Notes from Nudgeapalooza: Behavior Change for Good

Posted by Melissa Cavanaugh on December 17, 2018.

Capital H’s first dispatch from Nudgeapalooza looked at the ways in which behavioral nudges can make a difference in influencing consumers, employees, and citizens—and where their impact might fall short. In the second keynote of the day, Katherine L. Milkman, professor at The Wharton School of the University of Pennsylvania, illuminated how her research is testing various ways to help people make positive choices and make them stick.

Milkman’s research focuses on the ways in which interventions can create lasting change in individuals’ behavior around health, financial wellbeing, and education. But the findings also have implications for organizations that hope to encourage particular behaviors in their workforce.

The Behavior Change for Good Initiative

Milkman’s quest started with a competition that offered a $100 million grant to fund a single proposal that promised “real and measurable progress in solving a critical problem of our time.” She was inspired by the fact that 40 percent of premature deaths in the United States each year are the result of behaviors that could be changed, such as smoking, exercise, diet, and failure to get recommended screenings.1 She didn’t get the grant, but she did continue with creation of the Behavior Change for Good Initiative (BCFG).

The initiative launched in a partnership with a Fortune 500 company that offered employees the opportunity to “kickstart a lasting habit” with a four-week program. Each employee gave the BCFG researchers a two-hour window in which they preferred to work out each day. Reminders to exercise were provided to each participant 15 minutes before their target window. Employees were divided into three groups based on incentive and preferred times. The control group received no incentive for exercise during their target window. A second group, the “routine” group, got a small incentive—either $3 or $7 per workout—for each day that they exercised in their target window. And a third group, the “flexible” group, got the same incentive for each workout in a week, regardless of the time of day.

Before starting the experiment, Milkman and the BCFG team polled a group of psychology professors who predicted that the “routine” group would have the best results both during and after the experiment, since they would have cemented a habit. Sure enough, the routine group with the higher, $7 incentive had the biggest increase in workouts per week during the length of the experiment. But when the team checked back several weeks later, it was the “flexible” group with the higher incentive that had the most lasting gains, since they had learned to be adaptable to changing circumstances and still prioritize working out.

This was an important finding about how to sustain behavior change. It could have implications for employers in terms of their wellbeing programs, as well as other workforce programs such as learning and development. But, as Milkman pointed out, it took four years to learn. Her next question was, “How can we learn much more, much faster?”

Partnership Reveals Effective Strategies

The answer to that question is taking shape in partnership with a variety of organizations—such as fitness chains and banks—that already collect data on individuals who are hoping to change their behavior. In pairing with data scientists who can help unpack the data, those organizations can learn how to make a bigger difference in their consumers’ lives.

An ongoing partnership with a fitness chain is similar to the one that the BCFG team conducted previously—but on a much broader scale. In addition to the control group, the partnership has 20 different experiments running concurrently. The experiments test a variety of different interventions: Is it better to do daily or weekly planning? Does making exercise fun have an impact? Does signing a pledge increase accountability? Each experiment will test its hypothesis in isolation, and then all 20 will be evaluated against each other in a mega-study.

With thousands of participants currently enrolled and tracking their workout habits through a BCFG app, the results will be extremely telling. Although final results are still being evaluated, early findings point to a few things that seem to be effective: making it fun, encouraging intrinsic motivation, and making concrete plans.

Another early indicator is that “temptation bundling”—in which individuals limit something they enjoy, like a favorite book or podcast, to their time at the gym—is proving effective. Milkman shared that she personally uses temptation bundling to motivate herself to go to the gym as well as encourage positive behaviors at work. For example, she saves lunch at her favorite burger restaurant for days that she asks mentees to join her.

Companies Look Inward to Spark Change

Such partnerships between industry and academia are yielding significant volumes of data and helping to make sense of it. In fact, in designing the experiments, Milkman interviewed product developers and software engineers to learn to design and pilot ideas quickly. Those insights helped the team make interventions digital and relatively cheap in order to scale them rapidly.

But companies are also seeking data insights within their four walls, particularly in terms of how to use data to help transition to the future of work. As leaders gain better understanding of how to move the needle on individual behavior, the next step is to understand how organizations can use that knowledge.

1 “Creating Enduring Behavior Change,” Behavior Change for Good Initiative,

Melissa CavanaughMelissa Cavanaugh leads the Bersin™ Insights organization of Deloitte Consulting LLP.

One thought on “Notes from Nudgeapalooza: Behavior Change for Good

Leave a Reply