So you want to start a behavioral science team…

Jason Hreha

All companies are in the business of changing behavior.

Target and Amazon are in the business of getting people to visit their stores and websites and purchase as many items as possible.

Google is in the business of getting as many people to use as many of its products as frequently as possible — so that they can continue to be the world’s largest data aggregator.

Toyota, Tesla, Audi, and Ford are in the business of getting people to say “yes” once every few years so that they sign leases and purchase agreements.

Traditionally, companies have hired marketers, merchandisers, and strategists to achieve these behavioral ends. But none of these disciplines are taught the most up-to-date psychological research, and thus bring inaccurate models of human cognition and action into their work.

This is one of the many reasons that so many of their products and programs fail.

How many times have you purchased a car or downloaded an app only to think “this must have been designed by an engineer”?

How many times have you set up a new camera, laptop, or piece of furniture, only to think “which MBA decided to create such a cruddy experience”?

Those who don’t have a firm grasp of human psychology are bound to do more harm than good for their businesses. Sure, they might create products or programs that show a short-term gain, but they’re bound to do an even greater amount of harm to the long-term prospects of the companies they represent.

A great example of this is Zynga, which used destructive behavior-change tactics to juice its engagement and revenue numbers back in 2011 and 2012. These tactics allowed them to become the darling of Silicon Valley for a brief moment in time, until users revolted because of their deceptive and ineffective practices.

Another company that has used a scientifically inaccurate model of behavior to guide its product decisions is Fitbit. The stock price graph, shown below, speaks to a basic lack of behavioral-science know-how.

As most users know, the two-week retention of Fitbit’s products is atrociously bad.

Each of these companies could have been helped by a proper behavioral science team. In fact, if behavioral science research had informed their marketing and product strategies they might have been able to sustain their growth and protect their good-will.

Unfortunately, very few companies have behavioral science teams, and most of these teams aren’t architected in an effective manner.

To create a well-functioning behavioral science team, the following things need to come together. I call them the “4 Rs”:

  • The right talent
  • The right reporting
  • The right structure
  • The right work

The Right Talent

99% of the people have studied behavioral science at a high level are NOT a good fit for an applied behavioral science role. The skills learned in the academic world are usually not translatable to the practical world.

For a behavioral scientist to operate effectively in a business environment, the following things are necessary:

  • Comprehensive knowledge of a variety of different behavioral science research domains (PhDs usually don’t have a good general knowledge of behavioral science. They’re an inch wide and a mile deep. This is why those with Master’s degrees are often preferable.)
  • Creativity (Most of those who have spent a substantial amount of time in an academic environment apply academic findings in a cookie-cutter manner — copying the interventions they’ve read about in the literature with minimal customization. This leads to small effect sizes — and failure — in the real world.)
  • Political skills (Getting work done in the business world requires tact and political skills. Figuring out how to get work done in a practical context requires behavioral scientists to make concessions and figure out how to change their programs in a way that gets the biggest behavior-change impact while also fitting into the constraints of the organization. This is not something academics are good at, since their experiments are usually designed in simplified, artifical environments with no real-world relevance.)
  • Practical skills (The ability to think and read research papers isn’t very useful in the real world. It’s a nice *addition* to one’s skill set, but it’s the sauce on top of steak — so to speak. Practicing behavioral scientists need to be able to design, code, or do real-world data analysis. They need to be doers, not just thinkers.)

The % of people with a behavioral science background who have all of these attributes is exceedingly small. Those who are missing any of these attributes are not worth hiring and are often dead-weight to the organizations they join.

The right reporting

Good behavioral science work is more than “nudging”. If you want to hire a behavioral scientist to just do nudge work you’re wasting your time and money. Nudging is wasteful in most practical business contexts. Behavioral scientists have the most to offer in the product or program creation process. They can infuse accurate behavioral thinking to the foundation of new programs that the company is creating, which is often the difference between their success and failure.

To gain the organizational leverage needed to shape the programs they work on, behavioral scientists must report into senior executives (or have a deep, trusting relationship with senior executives).

The right structure

There are only two different structures that I’ve seen work for behavioral science teams:

  1. Independent organization reporting into a senior executive
  2. A distributed, functional organization embedded within product teams

The first structure means that the behavioral science team is effectively an internal consulting firm.

The second structure means that behavioral science is a core competency mixed into each product team.

They each have their pros-and-cons, but these are the only two structures that work.

The right work

As mentioned earlier, behavioral science teams need to be involved in projects from the very beginning. The proper role of a behavioral science team is to influence the fundamental structure of new products and programs — so that they conform to the laws of human behavior and cognition.

If a behavioral science team is brought into pre-existing products/projects, they’re set up for failure. Here’s why:

  1. Nudges aren’t as powerful as reported in the scientific literature.

Usually, the scientific literature overstates how powerful nudge-style interventions are. There are a variety of reasons this is the case (which I won’t go into here), but, from my experience, this is unambiguously true.

2. Even if nudges are powerful, they’re almost never powerful enough.

Even if a nudge is able to increase the KPI in question by 100%+, this is usually not enough to save the product or initiative in question. Real-world business projects need to be wildly successful and have a substantial ROI in order to receive continued funding and support. Even 100% nudges are not going to be enough to save most sub-par projects. The only thing that could have saved these projects was proper behavior-science knowledge during the creation phase.

3. Nudges increase operational overhead and are almost never sustainable.

While most nudges seem small, they also usually entail administrative/operational overhead that usually isn’t sustainable. Good luck getting the operational support you need to support even the smallest nudges (especially in businesses that have some physical component).

Figuring out how to set up a team in the right way with the right people is something that only a couple of people in the world have done. I can safely say that my experience building the first applied behavioral science firm in Silicon Valley and setting up the first behavioral science division at a Fortune 50 company puts me in a unique position.

If you want guidance on how to build your own behavioral science team, please go here and send me and my team a note:

📩 📩

We’d love to hear more about your organization and see if we can help.

Jason Hreha

Written by

Head of Product, Behavioral Sciences @ Walmart. World leader in applied behavioral science. Entrepreneur (1 exit). My site:

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade