Exploration & Experimentation, the Startup Lab approach!

This is the fourth installment in a five-part series on the CAH’s Startup Lab, our academic incubator program for health and finance tech startups. We’re demystifying why and how a behavioral economics research center at Duke supports and collaborates with entrepreneurs.

We’re looking for startups that are eager to experiment, and demonstrate a passion for building research-backed solutions to health and finance challenges. Startup Lab applications are now open until June 30th at 5pm EST. Apply here.

Exploration Informs Entrepreneurship

The Startup Lab is about accessibility — making behavioral economics findings readily available and quickly actionable for entrepreneurs. The incubator was born out of a need to better connect product designers with the wealth of academic research in social science. To achieve this, we’ve developed an iterative process that leads our startups through behavior mapping, exploring literature, conducting experiments, and implementing findings. The Startup Lab exposes startups to existing research, encourages them to explore, then teaches them how to apply the findings to create products in line with the natural operation of humans.

The Startup Lab Process

“Too often, designers forget to find the library.” –Dan Ariely

Exploring the “library” of existing research is the first thing we teach entrepreneurs how to do. We want them to filter their ideas through the lens of behavioral research so they can build solutions from an understanding of how people make decisions.

But we don’t leave them to decipher the academic literature alone. We provide a framework for how to interpret and leverage research in the context of their business. Our approach to applying BE to tech solutions is an iterative cycle that uses existing literature and our behavior mapping process to inform experimentation, which in turn informs which changes to implement into the product. Let’s look at the components of this cycle:

  1. Behavior Mapping: Identify Problems and Brainstorm Solutions

Behavior mapping is a way for entrepreneurs to get familiar with the motivations and environments of their users. We use this as an identification tool. We lay out the customer’s journey (including each step of using a tech product, though the product is just one part of the process we are trying to understand) so we can articulate our goals for the user and focus in on problem areas.

What is the key behavior we’re attempting to influence? What are the biggest barriers to this behavior? Where in the customer’s journey do we need to intervene to help them complete that key behavior?

  1. Look to the Literature

There is a large body of academic research ready to inform the behavioral problems and solutions that vex many startups. Yet products are often designed without the help of social science or scientific rigor, and later fail to attract adoption or deliver on their promises when used by people.

Behavior mapping uncovers barriers to positive behavior so that we can identify potential behavioral interventions. But we always circle back to the literature to answer these questions:

Have other researchers tested similar interventions before? In what context? Can we apply aspects of what they did to experiment here?

In essence, we’re learning from those who have gone before us, having tried similar solutions in similar contexts. We then want to build on those solutions through new experimentation.

  1. Experiment

We use original research to fill in the gaps that the literature doesn’t yet cover.

Beyond learning from existing research, we aim to bridge the gap between where academic papers leave off and real world application begins. Our ultimate approach to applying research to startup products is collaborative, original research.

  1. Implement Findings

Once we’ve collected and analyzed the data from an experiment, we can determine if the intervention we tested was successful in effectively influencing the positive behavior we set out to influence. Before collecting data, we determined how effective an intervention must be in order to move forward in implementing it into the product. If it meets pre-set standards, we deem it worthwhile and integrate the change to create a more effective product.

Entrepreneurs and Experimentation

The Startup Lab creates the environment — the time, space, and resources — for entrepreneurs to think, explore, and learn about behavioral economics. Startups are pointed towards relevant reading, provided with an infrastructure to soak up resources and learning, and surrounded with researchers ready to talk about interesting problems and hypothesize solutions.

Collaboration happens when people share space. Innovation happens when collaboration spurs original research.

We equip entrepreneurs with the right tools and people to execute original research. Experimentation provides evidence-based insight into what really works. Experiments don’t have to take an intimidating amount of time, effort, and resources. It’s better to create a process of iteration, where we fail with the intent to learn. The goal is to pinpoint what works and constantly create opportunities to fail forward.

P.S. We like to make rigorous experimentation accessible to a broad range of companies. Our Experimenting in Business series provides tools and guidance for using research to make smarter business decisions. Our first post is about beginning your experimentation process with the end in mind: Read here.


The Startup Lab supports problem-solvers by making behavioral economics findings accessible and applicable.

Are you insatiably curious about what drives decisions, shapes motivation, and influences behavior? Does your startup’s success hinge on the ability to affect positive behavior change that helps people live happier, healthier, and wealthier lives?

Apply for the October 2017-June 2018 Startup Lab program by June 30th at 5pm EST. Apply Now.

To learn more, visit our FAQs (includes information on the Startup Lab investment structure and other logistics) or connect with us at startup@danariely.com