Interview with Jason Hreha

Behavioural Science is a rapidly expanding field and everyday new research is being developed in academia, tested and implemented by practitioners in financial organisations, development agencies, government ‘nudge’ units and more. This interview is part of a series interviewing prominent people in the field. And in today's interview the answers are provided by Jason Hreha.

Jason has been a practicing applied behavioral scientist in the technology world for nearly 15 years. A Stanford alumnus and former researcher in the Stanford Behavior Design Lab, he founded the first Fortune 100 behavioral science unit at Walmart. As its Global Head, he integrated behavioral science research and methodologies into corporate strategy and different initiatives across the company. A serial entrepreneur, Hreha has founded three successful companies, notably Persona, which uses behavioral insights to recruit remote talent for fast-growing companies.

---

How did you get into behavioral science?

I first got interested in it in high school when I was taking a literature class on “Ancient Eastern Thought.” This was around the same time that Richard Davidson of the University of Wisconsin was doing all his neuroimaging studies on monks and looking at the effect of meditation on brain activity. One day my literature teacher brought that work up by showing us the newest issue of National Geographic. The cover was a picture of a monk with electrodes covering his head. I was so fascinated by the topic that I borrowed his copy and read it. And then I got really interested in neuroscience and read everything I could on the subject that year. So I originally got interested in the brain through a literature class.

I ended up studying neuroscience at Stanford through a major called Human Biology. The cool thing about Human Biology is that you can choose your own concentration, and my concentration was neuroscience. Specifically, the neuroscience of behavior. I also worked in a neuro lab there–a neuroethology lab–where we looked at the neural correlates of behavior. My interest was always in taking the behavioral sciences, and neuroscience in particular, and applying them to practical issues—using science to make people's lives better. But my studies showed me that, outside of medicine, neuroscience wasn’t really all that useful for solving real-world issues. So toward the end of my time at Stanford, I got very interested in looking at who in the behavioral sciences was doing applied work that actually seemed meaningful. And that's when I encountered BJ Fogg. I learned about his work right as I was graduating. So after I graduated, I reached out to him to see if I could collaborate with him. He said yes, we started working together, and the rest is history.

What would you say that your proudest achievement is as a behavioral scientist so far?

I don't know if there's really one moment that stands out to me. I feel that my biggest contribution to the field over the years, though I don't know if it's truly taken off yet, is behavior matching.

Back around 2013, I realized that the most important determinant of behavior change is the behavior you're trying to get people to do. If you're trying to get people to exercise more, and you're just like, “hey, I want everybody to start running an hour a day,” well good luck—it's not going to happen. Not everyone wants to run. Not everyone will find the activity enjoyable or convenient. You have to pick the right behavior for them. I realized that behavior matching, or what I also call behavior selection, was the most important thing that a behavioral scientist could do to help a company or an organization achieve its goals.

I realized this because I was working with a startup that was building a social network that was focused on helping people accomplish their goals. They brought me in to help them build the product and maximize its retention and engagement. The basic idea behind the product was: create a social network that’s built around shared personal growth. People can join, choose habits they’re interested in forming, and collaborate with others on these shared aspirations. And what I noticed was that people would sign up for the app, see things like meditation and yoga, and get very excited. They would say, “Hey, I want to do that three days a week. So let me start meditating with everyone in the meditation group.” But very few people who joined the product ended up following through with these behaviors, even though they're things that they thought they should be doing. Almost nobody was doing them. To discover why, I did a series of a user interviews and it became very obvious to me that most users felt socially pressured to do these behaviors, but nobody actually cared (intrinsically) about those things. Even though they thought they should, as successful and driven people, do them, they had other more pressing concerns in their lives.

So at that point I got very interested in behavior matching. You can get the same benefits of meditation by doing a bunch of different things i.e. breath work, long walks etc.  So why don't we, instead of being prescriptive, help match people to the right behaviors for them; personalize the solution to them. In the following years, I kept beating that drum around personalization, behavior matching, and how behavior selection is the main determinant of behavior change. So I feel proud about that.

I also feel quite proud about the awareness I brought to the shakiness of a lot of behavioral science research. I think that a lot of practitioners have done a great disservice to people interested in the behavioral sciences by overselling the effect sizes and the potential impact of these interventions. And I do think that they're killing the credibility of the field in the eyes of governments and the business world. I know for a fact that in the business world, people are starting to roll their eyes and don't want to hear about behavioral science, because people come in promising these huge effects, and huge business changing possibilities when the stuff is anemic, at best, a large portion of the time. So I do feel quite good about bringing attention to this and increasing the skepticism toward behavioral science.

---

So what do you still want to achieve? Do you want to make sure that behavioral matching is properly taken up?

I think that would be nice, because it's a win-win. I think it’s important to shift the field to the perspective of “how do I pick something that the customer will enjoy that will also get them a result they want that will also accomplish my organization's goals?” I think shifting the field into that mindset will make everyone better off. This is different than what I think is the dominant mindset today, which is focused around getting people to do something that you and your organization want using various tactics and “hacks”. This approach is not only ineffective, but it’s less ethical. A lot of the practitioners out there seem to approach humans as if they are computers, and the work is about finding this “code” they can use to get customers to act differently, even against their own self-interest. This perspective is not accurate or effective—and it results in bad products and programs.

A user-centric mindset shift is the big thing that I would like to see happen within the industry. And I don’t think we’re quite there yet.

Where do you think this field is gonna go next, say in ten years’ time?

We’re still in the early days, and so I think it’s quite hard to see what the field will turn into. I do think that applied behavioral science, in its current form, will die. I’m not optimistic about the future of people who exclusively specialize in behavioral science. Currently, behavioral science, as an applied discipline, is garbage. A lot of its results are fake or deceptive. One quick example of deception: defaults. A lot of applied behavioral people talk about defaults as one of the most powerful behavior-change tactics. In doing this, they inevitably use organ donation as an example. There’s only one problem with this: organ donation defaulting is not an example of behavior change. You’re not getting anybody to actually change their behavior. You’re tricking people into giving you legal permission to do things to them after they are deceased. You’re not causing someone to actually behave differently and do things differently while they’re alive. Behavior change means getting people to behave differently. It’s getting someone who was not running to start expending energy and going on regular runs. It’s getting someone to drive to the store and make a new purchase decision. Given this, organ donation defaulting is, in no sense, behavior change. It’s just linguistic trickery on the part of behavioral economists. They’re labeling something as behavior change when they’re really just talking about a legal and paperwork change. And, if that wasn’t bad enough, these interventions aren't even effective. If you look at the organ donation literature, you’ll see that there’s no statistically significant difference in organ donation rates between countries with opt-out vs. opt-in systems. Even Spain, the world leader in organ donation, has a mixed system (combining elements of opt-out and opt-in). They got there by creating all these different programs to help families think through the decision-making process and support them in making what is, for them, the right decision. In other words, they’ve done a beautiful job creating a whole infrastructure that does a lot of education and hand holding to help people understand the implications of their actions.  That’s what led to success—not some tiny tweak. This is a great example of looking at a messy reality, and it being very different from the picture the behavioral scientists try to paint. This is just one example of why I'm not optimistic about the future of the applied behavioral sciences.

---

Do you have more frustrations with the field?

Definitely. A significant portion of the research is just bad. We had the Reproducibility Project which tried to replicate 100 different behavioral science studies and, if I remember correctly, they only had a 36% replication rate.

Since then, we've had other attempts to determine the replication rate of the behavioral sciences. A study published in 2022 found a 50% replication rate. Another study from 2018 found a 62% replication rate, and also found that the effect sizes are half of what was originally reported. Things don’t look great.

I've worked my whole career in business, in the applied world. Most of my career has specifically been working with startups. These are companies that are fighting for survival. They’re trying to achieve what we call Product Market Fit. This is when a product matches what customers want and need so well that they regularly use it. It’s all about consistent use. Retention and engagement. Generally, for companies looking to find product market fit, a 1-5% increase in user activity is just not enough to move the needle in any way that actually matters. Often, when I'm working with a company, they need to double or triple their retention or activity. If they call in a behavioral scientist, then it's because they're struggling. They need to make big changes to get the big results they need. They don’t need small tweaks like loss aversion framing and social proof messaging or some other gimmick. These things barely move the needle. I've always been frustrated by the weakness of these approaches in the real world.

Another problem on top of all this is that a lot of people in the field that talk about behavioral science as if it's magic. You can just sprinkle it on any product or problem and it will have these profound impacts. That's just not true.

What it can do is provide a helpful lens that can allow you to better interpret user research or company data. It can help you ask good questions and come up with targeted hypotheses as to what may be going on. Instead of looking into 50 possible effects, a skeptical and experienced behavioral scientist can cut it down to the 3 or 4 most likely ones.  So it’s not magic. It’s just a helpful filter for viewing the world.

Those are my three big problems in the field:

1.      unreliable research

2.      pitiful effect sizes

3.      individuals that sell behavioral science like it’s magic

Those three things are destroying the field’s credibility.

So what challenges do you think the field is going to face after these developments? And do you have any solutions that you'd like to suggest?

I think the field is already facing credibility issues. It hasn't gotten to a crisis point yet, but I think it will—especially in the business world. A lot of technology startups have been interested in the applied behavioral sciences. Startups are generally cash constrained. They don’t have that much money to waste. They need to get truly meaningful results in order to justify the expense, and behavioral science isn’t really providing them.

I've been brought in a couple of times to companies where they hired a behavioral scientist to help them redesign a product. They spent a lot of money getting some behavioral expert’s advice, and in each of these cases they just wasted their money. They had to throw out the entire work product from the behavioral scientist they hired. And for a smaller company, say a startup, that matters.

Situations like this are damaging to the field. These companies are now never going to hire an applied behavioral scientist again. Especially since they could have hired a UX designer to do the same thing much more effectively—using common sense and their design skills.If this keeps happening, and we fast forward a few years, applied behavioral scientists will become obsolete due to their bad reputation.

---

I almost don’t want to ask, but what advice would you give to younger people looking to enter the field?

Or general advice to younger behavioural scientists.Well, I don't think people should go into the field—at all. I don't think people should want to be behavioral scientists. People should become developers, product managers, UX designers, UI designers, data scientists, etc. I think everybody should pick a core position that they want their career to be based. That should be their focus. And then they should treat behavioral science know-how as a multiplier, as a bonus. I don't think it's a very useful skill set on its own.

Ok let's talk about specific skills then. What skills should already behavioural scientists get into then? Or is that dependent on the position that you choose, for example a product manager?

Once again, this goes back to personalization. It's hard to get blanket advice for this. It depends on your interests and what you want your core position to be. If you love doing statistics and you're interested in learning Python etc. be a data scientist. Figure out what skills you need to succeed in that position. And then, second, become a behavioral scientist. I think each person should figure out what are they interested in doing and what really calls out to them. But I would recommend each person picks a core role and excels at that first. Then you can “super charge” it with behavioral science know-how. So the skills depend on the core role.

---

Ok that is a pretty clear recommendation. With all of this, do you apply any behavioural insights to your own life?

Definitely. I do behavior matching all the time to help myself pick the right behaviors for my goals. Whenever I want to achieve something, I go through a behavioral strategy process to create what I call a “Goal Menu”.

I enjoy treating each part of my life like it’s a menu. When it comes to fitness, for example, most people don't like going to the gym or jogging. But every single person has their own strengths and skills. For each person, certain activities are going to be easier and more enjoyable than others—based on their physiology, talents, living situation, etc. And so, in different areas of my life I have menus I create. For example, my fitness menu has all the different health and fitness related things that I like to do that also move me closer to my goals. I have a whole process that I go through that helps me pick the behaviors that are optimized to my goals, living situation, personality, strengths, interests, etc. And so, whenever I want to progress in my fitness, I look at my menu and see which one of the activities I’ve curated is calling out to me today.

And I have this for every area of my life. If I have a goal I care about, I have a Goal Menu for it. Over the years I have behaviorally matched my whole life, and I take this quite seriously.

So your life is definitely behaviourally optimized. Where do you think you would have ended up if you hadn’t found behavioural science?

I think that even if behavioral science didn't exist, I would still be thinking about behavior and people, and trying to understand what makes us tick. I probably would have spent less time reading books and articles and spent more time looking out into the world and just observing what’s actually going on. But I don't think my life would be all that much different if I never discovered behavioral science. I just think I would have gotten here in a different way,

Last question then, applying a nice snowballing technique. Is there a behavioural scientist, or really anyone, that you think would make for a good interview, given the topics we have discussed? 

You have already interviewed so many great people. But I think you’re still missing BJ Fogg.

---

Thank you so much for taking the time to answer my questions Jason!

As I said before, this interview is part of a larger series which can also be found here on the blog. Make sure you don't miss any of those, nor any of the upcoming interviews!

Keep your eye on Money on the Mind!