. Moreover, store associates now work with customers to find complete wardrobe solutions instead of simply telling them where certain items are located or answering fashion-related questions about specific items of clothing.

Today the changes in processes, systems and behavior have become part of the company’s DNA. Several metrics suggest that the CSE initiative has produced substantial improvements in customer service. For instance, customer service scores from Apparelizm’s “mystery shopper” program have risen significantly since the initiative began, more store visitors are purchasing rather than browsing and the average register receipt has increased to record levels. For the first time in several years, Apparelizm’s comps exceeded 5% on a quarter-to-quarter basis. While many factors have contributed to those increases, the CSE initiative has played a key role in driving performance improvement.

Managers at Apparelizm engaged in processes during the earliest stages of the second CSE effort that did not occur during the first effort. Our analysis confirmed that the four antecedent processes noted earlier — chartering, learning, mobilizing and realigning — help to create an organizational climate conducive to a sustainable and enduring change implementation. The four processes are necessary but insufficient conditions for success; they must be combined with a clear set of objectives, sound project management capabilities, clear lines of accountability and oversight and control systems that ensure disciplined execution.

Chartering Process: Defining Project Scope and Team Members’ Roles

The chartering process has two critical components: boundary setting and team design. Boundary setting describes the steps the team takes to define the scope of the initiative. Many comparable initiatives falter because the scope widens over time, employees’ attention is diverted, and the core team loses focus.5 So the team responsible for designing and implementing the initiative must make tough decisions about what issues and problems it will and will not attempt to address. For instance, in Apparelizm’s second CSE initiative, the core team made a clear decision to stay focused on enhancing the quality and quantity of interaction in the store aisles between employees and customers. Recognizing that employees would become confused if the initiative entailed too many active elements, the team deliberately held off on revamping the process by which customers paid for their purchases.

Team design describes how the members of the initiative’s core group define their roles and responsibilities and how they establish rules for working with one another.6 An effective lead team must be very clear, for example, about how conflict will be managed and how difficult decisions will be made. In the second CSE initiative, the core team was comprised of three relatively new hires with MBA and consulting backgrounds and two managers with extensive field experience, nicknamed the “Apparelizm Ph.D.s” for their respected track records and deep knowledge of the retailer’s operations and history. Fortunately, the two types of team members complemented each other very well.

Teams with such diverse expertise have great potential, but they often stumble if the members cannot find common ground and do not make decisions in the same manner.7 The Apparelizm team established some clear norms and ground rules in order to work together more effectively. First, the co-leaders spent a lot of time discussing how they would work together and establishing some basic principles for making decisions if the team could not reach consensus. Second, the team leaders chose to de-emphasize differences in position and status, treating all members as equals. Third, they stimulated constructive conflict and debate, encouraging members to test long-standing assumptions. Finally, the leaders established rules for debates to ensure that interpersonal conflict did not fracture the group. For example, they demanded that everyone cite specific evidence and examples when presenting their arguments, and they challenged members who offered broad generalizations that could not be substantiated. The team adopted two mottos to reinforce the rules: “Challenge every organizational sacred cow!” and “Deal in generalities, rarely succeed; deal in specifics, rarely fail!”8 The guidelines created a culture of candor that enhanced thinking, learning and decision making.9

Learning Process: Discovering What Works, Then Experimenting

The learning process involves discovery and experimentation. Discovery refers to how managers gather data and information to define the goals of the initiative and the means of achieving those objectives. Experimentation refers to an iterative process of testing and refining the components of the initiative prior to full-scale organizational rollout.10

Discovery entailed a great deal of intelligence gathering about customer service in Apparelizm’s stores. There were four critical facets. First, the core team set out to establish a baseline measure of service in the stores, using an independent agency to document the company’s performance relative to its competitors. The team leaders recognized that Apparelizm’s culture valued a “show me” rather than “tell me” approach. Detailed and accurate baseline metrics enabled the team to track and demonstrate progress frequently during the full-scale rollout, thereby building support and swaying skeptics who wondered about the efficacy of the initiative.

Second, the core team actively engaged many field employees in defining the size and scope of Apparelizm’s customer service problem, going so far as to establish an advisory committee of 12 regional vice presidents to help determine the scale and range of service difficulties. Thus, problem definition became a collective exercise during the second CSE initiative; corporate managers were not simply telling the stores that service was inadequate. The team confronted employees with raw data about service levels and asked them to draw conclusions about the data. By asking the employees to interpret the data themselves, instead of telling them that there was a serious problem, the team built a deeper shared understanding of the issues Apparelizm faced.

Third, the team spent time trying to discover why the first CSE initiative had failed. In order to avoid the same mistakes, the team leaders listened to those who had experienced the first initiative’s failure and learned that systemic change, not piecemeal change, was necessary to make long-term, sustainable improvements. They also discovered that given the company’s culture — a legacy of the founders’ passionate and intuitive style of decision making and motivation — persuading people to change would require emotional as well as rational appeals.

Fourth, the core team encouraged Apparelizm staff to experience the customer service problems themselves. Specifically, the team leaders asked senior managers to “mystery shop” the stores, an exercise that opened many eyes as customer service problems affected the managers’ personal shopping experiences. One manager even went to a competitor’s store and asked shoppers why they were not shopping at his store. Prior to that exercise, some senior executives had downplayed the notion of customer service deficiency, with one even arguing that “mystery shoppers should not and cannot tell our managers how our stores should be run.” The experience not only raised understanding and built buy-in for the initiative, it also engaged employees emotionally. They had their own stories of disappointment and frustration that they did not want customers to experience.

The second element of the learning process — experimentation — addressed how the core team ran the experiments as well as the mere fact that the group chose to test and refine their ideas before full-scale rollout. The experimentation process had four distinctive features. First, the team spent weeks in the stores, working day and night in the aisles, implementing the initiative in the test markets. Members thus saw firsthand many of the problems and obstacles that the stores would face as they tried to implement changes; they also earned the respect of store associates for “getting their hands dirty on the front lines.” They heard directly from customers and employees, getting real-time feedback. As a result, they learned early on how to target specialized training to different categories of employees, make the stores easier to navigate and work with suppliers to get product into the stores more efficiently.

Second, the core team constantly measured progress on a variety of parameters in the test markets — such as the number of store visitors converted into buyers — and let the results be known throughout the company. By providing positive results before many employees knew the details of the initiative, the core team softened and even pre-empted likely resistance when changes in store operating practices were introduced. Third, they gradually shifted responsibility for the implementation to the store managers as they moved from the first test region to the second test region to a pilot to full-scale rollout. By gradually staging the transfer of responsibility, the core team ensured a smoother transition than usually occurs.

Finally, although they chose to conduct the first test in a geographic market in which it was believed people would welcome the changes, they implemented the 40-store pilot phase in a market in which they expected resistance to be quite strong. Resistance was expected largely because the regional division comprised many of Apparelizm’s original stores and longest-serving employees —often people with steadfast beliefs in “the way we do things around here.” Why did the team seek trouble? They wanted the initiative to prove its mettle before full-scale rollout. The pilot phase helped identify many difficult hurdles before national rollout, and its success created a group of disciples who were respected throughout Apparelizm because of their status as veteran employees.

Discovery and experimentation are important to the learning process for two reasons. First, what managers learn enables them to enhance the quality of the new processes; they can anticipate obstacles more easily and develop more robust solutions. Perhaps more importantly, how managers learn proves critical to the development of employee commitment and buy-in. The ways in which the core CSE team engaged others in the learning process fostered a sense of collective ownership and a deep, shared understanding of the solutions that were eventually implemented.

Mobilizing Process: Using Metaphors and Symbolism

Mobilizing involves garnering resources and building emotional commitment to the initiative. How did the managers at Apparelizm do this effectively in the second CSE program? They engaged in storytelling and symbolic action, which helped create a compelling account of the need for the initiative and explain the specific changes that would be made.

Metaphors were used very effectively, readily conveying details of the program and triggering excitement about and support for it. One of the most powerful metaphors concerned auto racing. The core team drew a comparison between a NASCAR race crew and store staff, even producing a video to bring the metaphor to life. In the past, went the metaphor, it was as if store associates drove the race car, pumped the gas and changed the tires during a race. With the new initiative, a “pit crew” was responsible for ordering merchandise, receiving inventory and restocking shelves. The “drivers” moved around the store, helping customers. The “racetrack manager” monitored the flow of traffic in the stores, breaking up “multicar pileups” that occurred as the associates clumped together, talking to each other rather than interacting with customers. The “racetrack manager” served as a “player coach” on the floor, teaching the associates how to become more effective at consulting with customers.

The metaphor was incredibly effective because employees could readily understand how their jobs were akin to the pit crews and drivers in a sport most of them loved. They knew that NASCAR racing teams had to be fast, responsive and highly knowledgeable, and they understood that NASCAR team managers, pit crews and drivers had to communicate and coordinate their actions very closely and truly work as a team. Employees also recognized that those behaviors were just as critical to effective customer service in the stores, and they could see how the CSE initiative helped them act more like an effective pit crew. The core team, of course, drew these links very carefully to make it easy for employees to make this connection.11

Symbolic action consisted of taking steps to bolster the credibility of the core team and its message. The symbolism was a key ingredient in the effort to make the initiative appear legitimate to many of the long-term employees. Symbols mean a great deal.12 They construct simple ways for employees to understand, interpret and explain what is going on.13 In so doing, they affect how well the initiative will be received. They also affect the extent to which employees view an initiative as a sustainable, longstanding effort to which top management will remain committed.

The core team paid particular attention to a few key symbols. For example, during the first initiative in the early 1990s, “CSE” stood for “Customer Service and Efficiency,” reflecting the corporate office’s belief that some measures taken to improve customer service, particularly through a new division of labor among store associates, could also lead to greater labor productivity. In the second iteration of the initiative, the core team chose to make the “E” stand for “enhancement” to downplay the productivity theme — a symbolic move emphasizing that the first priority should be improving customer satisfaction.

The team also took great care to involve several key “influencers” within the organization, “icons” whose highly respected track records stretched back to Apparelizm’s early days. They brought in those individuals — the two founders and some of the original employees — at critical moments to offer their testimonials and endorsements. The influencers shared storied moments from Apparelizm’s past, particularly instances of heroic efforts to serve customers well, and they described the current initiative as an opportunity to return to the company’s roots. This framed the initiative as a return to past glory rather than a challenge to its valued tradition and heritage. The team also worked with Apparelizm’s CEO to ensure that he would talk about the initiative periodically in public forums, such as discussions with the press and Wall Street analysts. These symbolic acts enhanced the credibility of the core team and solidified employee commitment to the initiative.

Realigning Process: Changing the Jobs and the Performance Metrics

The realignment process includes two important sets of activities: redefining roles and changing the way employees’ performance is monitored and measured. In the first CSE initiative, employees had been asked to perform new tasks, but they had not viewed their roles as having changed materially. The new CSE team recognized that the underlying structure of many store associates’ jobs had to be changed if customer service was to improve. New store positions also had to be created, and many aspects of a store manager’s job had to be redefined. Significantly, the team connected new role and job definitions to the metaphors used in the mobilizing process. For instance, the “racetrack manager” in fact became an official position in the stores, with the job description calling for that person to provide on-the-job training in customer service. Other employees became full-time members of the “pit crew,” dedicated to goods receiving and inventory replenishment. The jobs of many others were redefined to cover fewer operational tasks and place much more emphasis on selling and service.

The CSE team also worked with Apparelizm’s human resources department to alter screening, hiring and promotion practices and retrain employees to understand their new roles and acquire the necessary skills to fulfill them. The team leaders learned that it was critical to identify the people who could perform operational tasks very capably but neither enjoyed nor excelled at service and selling, and vice versa. Then they worked with store managers to help match people and positions effectively.

Changes in job definitions required an overhaul of the ways in which employees and managers were evaluated, which in turn called for new metrics. The metrics did not evaluate just sales data, because the team recognized that many factors could affect a store’s revenues. Moreover, store associates often did not see the connection between their behavior and aggregate sales data. The new measures instead examined very specific aspects of employee behavior and store appearance that were believed to be antecedents to improved customer satisfaction and increased sales. For instance, they examined the number of “greets” — that is, the number of times an associate approached a customer to ask if he or she needed assistance, rather than waiting for a request for help. The new metrics also tracked the number of times an associate walked with a customer to another area of the store rather than simply pointing the customer in that direction. And they monitored the frequency of a store’s “multicar pileups.”

The metrics made the initiative’s priorities clear and tangible, reinforcing the initiative’s permanence because the metrics had to be taken seriously. Managers could see the benefits that the CSE initiative was generating. The metrics also enabled the core team to provide evidence of small wins — near-term improvements — during the early stages of the rollout process, before the improved service began to translate into significant sales increases. In so doing, the team garnered support at multiple organizational levels.14

Redefining roles and reworking monitoring and measurement systems ensured a high degree of alignment within the company. Employees did not find themselves struggling with one set of demands from their own supervisors and other demands from the CSE team. All of the human resources practices and procedures reinforced the new behaviors that the core team expected from store managers and employees. Moreover, by matching employees to the jobs that best fit their skill sets and by retraining others, the CSE group minimized the resistance to change that often is caused by concern about competence in a new work environment.15

Creating the Enabling Conditions

To embed change in an organization, managers have to create the conditions that enable employees to take ownership of the new procedures and systems and integrate and apply the key principles of the initiative to the way day-to-day work is done. The chartering, learning, mobilizing and realigning processes help to create those conditions. These enabling conditions occur in three contexts: structural, procedural and emotional. (See “Laying the Foundation for Enduring Change.”)

Structural Context

The four antecedent processes reshape the company’s structural context — its reporting relationships, monitoring and control systems and reward and punishment systems that collectively affect behavior. This structural context influences how people choose to spend their time, what kinds of behaviors they choose to engage in and what types of projects they wish to pursue.16 Apparelizm’s CSE team primarily shifted the company’s structural context during the realigning process, particularly in terms of the array of new human resource practices. The learning process helped the team see how the structural context needed to change, and it began to redefine the relationship between the corporate office and field personnel. The storytelling activities and metaphors of the mobilizing process enabled the team to communicate how roles and responsibilities would change for the store managers and frontline employees. And the chartering process created a new culture of teamwork at the top. Gradually, those norms and ground rules filtered down to the rest of the organization during the initiative’s rollout.

Procedural Context

The procedural context refers to the perceptions of procedural fairness and legitimacy that emerge as new processes and systems are introduced. If employees believe the process of formulating an initiative has been fair, they are more likely to accept and adopt the changes it involves.17

Procedural fairness does not mean that executives abdicate their decision-making authority or that employees always get what they want. It does mean that people have had a chance to voice their opinions about the initiative and that they believe their views have been considered carefully and they have had a genuine opportunity to influence the design and rollout of the initiative. Procedural fairness is particularly important in soliciting commitment when a company’s leadership is asking staff to implement ideas and proposals that they may not have endorsed initially.18

Procedural legitimacy is equally critical. It means that employees believe the change initiative fits with the tenets of what the company stands for and how work should be done.19 If employees believe that the new processes and systems are legitimate and that the proponents of the changes are credible, they are more likely to accept and adopt the planned changes.

Apparelizm’s CSE team used the learning process to foster a sense of procedural fairness, engaging frontline employees such as store associates in ways that made them feel “heard.” Importantly, the team’s efforts did not come across as a “charade of consultation.”20 The CSE change leaders also bolstered perceptions of legitimacy through the chartering and learning processes. Their own credibility passed all the tests: Their track records and reputations had laid the groundwork, and their hands-on involvement in the stores during discovery and experimentation showed they could make timely and effective decisions as a team and be honest about the size and scope of the customer service problem. The team used symbolic action to show that the CSE initiative was consistent with many of the core values of the company and its founders, as well as consistent with ideals that were deeply rooted in people’s recollections of the organization’s storied history.21 Although symbolic action is clearly connected to the emotional context, it is also an important step toward establishing procedural legitimacy.

Emotional Context

Most substantive corporate changes stir up a broad array of feelings and emotions.22 They often make employees feel more hopeful and energized about the company’s future and emotionally invested in the implementation effort. At the same time, change initiatives often trigger widespread fear. Proponents of major change must create a climate in which employees’ fears are allayed, while their enthusiasm for their day-to-day work and for the company’s larger goals is aroused and channeled effectively.23

At Apparelizm, the chartering and learning processes helped to evaluate the customer-service problem and make it abundantly clear that major changes were necessary. The realigning process forced people to come to grips with new roles and responsibilities. Although those activities created anxiety and discomfort, they also helped build dissatisfaction with the status quo — a vital element of any change effort. The mobilizing process provided the hope. It generated positive excitement and tapped into the “can do” attitude of the store employees. The use of stories and metaphors energized staff, built commitment and made it easier for them to understand the nature of the changes being instituted. In short, the CSE team used emotion to ease the tough cognitive task of comprehending and digesting complex procedural changes.

OUR FINDINGS UNDERSCORE the importance of acknowledging and accommodating behavioral and emotional factors in any large-scale change initiative. They also show a significant departure from the conventional wisdom about programmatic change. Most prior research has emphasized early articulation of a sense of urgency, with institutionalization of new processes and behaviors occurring much later. But we found that for widespread changes to stick, managers need to put a series of processes in motion right at the beginning. A focus on four core processes at the outset will help to create an environment that is receptive to change — and one in which people do not drift back to old ways of working. As the Apparelizm scenario indicates, change leaders must reshape the organizational structure, take the lead with rollout processes that are perceived to be fair and legitimate and employ a range of open approaches to engage people’s emotions if they want strategic initiatives to become more than the latest flavor of the month.