Where Implementation Breaks Down

Where Implementation Breaks DowntcbrPDF normal

Why can’t companies get the job done?

By David A. Garvin

Just Say No

David A. Garvin is C. Roland Christensen Professor of Business Administration at Harvard Business School and author or co-author of ten books.

Executives today are enamored of innovation. Many believe that the only way to improve their company’s performance is by producing a steady stream of new products and services, identifying customers’ “pain points” and “jobs to be done” and then fulfilling unmet needs. Apple, with its iMac, iPod, iTunes, iPhone, and iPad, is the poster child for this approach done right.

Innovation is clearly essential if companies are to prosper in the long run. But they must also perform day-to-day. And there, effective implementation is the name of the game. Companies have to be able to deliver on their promises. Yet countless conversations with executives—as well as dozens of case studies I have written on firms in industries as diverse as automobiles, electronics, health care, retailing, and software—lead to the same dispiriting conclusion: Many companies flounder when it comes to executing their plans. They are unable to meet financial and operational targets; unable to roll out carefully constructed marketing, manufacturing, and sales programs; and unable to launch IT systems in a timely, cost-effective manner.

In short, they can’t get the job done.

Why not? Faced with shortfalls in performance, many executives round up the usual set of suspects. “The plan was overly optimistic,” they say. “The new technology didn’t behave as expected.” “The global economy took a turn for the worse.” “Our partners failed to give us the necessary support.” Sound familiar? It’s an all-too-common refrain, with the same underlying message: “It’s not our fault.”

Effective implementation is delivering what is planned or promised; on time, on budget, and at quality; with a minimum of variability; even in the face of unexpected events and contingencies.

There is, however, an alternative explanation—that the problems stem not from an unforeseeable future but from weaknesses in management. Imperfect foresight is not the main culprit—rather, it is flawed execution. Managers have simply not devoted the necessary time and attention to mastering the skills of implementation. In their zealous pursuit of transformation and reinvention, they have short-changed the basic blocking and tackling required to get things done. As Cassius puts it in Julius Caesar: “The fault, dear Brutus, is not in our stars, but in ourselves.”

Of course, planning and execution cannot always be neatly separated. In practice, the two activities frequently overlap; they are interdependent rather than linear and sequential. Still, identifying the distinctive tasks and challenges of implementation—the final stage of moving an idea, program, or initiative from concept to reality—is important because it allows managers to zero in on the barriers to aligned, effective action and devise ways to overcome them.

Are You Executing?

But first, what do we mean when we say that implementation has been effective? How will managers know it when they see it? Consider the following definition: Effective implementation is delivering what is planned or promised; on time, on budget, and at quality; with a minimum of variability; even in the face of unexpected events and contingencies.

This definition captures four essential truths about implementation. First, it recognizes that success must be measured against pre-established specifications, goals, or design objectives. Skill at getting things done cannot be assessed in a vacuum. Determining whether or not implementation has been effective requires comparing performance to concrete, agreed-upon deliverables—a 10 percent increase in a product’s market share, perhaps, or the training of one hundred electrical engineers in advanced circuit design. Second, the definition emphasizes that successfully delivering what is promised is not enough; the when and how of delivery are equally important. Effective implementation demands timeliness, cost sensitivity, and quality that conforms to standard. An innovative cell phone that provides all promised features but comes to market a year behind schedule does not meet this test (as Research in Motion learned with the BlackBerry 10), nor does a software installation that comes in over budget and is plagued by bugs (as unhappy customers have repeatedly told sellers of ERP systems).garvin3

Third, this definition highlights the importance of consistency as an element of effective implementation. Seldom does the success of a new program or initiative hinge on a single action or a one-time event. Rather, implementation unfolds and takes hold over time, putting a premium on coordination and repetition. Consistency is essential. Managers and employees must take the same actions and repeat the same behaviors time and time again, with limited variation. Otherwise, delivery will be uneven and ineffective. This is especially true of large, multisite organizations—retailers such as Walmart, restaurants such as Wendy’s, and banks such as Wells Fargo. These companies often roll out new initiatives simultaneously at tens or even hundreds of sites, each with subtle differences in design, equipment, and staffing. Yet the expectation is that, despite these differences, every site will introduce the initiative in much the same way and with much the same results. Consistency of execution is therefore a virtue, and excessive variability across sites is a sign of ineffective implementation.

The fourth element of the definition is likely to be the most controversial because it eliminates a popular excuse—that unforeseen events absolve managers from responsibility for implementation shortfalls. Management, as Leonard Sayles notes in Leadership: Managing in Real Organizations, is “a contingency activity; managers act when routines break down, when unanticipated snags appear.” For this reason, they can hardly be given a free pass when actions or events do not go as planned. Resilience—the ability to recover from the unexpected and redirect activities so they still yield desired outcomes—is thus a hallmark of organizations that implement effectively. So too is some degree of foresight and anticipation. While the future is uncertain, many organizations face a range of what Max Bazerman and Michael Watkins dubbed “predictable surprises”—events that are likely to happen but whose precise timing cannot be fixed with certainty. Some department will fail to meet its promised deadlines; some component will not match required specifications; some staff member will call in sick. Imagining and preparing for such contingencies is essential to getting the job done.

Taken together, these four elements create a demanding but workable definition of effective implementation. Note that despite its comprehensiveness, this definition does not equate effectiveness with omniscience on the part of managers, nor does it demand perfect planning. What it does require is an understanding of the ways that even the best laid plans can derail and how those implementation breakdowns can be prevented or overcome.


Why Things Don’t Work

Implementation requires action. But that action must be mindful and directed. Managers would be wise to heed Benjamin Franklin’s warning, “Never confuse motion with action.” Execution can go awry through misguided movement as well as inertia. Fortunately, most derailments can be mitigated or eliminated with the proper up-front attention or real-time remediation.

Lack of understanding. At times, implementation fails for the simplest of reasons: The path forward has not been clearly communicated to members of the organization. This problem takes two main forms: (1) managers and employees who do not understand the overall objective or intent of the plan and so go off in the wrong direction, and (2) managers and employees who do not understand the granular changes in behavior that are required if the plan is to be implemented effectively and so do not change their daily routines. These problems are mirror images of one another. The first reflects a lack of understanding of the big picture; the second reflects a lack of understanding of essential details.garvin4

New programs and initiatives are normally designed to serve larger ends—strategic objectives, customer needs, growth targets, and cost-reduction goals. Yet executives do not always communicate these broader imperatives when developing or rolling out implementation plans. Frequently, they fail to provide members of the organization with information on context or intent, or do so using vague, loosely defined terminology. Workers at an aging production plant may be told that they must now adopt a new manufacturing technology “to improve our competitive positioning,” while members of a rapidly growing sales force may be informed that they are now required to use a complex reporting system because of “the need for better controls.” Such vagueness is usually justified by the claim that operating personnel are likely to be distracted by strategic or contextual information; their primary focus is execution.

This approach may work when the environment is stable and unchanging, but it is deeply flawed when market, technological, social, or economic forces are in flux. Then, managers and employees often have to rework their implementation plans to adjust to emerging new realities, often quickly and without guidance from above. To make the right decisions, they require an understanding of context and intent.

The military has long understood this need. The U.S. Army’s After-Action Reviews, widely used by officers and soldiers to reflect on successes and failures immediately after a training exercise or mission and derive lessons for the future, always begin with the question, “What did we set out to do?” Lack of agreement is often the first hint of things gone wrong—it signals the leader’s failure to communicate a clear, well-specified, widely shared objective. Without such understanding, there is little basis for aligned action, especially in the face of breakdowns or unexpected events. For this reason, both Army and Marine Corps doctrine give great weight to the importance of having a clearly communicated “Commander’s Intent,” which Naval War College professor Milan Vego defines as “the description of a desired military endstate (or ‘landscape’) that a commander wants to see after the given mission is accomplished.” In training exercises, officers practice articulating intent so that their soldiers can complete missions as planned or respond quickly and appropriately to unexpected developments on the battlefield.

The lessons for corporate leaders should be obvious. “Commander’s intent” is strikingly similar to “strategic intent,” and “completing a mission” is closely akin to “implementing a plan.” In corporate as well as military settings, understanding precisely how success will be measured improves the odds of effective implementation. TopCoder Inc., an online intermediary that uses competitions within its community of more than 300,000 software programmers to produce complex code at low cost and with few bugs or last-minute revisions, has institutionalized the practice of clarifying intent. In its early conceptualization contests, positioned well before the development of detailed specifications, members of the community pose hundreds of questions to clients about their anticipated uses and needs, seeking to understand exactly what desired software programs must be able to deliver. Only then do they draft formal requirements statements and begin to write code.

Lack of understanding of the big picture is not the only communication breakdown that leads to implementation problems. Difficulties also arise when managers and employees lack an understanding of granular details. Strategies must, as Henry Mintzberg puts it, be “programmed” so that lofty objectives are translated into concrete activities. Otherwise, confusion and delay are inevitable because people will not understand what they are actually required to do. The process is one of decomposition, in which broad goals are broken down and converted to specific actions and behaviors. Skilled implementers such as Emerson Electric, Staples, and UPS routinely take this step; many other companies do not.

Lack of buy-in or commitment. At times, plans are not implemented because managers and employees have not been fully convinced of the need for new behaviors. They may lack the will or desire to change, especially if they believe that a proposed initiative is likely to impose additional work, require increased effort, or place them at a personal disadvantage. A program to improve customer satisfaction that requires front-line employees to develop relational skills—and that imposes tougher scorecards for evaluating performance—is unlikely to be implemented if employees are unenthusiastic or not fully convinced of the program’s necessity or benefits. As Donald Hambrick and Albert Cannella have noted in the Academy of Management Executive, implementation involves both “substance and selling,” and employees will expect answers to four questions: (1) Why do we need to change? (2) Why is this the right change? (3) Why do you think the organization can handle the change? And (4) What are you going to do to help me through the change?

Julie Morath, for many years COO of Children’s Hospital and Clinics of Minneapolis-St. Paul, started the process of obtaining buy-in even before she took the job. Morath was interested in introducing a patient-safety initiative but recognized that medical errors were a sensitive subject for doctors and nurses. So she began selling early and often. Morath recalled: “As part of my entry into the organization, I had carefully crafted conversations around the topic of safety with people who would have to be on board with the initiative. I did spade work, talking about how we could align the whole organization so that safety was not just a problem for people on the front lines but was owned by the administrators who designed and operated many of the hospital’s systems. I found that most people had been at the center of a healthcare situation where something did not go well. They were quick to recognize that the hospital could be doing things better than it was.”

Morath continued her education efforts during her first year as COO, sharing national data on medical accidents and convening confidential focus groups to draw out stories from the staff about their own experiences. Only after Children’s staff members were convinced of the importance of the issue did Morath launch a formal patient-safety initiative that included concrete changes in reporting, disclosure, root-cause analysis, infrastructure, and systems.

Lack of capability or capacity. Even with the best of intentions, implementation efforts can fall short if managers and employees do not have the necessary competence or skills. They may be out of date, unequipped for the challenge, or lacking in foundational knowledge. A corporate manufacturing group may have detailed plans for a promising six-sigma initiative, but if production managers lack basic math skills, are uncomfortable with numbers, or distrust quantitative reasoning, the program is unlikely to succeed no matter how well it has been designed. It is for this reason that effective implementation often begins with intensive education and training. Zensar Technologies, a mid-sized Indian software company that has for over a decade relied on small teams of young, recently hired employees, called Vision Communities, to work together for six weeks to generate high-impact strategic and organizational innovations, provides all participants with real-time training in brainstorming, mind mapping, lateral thinking, and other creativity techniques to ensure that they have the necessary toolkits for generating, developing, and implementing breakthrough ideas.

The problem of insufficient capacity exists at the organizational level as well. Senior executives often promote broad-based initiatives that require resources from multiple departments without first checking if those departments have sufficient capacity to handle the increased demands on their people or their time. If resources are lacking, implementation is unlikely to succeed. As Bob Frisch observes in Who’s in the Room?: “Failure to properly identify and manage dependencies [the costs and resources assumed to be available but carried on the budgets of other functions] is one of the primary reasons initiatives fail.”

Note that identifying and adding required resources is best done during the very earliest stages of a program or project. Doing so late in the game frequently has unintended side effects. The effort required to bring new people up to speed and ensure that they are coordinated and integrated into a process that is well under way makes schedule slippage virtually inevitable. Software engineer Fred Brooks found this phenomenon to be so pervasive that he immortalized it as Brooks’s Law: “Adding manpower to a late software project makes it later.”


Lack of aligned goals. At times, implementation founders because the key players are measuring themselves against different scorecards, pursuing incompatible targets or conflicting objectives. Members of each organization may argue for steps that are in their own best interests and resist those that are not; without some mechanism for ensuring common direction or a basis for compromise, gridlock is likely. This is a common challenge for initiatives that cross departmental or divisional boundaries. When the National Geographic Society began to pursue integrated, multimedia projects—using the same discoveries and stories as the basis for magazine articles, books, movies, cable television shows, websites, and video games—it ran into just this problem. Members of the new and old media clashed; employees described the organization as siloed and territorial, with frequent turf battles that slowed execution. The CEO responded with two, highly visible steps: He made “collaboration” the single biggest factor in calculating bonuses, and he subsequently removed two division heads, one responsible for traditional media and the other responsible for a digital product line, who were continuing to act protectively and defensively. Multimedia projects soon became a lot less contentious, and implementation improved dramatically.

Sometimes, the differences among groups are so deeply rooted that action will only be taken if decisions are elevated to a higher level. Then, effective implementation requires an escalation mechanism, as well as the existence of an objective arbiter or judge. UPS took exactly this step in 2003 when it found that several of its strategic initiatives had stalled. The CEO asked a senior executive, John McDevitt, to join the Management Committee and take charge of “strategic integration.” His role was loosely defined; UPS managers variously described McDevitt as “the champion of strategy execution,” a neutral “tiebreaker on the Management Committee,” and the executive who was “brought in to help resolve deadlocks.” McDevitt himself defined his job as providing “accountability with visibility.” He explained: “The job involved a lot of change, since various parts of the organization were working on different projects and often using competing metrics. Roles were not always consistently defined. There were also different opinions about what we were trying to accomplish and how to get these imperatives operationalized. My responsibility was to make sure that all the teams delivered what they said they were going to deliver. For instance, on Trade Direct [a new service offering that would take goods manufactured abroad and link them into the U.S. delivery system], we set up meetings with all participants and uncovered the problems. The issues were elevated to the management committee, and we made sure that accountability was assigned. I had access to all the players because I reported directly to the CEO.”

Lack of awareness of problems. No plan unfolds exactly as anticipated. At many organizations, shortfalls, delays, errors, and noncompliance are the norm, not the exception—parts arrive days late, committed new hires spurn job offers, designs remain sketchy or incomplete, and feuding departments refuse to compromise. Often, the difference between effective and ineffective implementation lies in when these problems are uncovered. Deviations found early in the process are relatively easy to fix; those found in later stages of execution (or not at all) are far more difficult to remedy.

For this reason, effective implementation requires continuous oversight and monitoring, as well as ongoing problem-solving. Managers must immerse themselves in operational details and keep feedback cycles short. To ensure faster response after the onset of the recession, the CEO of Office Depot began to review budgets monthly rather than quarterly. A general manager at a smaller firm expanded on the importance of close, attentive monitoring: “You need to keep your hand in the cookie jar. You must ‘trust but verify,’ as Ronald Reagan said. Every day The Wall Street Journal has stories on management fiascoes where top management didn’t know what was going on. They were so busy with the view from the balcony that they didn’t watch the store.”

Checklists are one simple way of identifying problem areas and making them visible to all members of a team. Early-warning systems such as red/yellow/green coding of progress are also helpful, as are regular review meetings and a culture that encourages the discussion of emerging difficulties.

Alan Mulally found all of these to be lacking early in his tenure as CEO of Ford. Mulally had come from Boeing, where early notification of problems was expected (a mid-level manager at Boeing explained, “If I’m at a status meeting and I find that someone has missed a critical milestone, the first question I ask is, ‘Why didn’t you tell me about the problem last week?’, not, ‘Why did you miss the milestone?’”). The norm at Ford was quite different: Keep problems hidden. As Bryce Hoffman describes in American Icon: Alan Mulally and the Fight to Save Ford Motor Company, formal meetings at Ford were viewed as “political theater”; the aim was to present a rosy, upbeat picture of one’s business or function whatever the circumstances. Only in side discussions were “truths too painful to put in a PowerPoint presentation shared.”

Soon after his arrival, Mulally initiated a weekly, mandatory business-plan review meeting, in which the entire senior team discussed progress against goals using a red/yellow/green coding system. To Mulally’s frustration, presenters at the early meetings continued to claim that all was well (coding their projects almost entirely green, indicating trouble-free operations) despite the company’s financial difficulties. Finally, the head of the Americas business announced at one of these meetings that his group was delaying a critical product launch because of possible suspension problems; he recalled that in the ensuing silence he thought he might be fired. How did Mulally respond? Not with criticism but with a hearty round of applause, followed by a big thank-you for the executive’s candor. Within weeks, other managers began to volunteer their problems as well.

Lack of response to changing conditions. At times, implementation fails because of rigidity—managers’ and employees’ unwillingness to deviate from pre-established plans despite changes in the internal or external environment. They continue to follow prescribed steps blindly, while struggling to convince themselves that they are still moving in the right direction. Unfortunately, the right plan for one set of conditions is seldom right for all others. Consider NASA’s insistence on proceeding with its planned Challenger launch despite unusually low temperatures.

Contingency planning is an obvious solution to this problem, as is broadening one’s planning assumptions. Before the recession, Whirlpool’s strategic plans included scenarios in which demand was 5 percent higher and lower than expected; as the crisis worsened, managers broadened their scenarios to include possible increases and decreases of as much as 15 percent.

Note, however, that the mere existence of contingency plans or alternative scenarios hardly guarantees effective implementation. Key actors must also be willing to admit when a contingency has become reality—and must then act upon that knowledge. This imposes two requirements: (1) repeatedly testing the currency and validity of one’s assumptions about the environment, and (2) building in triggers for action.

When Harvey Golub was CEO of American Express, he always asked his managers to be explicit about the assumptions they were making when formulating plans—as he put it, to describe “the rocks they were standing on”—so that in future business reviews he could assess whether the actions that were deemed appropriate during the planning stage still made sense or whether the underlying assumptions had changed sufficiently to warrant different behavior.

Xerox CEO Paul Allaire took a similar approach. He and his senior team spent a year developing twenty-eight shared “view-of-the-world” assumptions, organized into four areas—economy and society, technology and organization, markets and customers, and industry and competition—that they then used to guide the company’s strategic planning. Equally important, Allaire insisted that all assumptions be reviewed and revised on a regular basis to ensure that they remained evergreen.

Lack of discipline. Sometimes, implementation efforts start off well but then fail to stay on track. After an initial period of success, progress slows and critical tasks are left undone. There are many possible causes: unclear or shifting priorities, an overload of initiatives, an absence of reinforcement and encouragement. Managers and employees face mushrooming, constantly shifting to-do lists; they are barely able to start on one agenda item when another suddenly rises in importance. Priorities can change with dizzying speed, leading to the dreaded fad-of-the-month syndrome.

Such shifts interfere with, and frequently undermine, execution because they result in wandering attention and a loss of focus. Effective implementation requires stick-to-it-iveness and disciplined follow-through, not wavering, inconsistent commitments. It should therefore come as no surprise that a study by Steven Kaplan, Mark Klebanov, and Morten Sorensen of CEOs of buyout firms found that superior performance was positively and significantly related to CEOs’ “resoluteness”: personal traits such as persistence, proactivity, delivering on commitments, and holding people accountable, rather than team building or interpersonal skills. Reinforcement, repeated communication, and singular focus were all essential to effective implementation.

Senior executives, after all, hold a bully pulpit, and one of their primary jobs is to manage organizational attention. They do so in multiple ways. Golub, in leading a multiyear reengineering initiative at American Express that reduced costs by over $1 billion, focused attention through personal involvement. In his words: “I took ownership of the reengineering initiative; I did not delegate it. People had to feel it was their and my responsibility, and my behavior had to match the message. I went to reviews; I was present at project meetings; I attended training sessions. I contributed ideas and illustrated how to think about redesign. And I talked about reengineering all the time.”

Jack Welch of GE worked similarly, reducing distractions by focusing on one major improvement initiative at a time—Work-Out, then Change Acceleration Process, and finally, Six Sigma. Each initiative remained a primary organizational focus for several years. Staples uses a slightly different approach. It ensures that employees’ attention is focused through continuous, overlapping reinforcement by multiple levels of management, who manage by walking around. Division presidents, regional VPs, and district managers all make regular store visits of several hours’ duration. During those visits, which occur as frequently as once every one or two weeks, they walk the floor and engage managers and employees in discussions about the status of the company’s latest initiatives, such as the rollout of digital cameras or the expansion of Copy and Print Centers, to ensure that they remain a priority.

At times, implementation founders because the key players are measuring themselves against different scorecards.

Mapping Implementation

Effective implementers are cut from much the same cloth. Despite differences in specific practices, their underlying philosophies are much the same. They embrace a flexible process perspective, insist on high levels of visibility and transparency, and consider skilled management to be as important as visionary leadership.

A process perspective ensures that managers and employees view implementation as a succession of interrelated activities that take place over days, weeks, and even months, rather than a one-time event. Implementation doesn’t just happen—it unfolds. Execution is usually a drawn-out affair, which means it must be divided into discrete, separate steps. There are many opportunities for derailment—and for corrective action. Sometimes, the required tasks and activities are executed in the pre-established sequence; just as often, they are not.

Effective implementers accept this reality—that implementation is a process with a multitude of steps that need to be carefully mapped in advance (which requires high levels of granular detail), and that the predicted sequence of steps seldom unfolds precisely as planned (which requires real-time adaptability and flexibility from managers and employees).

Adaptability will occur only in an environment that encourages visibility and transparency. Effective implementers understand that deviations from plan are inevitable; the trick is identifying them early and then responding quickly. As the CFO of a leading bank observed: “Most problems don’t age well.” Skilled implementers are especially attentive to centrifugal forces such as competing visions, incompatible priorities, unshared assumptions, and suppressed disagreements that impede progress. They strive to surface these roadblocks and bring them into the open. Comfort with visibility, however, is surprisingly difficult to achieve, especially in conflict-averse or perfectionist organizations where airing disagreements or owning up to mistakes is viewed as career-limiting.

The culture that Harvey Golub faced when he became CEO of American Express had just these pernicious characteristics; to overcome them, he followed three simple rules: “Never beat up on anybody who brings bad news, never beat up on anybody who says ‘I don’t know,’ but do beat up on those who bullshit.” Like other skilled implementers, Golub recognized that shooting the messenger virtually ensures that execution efforts will eventually derail. Why? Because as scholar David Woods has observed: “You cannot solve your problems until you know what they are. And you will not know what they are unless you create an environment where people feel free to tell you.”

Finally, effective implementers celebrate management as much as leadership. In recent years, leadership skills—establishing direction, creating alignment, motivating and inspiring—have become the Holy Grail for executives, while management skills—planning and budgeting, organizing and staffing, controlling and problem-solving—have taken a back seat. Not so at companies that put a premium on skilled, disciplined execution. These organizations value executives’ ability to deliver equally if not more than their ability to craft a vision or offer strategic insights. Chuck Knight, the long-serving CEO of Emerson Electric, who produced a twenty-seven-year string of quarterly improvement in earnings per share, summarized this perspective in a Harvard Business Review article written over two decades ago: “The basis of [our long-term success] is management from minute to minute, day to day, week to week. . . . [W]hat we do at Emerson to achieve consistent performance at high levels is just solid management, rigorously executed.”

It was true then, and it remains true today: The ability to get things done—on time, on budget, at quality, and with a minimum of variability—is a cornerstone of corporate success.