McKinsey Quarterly

The four building blocks of change

| Article

Large-scale organizational change has always been difficult, and there’s no shortage of research showing that a majority of transformations continue to fail. Today’s dynamic environment adds an extra level of urgency and complexity. Companies must increasingly react to sudden shifts in the marketplace, to other external shocks, and to the imperatives of new business models. The stakes are higher than ever.

So what’s to be done? In both research and practice, we find that transformations stand the best chance of success when they focus on four key actions to change mind-sets and behavior: fostering understanding and conviction, reinforcing changes through formal mechanisms, developing talent and skills, and role modeling. Collectively labeled the “influence model,” these ideas were introduced more than a dozen years ago in a McKinsey Quarterly article, “The psychology of change management.” They were based on academic research and practical experience—what we saw worked and what didn’t.

Digital technologies and the changing nature of the workforce have created new opportunities and challenges for the influence model (for more on the relationship between those trends and the model, see this article’s companion, “Winning hearts and minds in the 21st century”). But it still works overall, a decade and a half later (exhibit). In a recent McKinsey Global Survey, we examined successful transformations and found that they were nearly eight times more likely to use all four actions as opposed to just one.1 Building both on classic and new academic research, the present article supplies a primer on the model and its four building blocks: what they are, how they work, and why they matter.

Fostering understanding and conviction

We know from research that human beings strive for congruence between their beliefs and their actions and experience dissonance when these are misaligned. Believing in the “why” behind a change can therefore inspire people to change their behavior. In practice, however, we find that many transformation leaders falsely assume that the “why” is clear to the broader organization and consequently fail to spend enough time communicating the rationale behind change efforts.

This common pitfall is predictable. Research shows that people frequently overestimate the extent to which others share their own attitudes, beliefs, and opinions—a tendency known as the false-consensus effect. Studies also highlight another contributing phenomenon, the “curse of knowledge”: people find it difficult to imagine that others don’t know something that they themselves do know. To illustrate this tendency, a Stanford study asked participants to tap out the rhythms of well-known songs and predict the likelihood that others would guess what they were. The tappers predicted that the listeners would identify half of the songs correctly; in reality, they did so less than 5 percent of the time.2

Therefore, in times of transformation, we recommend that leaders develop a change story that helps all stakeholders understand where the company is headed, why it is changing, and why this change is important. Building in a feedback loop to sense how the story is being received is also useful. These change stories not only help get out the message but also, recent research finds, serve as an effective influencing tool. Stories are particularly effective in selling brands.3

Even 15 years ago, at the time of the original article, digital advances were starting to make employees feel involved in transformations, allowing them to participate in shaping the direction of their companies. In 2006, for example, IBM used its intranet to conduct two 72-hour “jam sessions” to engage employees, clients, and other stakeholders in an online debate about business opportunities. No fewer than 150,000 visitors attended from 104 countries and 67 different companies, and there were 46,000 posts.4 As we explain in “Winning hearts and minds in the 21st century,” social and mobile technologies have since created a wide range of new opportunities to build the commitment of employees to change.

Reinforcing with formal mechanisms

Psychologists have long known that behavior often stems from direct association and reinforcement. Back in the 1920s, Ivan Pavlov’s classical conditioning research showed how the repeated association between two stimuli—the sound of a bell and the delivery of food—eventually led dogs to salivate upon hearing the bell alone. Researchers later extended this work on conditioning to humans, demonstrating how children could learn to fear a rat when it was associated with a loud noise.5 Of course, this conditioning isn’t limited to negative associations or to animals. The perfume industry recognizes how the mere scent of someone you love can induce feelings of love and longing.

Reinforcement can also be conscious, shaped by the expected rewards and punishments associated with specific forms of behavior. B. F. Skinner’s work on operant conditioning showed how pairing positive reinforcements such as food with desired behavior could be used, for example, to teach pigeons to play Ping-Pong. This concept, which isn’t hard to grasp, is deeply embedded in organizations. Many people who have had commissions-based sales jobs will understand the point—being paid more for working harder can sometimes be a strong incentive.

Despite the importance of reinforcement, organizations often fail to use it correctly. In a seminal paper “On the folly of rewarding A, while hoping for B,” management scholar Steven Kerr described numerous examples of organizational-reward systems that are misaligned with the desired behavior, which is therefore neglected.6 Some of the paper’s examples—such as the way university professors are rewarded for their research publications, while society expects them to be good teachers—are still relevant today. We ourselves have witnessed this phenomenon in a global refining organization facing market pressure. By squeezing maintenance expenditures and rewarding employees who cut them, the company in effect treated that part of the budget as a “super KPI.” Yet at the same time, its stated objective was reliable maintenance.

Even when organizations use money as a reinforcement correctly, they often delude themselves into thinking that it alone will suffice. Research examining the relationship between money and experienced happiness—moods and general well-being—suggests a law of diminishing returns. The relationship may disappear altogether after around $75,000, a much lower ceiling than most executives assume.7

Would you like to learn more about our People & Organizational Performance Practice?

Money isn’t the only motivator, of course. Victor Vroom’s classic research on expectancy theory explained how the tendency to behave in certain ways depends on the expectation that the effort will result in the desired kind of performance, that this performance will be rewarded, and that the reward will be desirable.8 When a Middle Eastern telecommunications company recently examined performance drivers, it found that collaboration and purpose were more important than compensation (see “Ahead of the curve: The future of performance management,” forthcoming on McKinsey.com). The company therefore moved from awarding minor individual bonuses for performance to celebrating how specific teams made a real difference in the lives of their customers. This move increased motivation while also saving the organization millions.

How these reinforcements are delivered also matters. It has long been clear that predictability makes them less effective; intermittent reinforcement provides a more powerful hook, as slot-machine operators have learned to their advantage. Further, people react negatively if they feel that reinforcements aren’t distributed fairly. Research on equity theory describes how employees compare their job inputs and outcomes with reference-comparison targets, such as coworkers who have been promoted ahead of them or their own experiences at past jobs.9 We therefore recommend that organizations neutralize compensation as a source of anxiety and instead focus on what really drives performance—such as collaboration and purpose, in the case of the Middle Eastern telecom company previously mentioned.

Developing talent and skills

Thankfully, you can teach an old dog new tricks. Human brains are not fixed; neuroscience research shows that they remain plastic well into adulthood. Illustrating this concept, scientific investigation has found that the brains of London taxi drivers, who spend years memorizing thousands of streets and local attractions, showed unique gray-matter volume differences in the hippocampus compared with the brains of other people. Research linked these differences to the taxi drivers’ extraordinary special knowledge.10

Despite an amazing ability to learn new things, human beings all too often lack insight into what they need to know but don’t. Biases, for example, can lead people to overlook their limitations and be overconfident of their abilities. Highlighting this point, studies have found that over 90 percent of US drivers rate themselves above average, nearly 70 percent of professors consider themselves in the top 25 percent for teaching ability, and 84 percent of Frenchmen believe they are above-average lovers.11 This self-serving bias can lead to blind spots, making people too confident about some of their abilities and unaware of what they need to learn. In the workplace, the “mum effect”—a proclivity to keep quiet about unpleasant, unfavorable messages—often compounds these self-serving tendencies.12

Even when people overcome such biases and actually want to improve, they can handicap themselves by doubting their ability to change. Classic psychological research by Martin Seligman and his colleagues explained how animals and people can fall into a state of learned helplessness—passive acceptance and resignation that develops as a result of repeated exposure to negative events perceived as unavoidable. The researchers found that dogs exposed to unavoidable shocks gave up trying to escape and, when later given an opportunity to do so, stayed put and accepted the shocks as inevitable.13 Like animals, people who believe that developing new skills won’t change a situation are more likely to be passive. You see this all around the economy—from employees who stop offering new ideas after earlier ones have been challenged to unemployed job seekers who give up looking for work after multiple rejections.

Instilling a sense of control and competence can promote an active effort to improve. As expectancy theory holds, people are more motivated to achieve their goals when they believe that greater individual effort will increase performance.14 Fortunately, new technologies now give organizations more creative opportunities than ever to showcase examples of how that can actually happen.

Role modeling

Research tells us that role modeling occurs both unconsciously and consciously. Unconsciously, people often find themselves mimicking the emotions, behavior, speech patterns, expressions, and moods of others without even realizing that they are doing so. They also consciously align their own thinking and behavior with those of other people—to learn, to determine what’s right, and sometimes just to fit in.

While role modeling is commonly associated with high-power leaders such as Abraham Lincoln and Bill Gates, it isn’t limited to people in formal positions of authority. Smart organizations seeking to win their employees’ support for major transformation efforts recognize that key opinion leaders may exert more influence than CEOs. Nor is role modeling limited to individuals. Everyone has the power to model roles, and groups of people may exert the most powerful influence of all. Robert Cialdini, a well-respected professor of psychology and marketing, examined the power of “social proof”—a mental shortcut people use to judge what is correct by determining what others think is correct. No wonder TV shows have been using canned laughter for decades; believing that other people find a show funny makes us more likely to find it funny too.

Today’s increasingly connected digital world provides more opportunities than ever to share information about how others think and behave. Ever found yourself swayed by the number of positive reviews on Yelp? Or perceiving a Twitter user with a million followers as more reputable than one with only a dozen? You’re not imagining this. Users can now “buy followers” to help those users or their brands seem popular or even start trending.


The endurance of the influence model shouldn’t be surprising: powerful forces of human nature underlie it. More surprising, perhaps, is how often leaders still embark on large-scale change efforts without seriously focusing on building conviction or reinforcing it through formal mechanisms, the development of skills, and role modeling. While these priorities sound like common sense, it’s easy to miss one or more of them amid the maelstrom of activity that often accompanies significant changes in organizational direction. Leaders should address these building blocks systematically because, as research and experience demonstrate, all four together make a bigger impact.

Explore a career with us