The Future of Nudging: Crisis or Transformation

Over the past 14 years, nudging has become a popular method within organizations, government and academia, but how effective is it when it comes to changing behavior? Recent research shows that the effect of nudging might be much lower than we previously thought. This has caused quite a stir, and rightly raises questions about the future of nudging. But as is often the case with something as complex as human behavior, the situation may not be as black and white as it may seem at first glance.

What is nudging again?

From fresh fruit and vegetables at eye level and the famous fly in the urinal, to the reminder for a deposit on small plastic bottles: nudges are behavioral interventions that help us make better choices. A push in the right direction, without limiting freedom of choice. Think of an elephant mom who subtly nudges her child in the right direction.

Since the publication of the book Nudge by Richard Thaler (University of Chicago) and Cass Sunstein (Harvard University) in 2008, hundreds of studies have been published in which nudges are tested. governments, for-profits and non-profits around the world work with official Nudge Units to positively influence behavior.

Is there a nudge crisis?

In January 2022, PNAS, the official body of the US National Academy of Sciences, published an analysis examining more than 200 nudge studies and looking at the overall effect size. Also known as meta-analysis, this form of research demonstrated that nudging has an average effect size when it comes to behavioral change. In social science, this is robust and even higher than other behavioral change interventions such as communication campaigns.

The authors did emphasize the problem of publication bias; Research that can demonstrate certain results is more likely to be published than research that produces negative or inconclusive results. In the case of nudging, this would mean that we mainly see results in which nudging works, while research in which nudging does not work disappears into the bottom drawer.

In July 2022, things got even more intense. PNAS published a response to the first study in which Eric-Jan Wagemaker (UvA) among others argued that if publication bias is taken into account in the analysis, the total effect of nudging is virtually zero.

Does nudging work or not?

This news about nudging understandably raises a lot of doubts and demonstrates that publication bias is a real problem (that impacts almost all disciplines of science). Yet there is another side to this story: Also in January 2022, The Office Evaluation Sciences and The US Office of the Behavioral Insights Team (two leading behavioral change institutions) published a similar study in which a strong positive overall effect of nudging was measured. To exclude publication bias, the data and results of 350 nudging studies involving more than 24 million participants were analyzed, including all studies that were not selected for publication. For more details, see Michael Hallsworth’s article on the future of nudging debate.

So there is scientific evidence that nudging doesn’t work and does work, and a discussion that could theoretically go on forever (also a problem that almost all scientific disciplines struggle with). Fortunately, rather than a yes/no debate, there are more productive ways to find out what nudging can and cannot do for us.

To create clarity about the effectiveness of nudging, it is important to understand how and where nudging is applied. While the first nudges were simple and one-sided (think of the fly in the urinal), there is now evidence that nudges can also have a structural impact. For example, a nudge where UK residents automatically save for retirement has resulted in a £7bn increase in retirement funds in one year. In addition, there is evidence of positive effects of nudges within finance, public health, and the private sector.

In short, nudging comes in all shapes and sizes. By lumping all the nudges together and adding a total effect to it, we run the risk of comparing apples to oranges. In addition, the effectiveness of a nudge on behavior often varies within population groups and context. Instead of arguing about an overall effect, it’s time we looked at which specific nudges work best for which problems.

The Future of Nudging

Since 2008, nudging as a behavior change technique has grown exponentially, but is still young and full of potential for improvement. For example, computer science can be used to map differences within population groups (hypernudging), and meta-analysis can be used to study the effects of nudges within a specific context. In addition, more and more organizations are experimenting with internal or external nudge units to demonstrate the benefit of nudging on compliance, business transformations, and culture change.

The publications of PNAS do not seem to be reasons to immediately throw the nudge overboard. It does, however, provide a strong motive to continue to look critically at the broader challenge of behavioral change. Nudging is just one tool in the brain and behavioral science toolbox in which testing, measuring, iterating and optimizing are central. The future of nudging and understanding our behavior will not be a discussion about whether nudging works or not, but a dialogue between different disciplines in which there is room for nuance and curiosity.

Enjoyed this article? Consider sharing it with a colleague. If you’re looking for a behavioral business partner to drive organizational change, we’re happy to schedule a call. Or read our blog to learn more.

About Neurofied

Neurofied is a Dutch behavioral science company that helps organizations change and professionals develop. We apply insights from psychology and neuroscience to improve areas of expertise such as change management, leadership development, and AI adoption.

Since 2018, we have helped 100+ teams and 1000s of professionals at organizations such as Johnson & Johnson, Deloitte, Novo Nordisk, and the Dutch government to drive measurable business outcomes with behavioral science. Read more in our client cases.

We frequently provide custom training for teams and speak at universities & conferences.