You’re Creating Products the Wrong Way. Innovate by Tapping Into Behavioral Science Instead

The best product ideas don’t come from your customers.

At least, not in the way you think.

Instead of asking customers what they want, you should be putting on your detective hat and asking through the lens of behavioral science instead.

That’s why I started Creative Science — and it’s what makes me excited to do this work every day.

When product managers and designers develop a deep understanding of human behavior, they can innovate in ways they never thought possible.

The best product ideas don’t come from your customers.

Behavioral science vs. data science

I first want to break down the difference between behavioral science and data science.

Behavioral science studies the social, emotional, and cognitive factors that drive us to take action.

There is a lot that goes into our decision making that we’re not consciously aware of.

Behavior science shows us the things that are going on in our environment and are affecting our decisions.

It’s the ying to the yang of data science.

Data science tells us who is likely to do what. Behavioral science tells us why a person does something and how to get them to do it.

Behavior-first problem solving

When a change needs to be made in your product, it’s easy to get in the habit of thinking solution-first.

But that doesn’t solve the underlying issue. It just adds layers of processes.

It’s always better to think behavior-first.

Zero in on the behavior that you’re trying to change and then work backwards from there.

For example, if there’s an insurance company trying to reduce the number of fraudulent claims, most people would say that there needs to be better checks and balances for catching bad actors.

A behavior-first approach would say that the real problem is a broken system of trust.

The solution to the problem becomes: how do we create a system that makes people more trustworthy?

how do we create a system that makes people more trustworthy?

This problem has been solved with behavior science in real life.

Dan Arriely, the Chief Behavior Officer at Lemonade insurance, realized that people are less likely to lie on video than in writing.

So Lemonade started requiring people submit insurance claims using video instead of a written form. The result? An immediate recursion in false claims.

The importance of “why” data

Most companies are missing an important piece of data when they’re building a product.

They usually collect two types: “say data” and “do data.”

  • Say data is everything that a user tells you. You collect that from surveys, interviews, and focus groups.
  • Do data is analytics. Companies will collect that from Google Analytics, heat maps, and tracking tools.

Often, the two don’t match up.

For example, the user told you that they’ll use a feature if you build it. But then you see in the do data…and they don’t use it like they said they would.

That’s where a third type of data comes into play: “why data.”

Why data explains the subconscious factors that are actually driving user behavior.

Once you understand that, it creates a framework to then create products and make product decisions.

How to gather “why data”

One of the ways you can get that data is by split testing using a behavioral science concept.

Say you have an app that helps people save money and you find that people are staying signed up, but they’re saying they’re not saving because they don’t have enough money. Dig into the academic literature to garner ideas of what could be causing the issues you’re seeing.

You can test a theory like mental accounting. It says people tend to bucket money differently depending on where it came from.

Knowing that, you can set up a landing page or Facebook ad that tells people to sign up and save their tax refund money.

If you get more conversions, you know you’re onto something and can move forward with similar ideas.

3 behavioral science concepts every product manager should know

Now that you know what behavioral science is and why it’s important, let’s get into some of the concepts that influence product design.

1. Storytelling: The identifiable victim/beneficiary effect

This concept comes from the nonprofit space. It’s the idea that we have a tendency to offer greater aid to a specific individual versus a group of people.

We develop deeper connections with people individually — and there’s no greater way to create that connection than through story.

As a species, we’re not great with numbers and data. We think about the world and we process information with stories.

Think about the best startup pitches you’ve ever heard. My guess is that they revolve around stories.

2. Herd effect and social norm theory

This one is pretty straightforward: we follow the crowd.

The ways in which this concept can be leveraged are infinite.

If you want someone to do something, give them insight into what other people like them are doing.

3. Present bias and hyperbolic discounting

These two concepts work together.

Present bias is the idea that we heavily over-weigh our present situation versus the future. We prefer immediate rewards versus delayed gratification.

Because of the way that time works, we heavily discount anything that happens in the future. This is known as hyperbolic discounting.

This means that any pain or gain that happens in the future is downsized in our minds.

These concepts have a big impact in the business world. Think about a digital product that asks a user to take an action today that has a potential long-term benefit.

If you have an app that encourages users to save money, eat better, or learn a new skill, you want to have a deep understanding of how to counteract these biases.

One way that companies account for this is with gamification models or reward paradigms.

We want rewards for the actions we take. But often the reward for saving money or eating better is so far in the distant future that we have to substitute that in some way.

That’s what those models do. To use them effectively, you have to understand the science. Otherwise, applying them can backfire in the wrong hands.

Check your ethics

Companies mostly use their knowledge of behavioral science for good, like a FinTech app nudging a person to save more money.

But it is possible to use these ideas for evil. Some giant tech companies have already used them for not so good things, like infinite scroll.

These tactics and nudges should be used to help users do the things they already want to do.

If you’re going to integrate these concepts, you have an ethical and moral responsibility to ensure that you’re doing it in a way that is bettering the end user.

These tactics and nudges should be used to help users do the things they already want to do.

My baseline goes back to these two questions: do the types of things that you’re trying to get your users to line up with their initial intent? Are your tactics helping your user to achieve their goal?

Personas and behavior

It’s not clear how valuable personas are to predicting user behavior.

What we do know is that personality profiles may not that useful.

Behavior might be driven less by a person’s personality, and more by what’s going on in their environment.

Personas can be informative for applying behavioral science tactics though.

Many cognitive biases are common to all people, and can be used to predict and nudge behavior across the board.

But what we’ve learned in recent years is that some cognitive biases have a strong correlation to certain cultures, religions, and other demographics.

I’m still not 100% sure how much we can use personas to make those decisions, but it’s definitely something that we continue to research on an ongoing basis.

3 ways to generate ideas and level up

Product managers who want to get better at applying these ideas to their products need to do three things.

You become really good at this just like anything else: it’s pattern recognition.

  1. Read. Read books, articles, and everything you can find on behavioral science. I recommend starting with “Nudge” by Richard Thaler and Cass Sunstein, “Thinking, Fast and Slow” by Daniel Kahnemman, and “The Elephant in the Brain” by Kevin Simler and Robin Hanson.
  2. Get your hands dirty. Start using the ideas you learn about in your products and test them out. You’re not going to learn this until you start banging it and banging it and banging it and seeing what works and what doesn’t.
  3. Reverse engineer. Go and look at different products and check out their user interfaces. Find what behavioral science has been built into them. Our biggest breakthroughs come with a new intervention or a new design that is pieced together from four or five different things that we have seen work.

You become really good at this just like anything else: it’s pattern recognition.

Over time and with experience, you build up that knowledge bank. And that’s what becomes really powerful.

Innovation is bringing different disciplines together

The way to keep innovating and keep generating great ideas is to apply the behavioral science ideas you learn to all the different disciplines you’re working with.

Where I get most excited is the cross-disciplinary application of various ideas.

Behavioral science itself is layered upon anthropology, sociology, and many different disciplines.

Everything can be so siloed. But innovation happens when you start to understand how everything fits together.

Decoding the Why — How Behavioral Science is Driving the Next Generation of Product Design.

If you’re interested in learning more, grab a free e-copy of my book Decoding the Why — How Behavioral Science is Driving the Next Generation of Product Design.

Promo Code: Medium2021

Decoding the Why Book

If you want to connect, you can reach me here via email (, or connect with me on LinkedIn.

Entrepreneur blending behavioral economics, data and technology to help changemakers inspire and move people to action. Co-founder @creativesci & member @yec.