McKinsey Quarterly

Lessons from the front line of corporate nudging

| Article

If you’re serious about setting up a behavioral-science team—or a nudge unit, as we’ll more colloquially refer to it in this article—you need to ask yourself some tough questions, such as what it should do, where it should sit, how you’ll know it’s succeeding, and whether you’re ready for the ethical tensions it may raise.

Subtle interventions to help people make better decisions are hardly new. Since the 1950s, behavioral scientists, using a mix of economics and psychology, have studied human irrationality and devised ways both to improve the choices made by consumers and influence how employees react in the workplace. Increasingly, over the past two decades, companies have used the insights of behavioral science to reduce bias in boardrooms, improve strategic decision making, provide benefits for customers, enhance the effectiveness of marketing campaigns, and avoid making bad bets on major acquisitions or investments.

The publication, about ten years ago, of Cass Sunstein and Nobel prize winner Richard Thaler’s book Nudge: Improving Decisions About Health, Wealth, and Happiness (Yale University Press, 2008) increased the public sector’s focus on these issues. The UK government established a nudge unit (which Thaler helped set up) and famously used behavioral science to encourage citizens to pay their taxes on time and to vote, as well as to address public-health issues. Governments in Australia, Denmark, and Singapore have also been enthusiastic nudgers in selective policy areas.

The word nudge has since caught on in the corporate world and, along with better data analytics and improved intervention techniques, has helped accelerate the advance of behavioral science into the corporate mainstream. As nudge units sprout up in the private sector, they are helping companies promote change and increase productivity on the factory floor, design better products, drive higher sales, and enhance decision-making processes (exhibit). Nudge units create win–win outcomes for companies, employees, and customers. One insurance company, for example, has used nudges to great effect to promote the advantages of using a partner network of car-repair shops; another has encouraged customers to exercise and lead healthy lifestyles, thereby reducing claims. A German utility company has used related techniques to combat irrational decision making among employees at all levels of the organization.

Companies are using behavioral science to improve customer and employee behavior and well-being.

To gain a better understanding of how to build a successful nudge unit, we recently talked to 14 experts who have led initiatives in sectors from financial services and healthcare to advertising and retail (see sidebar, “Fourteen experts forging the future”). Most stressed that while nudging is a catchy term, it does not do full justice to the broad applications of behavioral science to the businesses for which they and their units are responsible. Behavioral science, for instance, encompasses debiasing and other tools for driving behavioral change, including incentives, education, and awareness.

That said, nudge and behavioral-science units, with their scientific approach—evaluating interventions based on data and experimentation—should also be distinguished from the ad hoc bundle of marketing communications, HR, and training initiatives that has often sought to change customer or employee behavior in the past.

Combined with the best practices we’ve already gleaned from working with companies, here are some key takeaways from the conversations we had with the 14 experts on strategic choices, organizational structure, team composition, measurement challenges, and ethical considerations. Our experience, and that of our interviewees, strongly suggests that there is no one-size-fits-all approach for creating effective nudge units. But there are some common challenges and trade-offs that most organizations will face—hence these six questions that any leader seeking to launch or scale a nudge unit should ask.

1. Where should the nudge unit focus?

A first step is to clarify the value proposition of a nudge team: Will it be employee focused (dealing with questions such as motivation, better decision making in boardrooms, healthier eating, and healthier living), customer focused (dealing with issues such as encouraging better pension provision or infusing behavioral science into the marketing mix), or a combination of the two? Most of our interviewees said they focus on customer behavior, but some cited applications to “increase the internal capabilities of employees” and to “learn about employee engagement and performance.” To span the experiences of customers and employees, teams must pursue initiatives in areas as diverse as in-store service, digital operations, and HR processes (a trifecta advanced by one nudge unit we learned about). And, of course, the teams focused on customer behavior must also involve employees—training them and securing their buy-in to experimentation with proposed nudges.

Some nudge units we know are opportunistic, responding to different demands across their businesses. The vast majority of those we talked to, though, insisted that integrating behavioral science into long-term strategy delivers superior results. “The best way to make better products or programs is to build them with an accurate behavioral foundation,” explained one. “That’s why we work on projects from the very beginning, helping teams better define their strategy, and then design and run experiments to determine how to get the desired behavioral outcomes.”

2. Where in the organization should the nudge unit sit?

The strategic purpose of a company’s nudge unit will likely determine its location in the structure of the organization. Some businesses have small targeted units within areas such as R&D or marketing, or they house their behavioral-science and data-science teams together to capture synergies. Others seek to reap broader benefits by creating a global nudge-operations hub with a cross-functional remit, often reporting directly to senior leadership.

Discover and subscribe to McKinsey Quarterly Audio

Five ways to subscribe:

Where the unit sits is important. One nudge-unit leader said her team recently benefited after moving away from the corporate center so as to be closer to products and services. “This meant we had greater access to customers and have been able to add more value to them in a tangible way.” Another practitioner agreed that the right place for a customer-focused behavioral-science team is the product-management domain—putting a nudge unit in the data-excellence center (as favored by some companies) risks a stronger emphasis on analyzing patterns of past behavior rather than coming up with fresh, business-relevant ideas and rapidly testing them.

No matter where the team is located, the practitioners we spoke with agreed that flexibility and integration with other methods of behavioral change were critical: behavioral scientists, said one, gain from the insights of fields such as cognitive neuroscience, social psychology, audience profiling, and segmentation using personality-trait science, cultural cognition, and cognitive styles. Behavioral science “should be part of everything we do,” he added. “There is probably no challenge that could not be rendered more effective through its use.”

3. How to populate the nudge unit?

Hiring the right people, whether inside or outside the organization, is a critical path to success. The majority of respondents did both, with most expressing a preference for going outside, at least at the outset when foundational skills might not yet be in place.

Most of the experts we talked to work in units with three to eight members. One consists of a single person, “very highly leveraged” by “product managers and designers across a lot of independent teams.” The larger organizations had as many as 50 in their behavioral-science teams, often spread among different geographic locations. The only nudge-unit head who unequivocally said she had solely recruited internally did so because “institutional knowledge” was critical in her organization, and the learning curve for acquiring it extremely long.

Under just about any set of circumstances, finding the right talent takes time. “Our first recruits came from five different continents, and it took us over a year to identify, recruit, and relocate them,” said one interviewee.

Many companies say the best teams mix behavioral scientists with specialists in other areas, such as psychology, marketing, and advanced data analytics. “Our people must bridge the worlds of behavioral science, marketing, and communications,” said one of the practitioners we spoke with.

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

PhDs are often favored, while some respondents pointed out that people with practical experience in experimentation and a relevant academic background (for example, in psychology) can be very useful in certain roles, even if they don’t have an advanced degree. Personal skills, such as curiosity, a can-do attitude, a talent for problem solving, a willingness to “own” solutions, and the ability to communicate with stakeholders at various levels of the organization, are as helpful in nudge units as they are in most corporate contexts. “People who are entrepreneurial are key,” said one interviewee. “Specifically, we find that 90 percent of the [unit’s] effort is in implementing the behavioral design correctly while surmounting various operational, financial, legal, buy-in, and other barriers that come up.”

A number of our respondents also suggested that links with academia are helpful, especially in the early days, when mentions in academic journals help give corporate initiatives credibility both inside and outside the organization.

4. What is being done to set up the unit for success?

Cross-functional involvement is key to any behavioral-science agenda. “In my experience of putting on training workshops for large companies, the more people who attend from different functions the better,” observed one practitioner. “This is primarily to create a shared language among people at the company. We do this for every project kick-off—the more people attending the better. Data scientists and researchers are especially important, as they can support the initial behavioral diagnosis efforts to uncover customer-behavior insights and report on experimental efforts. Designers are critical, as they help implement the details required to drive behavior change.”

Misunderstandings about the true potential of behavioral applications—combined with patchy support at senior levels and pressure for short-term results—can make the early days for a new team very difficult. That’s why it is important to articulate clearly a vision and value proposition, and to communicate them broadly before rushing off to initiate projects with “low-hanging fruit.” One team we know acknowledges it was too hasty to roll out behavioral science across the organization before expectations were set; the result was a bottleneck in the approval and delivery of projects. It recommends that units do a soft launch in the first 90 days to ensure the “right scaffolding” for the team.

“We grossly underestimated the challenge of implementing even seemingly simple innovations in large organizations,” said another team leader. “We initially went after quite ambitious innovations, and then had to abandon them.”

Comfort with experimentation is critical in every behavioral-science effort. The academic approach, after all, requires thorough testing through randomized controlled trials, particularly, as one respondent pointed out, in complex domains, such as finance, when behavioral interventions can produce unexpected second-order effects. Randomized controlled trials should also be attempted as much as possible. In practice, of course, it’s not possible to test every nudge thoroughly, but our practitioners offered a number of pragmatic solutions:

  • pursuing simple, straightforward interventions (nudges directly implemented, with their results compared against stable historical baselines)
  • piloting nudges using limited sample sizes in preparation for a proper test
  • employing “lean experiments” for simple proceed-or-stop decisions

One respondent said being humble about what you really know about human behavior, along with a thirst for experimentation as a means to finding out what truly works, are among the most critical success factors for nudge efforts.

5. How to demonstrate impact?

Nudge teams won’t win support from the top, or gain the trust of other teams, if they don’t provide evidence of their impact. That means translating changes in behavior, unearthed through data monitoring, into measurable value—for instance, increased retirement savings of employees.

Stressing the financial benefit, though, is not the only way to formulate and communicate results. In our experience, it’s also possible to inspire people across the organization to support behavioral-science initiatives with news of a specific outcome (such as “increased retention by 10 percent among students at risk of dropping out of a university”) or through an impressive anecdote. One insurance company says that after it implemented behavioral measures to increase trust, customers who felt they had been excessively reimbursed for their claims (in one case for a broken pipe) actually started sending money back to the company.

How to nudge your way to better performance

How to nudge your way to better performance

Most of our experts acknowledged that it takes time to demonstrate impact, that internal skeptics can slow things down, and that it is hard to measure some outcomes directly. “Analytics is integral to our efforts,” says one nudge-unit head. “Our analytics team is involved long in advance of a project, acting as a reconnaissance unit to get insights into certain behaviors ahead of time; then, later on, they help us determine whether our interventions were successful.”

Nudge units need to bring other colleagues with them. “If I had it to do over again,” said one respondent who admitted scaling up an initiative too quickly, “I would have spent more time with senior leaders, external stakeholders, and staff to understand ways of working, constraints, and organizational priorities. A small number of manageable and highly visible projects would be very carefully selected and delivered with enough time and resourcing to demonstrate the unique value of the team. Once these new elements were in place, the team would be in a position to conduct a more ambitious profile-building program.”

6. Are ethical challenges being faced head on?

Some people worry that nudging may be a paternalistic practice or, worse still, open to manipulation. An important task of the nudge unit is to deal with this responsibly. Some organizations in recent years, for example, stand accused of harnessing predictable human errors for their own gain, sometimes in covert ways. One respondent said that he had seen several applications of behavioral science to sales campaigns, which showed little regard to whether the approach was helping solve the needs of customers or whether customers understood what was happening.

Proponents of nudging, on the other hand, argue that many nudges, like timely educational prompts, reminders, calorie labels, or active choice forms, can encourage positive behavior. Some have set up ethical codes to cover the work they do, including that with academic partners. One formal document, for example, titled “Ethical imperatives,” spells out the twin principles of intent (why you are doing it and for whose benefit) and respect (governing how the action is carried out). Another interviewee said it was “always right to worry about the incentive model” and ensure that actions to deliver value to shareholders were aligned with delivering value to the customer. “If done ethically, the nudge unit can be a force to help companies improve customer outcomes and drive measurable increases in well-being,” she added. The ultimate test of nudges, another added, “is always whether they would stand the test of revelation to the public—would the public be happy with what you are doing?”

Companies, perhaps concerned by perceptions of the “dark side,” have generally been more coy than the public sector about communicating the positive benefits of nudging. Yet publicizing the work of a nudge unit can promote a reputation for innovation. A global bank, for example, received positive press coverage for its nudging finance app that promoted healthy financial practices.


Creating an effective nudge unit requires much more than hiring a few experts who understand heuristics and statistics. It’s up to senior management to create the conditions for success by helping to focus the unit, situate it in the organization, celebrate its impact, and hold it to high ethical standards. Leaders who tackle these challenges boost the odds that the unit’s nudges will cause real movement in the organization, and in its performance.

Explore a career with us