How to apply behavioral insights at an organization-level: Tips from Dr. Elizabeth Linos

In this article you’ll learn:

• Common themes among government agencies looking to improve policies or legislation
• Advice for how to apply behavioral economics principles at an organizational level
• Insights into supporting agencies to adopt best practices
• Tips for disseminating research so that it has real-world impact

Dr. Elizabeth Linos is the Emma Bloomberg Associate Professor for Public Policy and Management, and Faculty Director of The People Lab at the Harvard Kennedy School of Government. She spoke with CHIBE for a Q&A about her lab, which works to strengthen the government workforce, improve resident-government interactions, and take best practices to scale.

In 2017, you founded The People Lab, which collaborates with government agencies and other organizations to respond to challenges. What sort of patterns have you seen over the years – are their common themes with the types of problems or solutions these organizations face or need?

We are so privileged to work with exceptional leaders at all levels of government as well as with individuals in the private and nonprofit sectors who align with our mission. One of the most striking patterns we’ve observed is that nearly every agency thinks they are uniquely struggling—whether due to their size, structure, or the specific communities they serve. But in reality, many of the challenges they face are remarkably similar. No agency has figured it all out (nor has the research community).

A lot of the questions we hear are how questions: How do we reach residents and foster positive interactions? How do we better support our frontline staff? How do we make programs easier to access without adding new burdens?

These questions highlight an important reality: improving government isn’t only just about designing better policies or legislation—it’s about ensuring they can be effectively implemented. As researchers, our job therefore isn’t just to offer high-level policy ideas but to help operationalize those ideas in ways that work within real-world constraints.

One thing that has been particularly surprising is how much the appetite for data, evidence, and rigorous evaluation has grown in government. Agencies don’t just want a stamp of approval on their projects, they want to find ways to improve their services and are eager for support in building the capacity to use evidence in everyday decision-making. This was not the norm 15 years ago, but today, more agencies are investing in evaluation, testing new approaches, and embedding behavioral insights into their work. 

Do you have any general advice for how to apply behavioral insights at an organization- or system-level?

The best advice I have for applying behavioral insights at an organization or system level is to start with the problem, the experience, and the people navigating it every day. The behavioral nudge should come last.

At The People Lab, we spend the majority of our time trying to deeply understand and scope the problem—before ever proposing an intervention. This means talking to frontline staff, observing how people interact with systems (or even trying to follow the user journey ourselves), and mapping out existing constraints. Too often, there’s a rush to jump straight to solutions—whether it’s rewording a letter, tweaking a deadline, or adding new choice architecture—without first understanding what is driving behavior in the first place.

At the same time, I encourage anyone working in this space to be patient with the number of constraints that organizational leaders face and to become more comfortable with incremental but consistent tweaks. If an obvious, high-impact solution hasn’t been implemented yet, there’s usually a reason—whether it’s resource limitations, legal barriers, or competing priorities. The most effective change may happen through sustained, iterative improvements rather than one sweeping fix.

You co-wrote a paper showing that government organizations who use RCTs to test innovations are most likely to incorporate the results into policymaking if there’s already some pre-existing communication built into place. Do you have any ideas for how to spur more inertia if there isn’t some pre-existing infrastructure?

The paper you’re referring to follows agencies that had already rigorously tested interventions in their own contexts. We found that those interventions were much more likely to be adopted—up to 5 years later—if they were part of an existing communication infrastructure. In other words, when agencies already have systems in place, integrating research findings becomes a relatively incremental change rather than a disruptive one. But what about when they don’t? That’s the question we’re actively researching.

One of the biggest challenges is that using evidence in government is rarely just a technical problem—it’s also an informational, translational, and organizational one. Leaders need to know what evidence exists and understand how to interpret its value. They need to translate findings into their own contexts. And they must overcome a series of organizational hurdles—securing buy-in, managing logistics, obtaining funding, and navigating political will—to implement meaningful change.

The good news is that an entire ecosystem has emerged to support agencies in adopting best practices. Players in this ecosystem are embedding research teams within agencies, building peer networks to facilitate knowledge sharing, and providing technical assistance to help leaders interpret and act on evidence. Our next step is to rigorously assess which of these approaches is most effective, for whom, and under what conditions. Stay tuned—we’ll have more insights on this soon!

Another paper you wrote showed that government communications are more effective at influencing individuals’ behavior if they are formal rather than informal. Can you tell us how you disseminated your results so that organizations can implement these findings?

Our research on government communication and formality has gained significant traction, with multiple agencies and nonprofit teams reaching out to learn more about how they can apply these findings. We were surprised by these findings too! When governments communicate in a formal, more official tone, people are more likely to take the message seriously—because formality acts as a heuristic for credibility and importance.

But generating research findings is only half the battle. The real challenge is ensuring those findings actually influence practice. We’ve taken multiple approaches to disseminating this work and other research projects:

• Direct engagement with policymakers – We present our findings at government convenings, workshops, and meetings with agency leaders.
• Policy briefs and practitioner-friendly materials – We translate our research into accessible formats so that public servants can quickly grasp key takeaways.
• Collaborations with government teams – Beyond sharing insights, we work alongside agencies to test and refine these ideas in their real-world contexts.

Despite these efforts, we know that dissemination remains an ongoing challenge. Research alone doesn’t change practice—embedding evidence into policy decisions requires sustained engagement and collaboration. One of the big questions we’re working on now is how to design more effective pathways for getting evidence into the hands of the people who can use it.