Jason Hreha is a behavioral scientist, formerly Global Head of Behavioral Sciences at Walmart, who uses his knowledge of human behavior to build better products. With a decade of applying behavioral science to tech challenges, he co-founded Walmart's Behavioral Science Team and pioneered Behavioral Strategy as an interdisciplinary approach. He studied human biology and neuroscience at Stanford University, and is currently co-founder and CEO of Persona, the Startup Assistant company.
Jason Hreha is a behavioral scientist, formerly Global Head of Behavioral Sciences at Walmart, who uses his knowledge of human behavior to build better products. With a decade of applying behavioral science to tech challenges, he co-founded Walmart's Behavioral Science Team and pioneered Behavioral Strategy as an interdisciplinary approach. He studied human biology and neuroscience at Stanford University, and is currently co-founder and CEO of Persona, the Startup Assistant company.
AMY: [00:00:20.9] Jason Hreha is a behavioral scientist and entrepreneur who knows how to leverage psychology to build better products.
For over a decade, Jason has been at the forefront of habit design, working with industry leaders like Google and Walmart and launching startups, including his current company, Persona.
Jason knows that building habits and driving retention always starts with a deep understanding of your users.
Jason: We're in the product design game to build solutions that make people's lives better. We're here to really understand our users, understand the people that we're building for, and we really wanna create something that is, is meaningful for them. And I can't think of a better [00:01:00] way of really measuring that than are people, sticking with us? Are they here, six months later, a year later, et cetera. I think habit formation is just really a wonderful way of just like boiling down what matters.
AMY: Join us as we talk about habits, rewards, and why convenient theories like the power of nudges don't usually work in the real world.
AMY: Welcome Jason. Before we dive into habit design, tell us about your background.
Jason: So, at Stanford I stayed neuroscience. I did this major called human biology, but I was interested in behavioral neuroscience.
So that was my focus. I really had a very in-depth behavioral neuroscience background, just really looking at like the brain, how does the learning system, the reward system modify, how animals learn new things, adapt, behave, how does that relate to human behavior? And that was basically for, that academic period of my life, my real focus.
And then I met BJ. When I was at Stanford, BJ Fog. And I started working with him actually after I graduated. So, I was in the Stanford Persuasive Technology [00:02:00] Lab. After I graduated, and that's just when we were really building this field of behavior design.
That's actually when I was in the lab is when the term was coined. There was actually like a kind of, like a vote on it and everything like that. So, it was very, it was very cool just to be around at that time during this dynamic period when all these different companies like Facebook, Instagram, and stuff like that were popping up and engagement. Retention behavior was really just becoming a focus of product design.
Right? And that was like really just, it was just very cool just to be in the valley at the time. And just to see people just really starting to think in a very sophisticated, nuanced manner about these kind of psychological and behavioral issues as it came to product design. So, I was there and then just over the years.
I, I was hooked at that point after working in the lab with BJ, after seeing that kind of human psychology and human behavior insights about those things could be applied to product design, just, I was just fascinated after that. So, I've spent my entire career just applying research from different areas of the behavioral sciences.
AMY: So tell us what you've [00:03:00] learned about using behavioral science to build better products.
Jason: So over the years, I've been really fascinated by like evolutionary psychology, social psychology, behavioral economics, you name it. Just if there's an interesting or a seemingly useful idea from any behavioral science subfield, I'll take it and I'll try it. And I've worked at Dan Ariel over the years helping him apply behavioral economics research to product design.
I've collaborated with a lot of other people as well. Walmart recruited me to build out their applied behavioral science group. So, we were like the first applied behavioral science group at a Fortune 50 company. Walmart was the fortune one when I was working there, so it was the biggest company by employee based in the world, and they get most revenue of any company in the world. And so it was really cool to, at that scale run experiments, run tests and just apply these ideas. And so I got to actually apply a lot of these ideas in store. So in physical environment, not just a digital environment.
Not really, did actually change quite a few of my ideas about behavior design or like a applied behavioral science, if you'd say.
Everything I talk about, I've [00:04:00] tested in the real world. This isn't just some like academic thing where it's just me pondering, right? The reason I'm against, so against nudges, as somebody mentioned earlier, is I've tested nudges with huge sample sizes and the nudge literature will say, hey, you're gonna get an eight or 10 or 20 or whatever percent increase in the outcome variable that you care about if you know you implemented this nudge. And I've implemented so many different nudge experiments and I just would always see that you'd get like no effect or maybe a tiny effect, positive effect or a tiny negative effect.
I've just seen time and time again, just this nudge stuff in the real world in a complex environment just doesn't work. And so that's been my battle a lot of the times with people that are more academically minded is just they're running these experiments. At least very tightly controlled conditions, or they're torturing their data analyses to try and get an effect.
And then I go out into the real world with companies and try this stuff, and I just see there's just no effect.
AMY: So why should product designers care about habit design?
Jason: Habit formation is the ultimate proxy that you're really solving a real problem, that [00:05:00] you've really built something meaningful and something that people really find value in, and that people really enjoy practically, right?
Habits are just very useful for a business. If you're a business and you have habitual users that are coming back, day after day, week after week, you're gonna get more money, you're gonna get more brand loyalty. You're building that relationship with your users.
It's very obvious. Of course, you're gonna get greater retention, right? There's more of a switching cost if somebody really has this tight bond between a problem in their lives and in your product, and they formed a habit around that, right? They're just, the switching costs are increased.
It's just harder for people to break a habit than it is for them just to break any other sort of behavior. And, as I mentioned just a little bit earlier, it's a sign you built something people actually want. And at the end of the day, that's like why we're in the game,
That's why we're in the. Product design game is to really build solutions that make people's lives better. We're here to really understand our users, understand the people that we're building for, and we really wanna create something that is, is meaningful for them. And I can't think of a better way of analyzing that in most cases.
Of course there are exceptions, [00:06:00] but in most cases, I can't think of a better way of analyzing or of really measuring that than are people sticking with us? Are they coming back constantly? Are they here, six months later, a year later, et cetera. I think habit formation is just really a wonderful way of just like boiling down what matters.
AMY: So what's your definition of a habit?
Jason: In order to discuss habits, we have to really understand like what a habit really is, right? It seems like every habit expert has their own definition. I have my own definition as well. But I think that, the definition that I have really gets at what a habit really is.
So, James Clear actually quotes me in his book, Atomic Habits. He quotes my definition of habit which it's a reliable solution to a recurring problem. And so that's I think really for me, the most crystal clear, simple way of describing what a habit is, and it has huge implications, right? For habit formation design, and for product design.
And so for me, problems are like the foundation, they're the core of all habits. The reason that, [00:07:00] we as human organisms do things, the reason that we evolved to, have limbs and evolve a brain that's plastic and can actually adapt and do different things is in order to solve problems in our environment, to solve survival problems, to solve other sorts of problems to adapt, to grow.
And I really think that approaching habit formation is fundamentally like habits are solutions to recurring issues that we encounter frequently. And so, if you just take that insight and then you unwrap it, there are lots of very interesting insights that you come to.
What separates a habit from a routine? I think a lot of this, especially in the self-improvement, self-development area, which is a little bit, it's separate from. Habit product design in, in the self-improvement and self-development world in particular, I think that this distinction isn't really made right.
Habit by its very definition is supposed to be automatic. It's supposed to occur without a whole lot of thought. It's supposed to almost be like a, almost like a reflex where, you get a stimuli in your environment that prompts you like, Oh, cool. This issue just came to mind, or this issue just prompted [00:08:00] my neurology.
And then suddenly before you know it, you're solving it. You're doing some behavior that takes care of the problem. So, this is all to say that a core feature of habits, just if you just look at the behavioral science literature is that their automatic don't require a whole lot of conscious awareness, if any conscious awareness.
But a lot of the things that people are trying to do from a habit perspective actually require a lot of thought or a lot of a lot of willpower. So for example if you're trying to start a new exercise routine, like to tap into the health behavior change angle for just one moment.
Most health behavior change problems. The reason that they're, it's so hard to form habits around these things is that they're really hard, right? If you're trying to get somebody from a health behavior change point of view to exercise, and they even exercise in years it's gonna take a lot of willpower, it's gonna take a lot of effort.
It's just gonna be really hard to get them to do this, right? Exercise by it's very nature is hard, right? That's the whole point, right? Is to do something effortful challenging for your biology that gets your biology to then adapt right, and get stronger. From that perspective, any let's say [00:09:00] exercise related behavior is gonna be effortful, it's gonna be challenging, it's gonna require conscious thought.
So almost every single health behavior that people try and get people to do is not really a habit. It's a routine, right? Maybe one day it could become a habit when you get stronger, when it's much easier for you to do kind of those exercise or health related behaviors. But in the beginning, at least those things are always gonna fall into the routine category.
And it's just very important to understand the implications of that if you're gonna effectively change behavior in that manner.
AMY: An important part of habit design is rewarding the user for performing the right actions. Tell us about the right way to do rewards, and if you could, share some common misconceptions that tend to steer designers in the wrong direction.
Jason: Rewards are very important, right? It's what cements the habit in the mind of the performer. But my, my conception of rewards are a little bit different than that. I think of most of the people that talk about this stuff.
For me, rewards are [00:10:00] really inherent to the behavior itself. So like I know somebody earlier mentioned intrinsic rewards I think that for habit formation in particular, I really think that the reward has to more or less be intrinsic.
It's a big topic, but I do think that rewards is classically conceptualized by a lot of people in the habit space or in the product design spaces like, points, money coupons, incentives, the likes social rewards, et cetera.
I think that those in general are like, at least from my experience, are pretty ineffective at driving habit formation for a majority of the different habits that you'd be interested in kinda getting people to, to form. My perspective is really that you have to pick for the target user group that you're designing for, you really have to pick the right behavior that is intrinsically rewarding for that target user group and for the target problem that you're trying to solve. I know that I just said that I just said a mouthful. There's a lot to unpack there, but in my perspective really [00:11:00] is that if you have to rely on points or any of these what would be called extrinsic rewards for your habit formation design process you've failed already. Like it's not gonna work out.
So, I really see that rewards have to be intrinsically a part of the behavior or the activity that you've selected. If you already have a product and you've already selected a behavior, there are often ways of just tweaking it slightly to make it intrinsically rewarding.
So, there's been a lot of talk over the years. Variable rewards, which is like a type of reward schedule being very important for habit formation.
I think it's been largely overblown and I actually think that if you are consciously trying to inject variable rewards into your product, I think it's in almost every case gonna make the product worse. Certain types of rewards that are very compelling to us.
Humans are variable by their very nature. But I think focusing on the variability is just a mistake. For example, social activity, if you interact with another [00:12:00] human it's gonna be a variably rewarding experience. People are unpredictable if you're interacting with another person, having a conversation with another person.
Let's say you start talking with a stranger, a lot of that conversation might not necessarily be all that interesting, but then maybe randomly they'll give you a compliment halfway through and it's like Whoa. That was cool. So just interacting with unpredictable humans just is like this amazing variable reward system.
And if you look at like, all the products that are focused on variable rewards, they're all social products basically. Maybe there's some games, but a lot of games are social too, right? Like Fortnite is like a, it's a social experience just as much as it's a game, right?
And so I really think that focusing on variable rewards has been a case of people missing the forest for the trees. It's like they're so focused on the reward schedule, and the timing of the rewards that they don't see. The thing that's actually like really compelling about a lot of these, quote unquote, variable reward products is that, they're social and we've evolved to be obsessed with other people and to, to enjoy interacting with other people.
And people are unpredictable and they [00:13:00] give us rewards at unpredictable moments, on unpredictable schedules.
The most effective reward when you're trying to build habits is often just some simple feedback, right?
one of the most compelling products ever is Uber. And one of the great things about Uber is, I remember when I was living in San Francisco, and this is pre-Uber, to get to my house in Beal Heights.
I would call a cab company and say, hey I need to get a ride. And half an hour, can you send somebody? And. You never knew if the taxi driver is coming, right? Sometimes like 45 minutes would pass. And I'd have to call again, and I'd say, hey, is, I need to get a ride.
I need to get to this party that I'm trying to go to. That's like my reward is getting to the party and having fun. And I called the cab company, I'm waiting for the cab. No one's ever coming. And so, the great thing that Uber did was give you feedback. hey, you need a ride.
Cool. Like you just press this button and I'm showing you that actually while your problem hasn't been solved yet, while you haven't gotten to the party, which is like the reward that makes like the whole experience so great. But we haven't gotten there. I'm [00:14:00] taking care of it. Here's the person, here's their name, here's them driving down this street, here's them driving down that street.
And so that's a great example of like feedback, bridging the gap between your current state and where you wanna be, which is the rewarding state. And a lot of the most habit-forming products for the products that just give us, just create a wonderful experience is that they provide us feedback along the way.
And so a lot of the times actually the best reward is just some simple feedback. I actually think that all the focus on variable rewards and just these slot machine dynamics just really misses the mark in that actually, um variable rewards are the norm in nature.
When we were like hunter gatherers and we were like going out and hunting, eight times outta 10, let's say, you wouldn't, score a kill. You wouldn't get a hunt, You wouldn't get like an animal. And if you just look at like most of, for most of human history most of our activities, especially our jackpot-oriented activities, didn't really result in anything.
Most of the time. Variable rewards were just the norm for most of kinda human history. And then with technology, the great thing is we've [00:15:00] created these systems where you can get rewarded every time. Where it's I can go into Instacart and get a steak every time, basically. And so actually I think that in many ways consistent rewards, which would be called continuous reward schedules are actually the evolutionary novel thing.
And they're the thing that like, I think really separates the all-star products that really become very habit forming from the products that actually fail. And so I actually think that like the people that I focused on variable rewards over other types of reward schedules have gotten it exactly backwards.
AMY: Okay, so now let's tie this all together. What are some practical steps that product designers can take to build lasting habits into their products?
Jason: Let's pretend you don't have a product yet. And you're just like you're a founder or you're somebody that has a new idea and you want to, build a habit-forming product from the ground up.
What should we think about first? What should we think about second? When it comes to habit formation, I really believe it's all about what I call behavior matching.
I don't believe in universals in general, right? The one [00:16:00] thing that product design has really taught me over the years is that if you're building a product, you have to really get into the weeds. You have to understand your users, you have to talk to your users. You have to ingest the content that your users ingest. You have to really do everything you can to gain that, like deep empathy and understanding of them.
There's some things that are universal, which is all humans want to communicate with other humans and be social. And that's why messaging apps are like reten and ubiquitous. And it's like you can build one app, like what's that?
And have, billions of people use it or Facebook, right? Everybody wants to keep in touch with people that they know or people that they've met. They wanna see how their lives are evolving, et cetera. But for most products, most products are more niche, right?
And you're really building for not all of humanity, but you're building for like a specific target market or a specific user group in a specific place. And you really need to go down into the weeds, understand them, and just really empathize with them and understand everything about them.
And the reason that's so important [00:17:00] is because you wanna know are we asking them to do something that one they're comfortable with? Are we asking them to do something that they value, that they think is really going to add, value to their lives? Are we asking them to do something that's like within their abilities, right?
Are we asking to do something that's too complex or too challenging given their specific background skill sets, et cetera. And. From my perspective habit formation design, a lot of it is really about picking the right behavior, the right solution for your target audience to do.
And if you pick the right thing, if you pick something for them to do that they see the value in that they enjoy doing, that's within their kind of skill set of being able to do. If you pick something for them that really effectively solves their problem, and you shape it the right way most of your work's been done for you picking the right behavior.
Behavior matching is really so much of what, like habit formation design is all about. If you pick the wrong behavior, if you pick the wrong activity, the wrong solution for them , The audits are against you, right? You really have to, I think, pick from the very beginning, the right [00:18:00] activity, the right thing for the target user group that you're going after. Or, you know a lot of times what happens, right? You see this with startups all the time, is they'll build something for a specific target user group and then they pit that user group doesn't want what they've built, and so they just pivot on the user group, right?
They say, oh cool instead of this being for high school teachers, let's pivot it to like restaurant owners or whatever, right? So they pivot on the user group side of things in order to get that behavior match. So I talk about behavior scoring, which is the process of taking the different possible solutions or behaviors that we want to ask that we're planning.
Asking the target market to do and then figuring out, okay, how good is this idea? How good is that idea? And then rank ordering them and figuring out which one we're gonna use for habit design process.
Okay, cool, let's construct a product, a habit forming product from the ground up. Taking all of the principles and all the ideas that we've gone over so far. Let's build something from the ground up. Based on that, this one is really all about.
Okay. Most product managers are working on [00:19:00] products that already exist. And so there's a different approach that you would take to like analyzing a product that already exists and applying these lessons in a way that's impactful.
So this one is all about, okay, we're not building from the ground up. Instead we're taking something that exists already and we're deconstructing it given kind of these perspectives and given this research. And we'relearning how to apply these methods and these findings to our current product.
And one of the things I talked to here talk about here is what I call the four E's of habit problem fit. I really think that they're a kind of four major variables that determine whether or not you have, behavior problem fit or habit problem fit, which would be just.
Is the behavior slash the solution that we've picked, matched to the target user group that we're building for. The four E is really just breaking that down further so that we can actually think about it in a very concrete, granular way.
AMY: Thank you so you're overlapping with a lot of our most favorite topics and it's awesome. Like [00:20:00] feedback being the best former reward and debunking that external rewards are the solution. And then you touched on variable reinforcement mm-hmm and how important it is to pick the right behavior if you're gonna solve a habit problem And Many of those issues are commonly put forth as the solution to your engagement problems Mm-hmm and a lot of it boils down to Skinner box mechanics backfiring, Mm-hmm
AMY: That's my takeaway a lot of people have been trained in behavior science. Some people lightly some people deeply, right. Like you mm-hmm and me to some extent I did lab in school and all that You and I have very similar background and the thing is at like as a Psych student when I first did behavior lab and saw that those mechanics could make micer pigeons do things it blew [00:21:00] my mind. I thought Oh my God this is how you shape behavior, this is amazing. And I remember feeling that but then as a product creator trying to implement it, I could keep seeing what happened Yeah And so I think that your experience elbow deep in product creation versus punditry Mm-hmm I would like you to say un distilled Why doesn't it work Why isn't that the silver bullet that's been solved.
Jason: Yeah It's just the wrong level of analysis in general Like if you're like I think that thinking about reward schedules is um it's almost never useful in like a in a product design process the time should be spent really thinking about how do I reliably solve a key pain point or a key issue for my audience. It shouldn't be thinking about how do I delight them unpredictably in some of the time, right I really think that the focus for product designer should be on continuous reward [00:22:00] schedules let me just zoom out a little bit So some people create games that's great. Some people create the equivalent of like let's say a digital slot machine And for them it's like using kind of like Getting obsessed with reward schedules and uh building all these overlapping reward schedules may be a worthwhile activity but I think that what most of us are trying to do is create what I'd consider like utilities right We're trying to create products that solve some problem in users well and make our customers lives better
AMY: Yeah for sure That's not how you make your customers
Jason: Lives better, No, No, if you're building a utility your focus really should be on every time somebody comes into my app Or my product I want them to get what they need Like if I try to make Instacart uh variably rewarding um what I would do is make it so that like some of the time you can find all the groceries that you want or that sometimes you order and your order gets canceled but the next time that you order it comes through
There's nothing useful about arbitrarily [00:23:00] adding variability into your reward schedule your reward system in a utility app which is what most apps are. I think instead the focus really should just be on every single time person comes in I just want them to have a great experience get what they need you solve their problem which is the reward right for most apps. I can't think of a single instance in which I've seen somebody take the variable reinforcement idea apply it to a non-game. Um and for it to make the product better in any way. I just think it's kind of like one of these ideas that like if you aren't well studied in in kind of the behavioral sciences if you haven't thought about this stuff for a long time you're just like, oh wow. that seems that sounds scientific and compelling but when you actually try and practically apply it like it's just kind of a dead end.
AMY: Well if you're gonna take that toy away from me what can I do And that's where your you're step-by-step system comes in et cetera et cetera.
Jason: Yeah Yeah I think it's all about picking a solution or a behavior that's intrinsically rewarding and intrinsically compelling for the [00:24:00] target market you're building. For my approach is really all about yeah what I call behavior matching or behavior selection It's really about just like making sure that the core activity the core action inside of your app your product your service is the thing itself is rewarding is like uh solves their problem in a way that's uh you know enjoyable easy effective in a way that's exciting that's really like what my system is all about It's just doing that in a very systematic and clearheaded way, right?
AMY: And that's why you have to know who you're designing for. So, thank you so much for your time Jason.
Jason: Oh, it's my pleasure.
AMY: Bye
[00:25:00]