What exactly is a
■ I’ve found it means different things in different companies. I run experiments to try and understand how we can help people save money and get out of debt — I mean formalized experiments in randomized controlled studies to show what are the obstacles that people face. Everyone knows they need to save. Everyone wants a secure retirement. Everyone knows getting out of debt is important. So telling people that’s important doesn’t have any impact. How do we help people to take the actions that they already want to take?
How do your
■ First, we identify what is the particular user problem that people are struggling with. Is it debt? Is it retirement? We do a mix of quantitative and qualitative research. There’s another person at HelloWallet who does user research. We look at what might be behind that problem. Is it a lack of attention because people are so busy with other things? Is it a fear or baggage from the past? Or is it a problem of not being urgent, like retirement, something that is important but never seems urgent?
Every company is doing some kind of experimentation on their product because that’s the core of improving your product. We just do it with a science. It gives us a guidance. It’s just the normal experimental method.
Give me an example.
■ A core feature of our product is the HelloWallet Wellness Score, which is a score of how you’re doing financially. That’s something that came from this research. The score is an object, clear and simple single number that a person can focus on, get immediate feedback when they make progress and small wins. We also throw in an angle of peer comparison; you can see if you’re doing well or not compared to others. That tends to be motivating.
How did you come to learn about behavioral science and choose to pursue it professionally?
■ I was doing a PhD in political science and was studying economic models of voting behavior — until I learned that economic models which assumed a rational person calmly deciding about the costs and benefits of voting just didn’t work. People aren’t rational optimizers in their daily lives; there’s something more going on. I learned that behavioral science was on the cutting edge of figuring out what that “something more” is — by systematically studying how our environment and our cognitive biases affect our decision making.
Did you have an “aha moment” when you realized that behavioral science had business implications?
■ Absolutely. At HelloWallet, we started off with a very standard approach of calculating the optimal financial behavior for a person and then yelling at them (digitally) to go do it. That didn’t work at all. So, we had to find something that did work or our business model of unbiased guidance would fail, and I’d be out of a job.
We turned to behavioral science because without it, we wouldn’t exist — we need to actually help people take action on their finances for our clients to pay for our software. And, as I was researching my book, I realized that many other companies were in the same boat — if they didn’t help their users change behavior, they wouldn’t be in business. This is true whether for companies that help people exercise more, such as Jawbone’s Up, an activity monitor worn on the wrist, or learn a new language, such as Rosetta Stone’s software.
Should all companies use scientific testing then?
■ Companies are still trying to figure that out. Companies are just starting to find people who can do that dual role; who know the research and have a strong focus on products that can apply the research. The book I wrote is all about the problem of not having clear guidelines for companies to apply behavior sciences in their business. How do you take all of this shiny new research and how do you apply it to a product to help people take action?
What’s the first step?
■ Identify what obstacle people are facing first. Is it a lack of attention, desire or resources? Is it procrastination? Then, pick the tool that fits the problem. Often, you can use simple tools, such as building default settings or reminders into the product, to solve the problem quickly. Just remember that human behavior is tremendously complex, so plan on getting it at least partially wrong the first time. Plan on testing, iteration and refinement.
How is behavioral science different from manipulation?
■ The reason you buy a product is because you want to change something about your life. Behavior science is about delivering the value that a company has promised when you want to change something, but you are standing in your own way. It is not how you trick them. That’s a very different thing, and thankfully it’s much harder. You’re not twisting their arm. You’re helping them do things they want to do.
Is it possible for companies to become too hung up on scientific testing?
■ You’ll never do as many tests on products as you want because the company has to adapt quickly. But everyone believes a new product or innovation is going to be better because you see the problems with the old product. That may not be the case if you haven’t tested it appropriately. Scientific testing makes a better product rather than just making new changes. It’s about building more effective products.