Making Smart Decisions Feel Sexy

Editor’s note: As you may know, I’m into money. I mean, not just in the way most people are “into money,” but also because it’s a great tool to study human behavior. Money is powerful – for better and worse – in part because it’s so specific and measurable. You can track changes to the decimal point. Other metrics, like job satisfaction and productivity and happiness, are harder to track (though we’re trying), but money … hoooooooo boy.

There are a lot of organizations trying to apply behavioral science to financial decision making … for better and worse. I’ve long had an eye on Dan Egan’s work at Betterment because they seem to be well intentioned with their interventions. So I got someone to interview him, and I think the details of how and why they use the science as they do can be quite useful for most industries, organizations and individuals. Get in the weeds with him and see what you discover.

As always with such PeopleScience articles, this is not an endorsement of a product or service. It’s a behavioral science story to share at our dinner party. Drink up and enjoy.


Daniel P. Egan wants you to retire by the beach somewhere, your kids college-educated, your car in working order, your finances sound. He wants to help you, and he has the technology to do it, but you have to behave.

Egan is a behavioral scientist with a focus in behavioral finance. After earning his MSc in Decision Science (“a combination of cognitive psychology, decision-making, probability and game theory,” he says) from the London School of Economics in 2005, he worked for several years as a behavioral finance specialist at a wealth management firm in the UK, where he noticed two things:

 (1) More and more financial advice and guidance was being done by technology behind the scenes;

(2) A client’s quarterly performance reports often were not tied or tailored to the initial savings plan drawn up when the client hired the firm.

 “We are fundamentally limited in how effective we can be if all we are doing is training human beings about biases,” Egan says. “We actually need to change the design, the interface that clients see in order to encourage good behavior, and to encourage thinking about the future.”


“We actually need to change the design, the interface that clients see in order to encourage good behavior.”

Enter Betterment, an online financial advisor where Egan has served as Director of Behavioral Finance and Investing for the past six years. Betterment plucks the algorithms out of financial planners’ heads, implements the algorithms into its software, and rolls that software out to half a million clients. It’s relatively cheap, the advice is available 24/7, and the technology creates opportunities for saving that a human advisor would not be able to create efficiently, if at all.

Egan spoke to PeopleScience about encouraging good behavior through design, the shifting role of financial advisors and getting people to move beyond just asking, “What happened last month? Did my stuff go up or down?”

 (This interview has been edited and condensed for clarity.)

Why are some people squeamish about a machine or algorithm managing their investments?

People inherently feel threatened whenever it’s said that a machine can do the things that they do. There’s a lot of pride wrapped up in our view of the essential novelty and thoughtfulness of what we do. But giving financial advice, portfolio optimization, constantly looking at portfolios and rebalancing them: This is the sort of calculating and mathematical heavy lifting that’s really boring for most people to do, and machines are great at it.

If you are a technology native — under age 35, grew up with smartphones and computers and the internet — you’re probably a lot more comfortable with the idea that technology can do these sorts of things. You might every once in a while want to talk to a human being, but it’s not scary or unusual to you that a machine manages your portfolio and gives you advice.

That said — if you’re older, or a certain type of younger person — there is a kind of weird trust issue where people don’t know to what degree the technology or algorithms know who they are or have asked the right questions, and they want reassurance and the ability to have a conversation with another human being who has that empathy.

That being the main blocker, you get into this really interesting new division of labor for financial advisors, even at Betterment. We have financial advisors, certified financial planners, who will talk with clients about their situations. These financial advisors are essentially the interface to the technology. For example, a client will say, “I set up these three goals. Are they right? Can you please check them and check that I put in the information correctly?” So the advisor checks their work and reassures the client that what’s happened is correct and everything’s fine. But even the advisors in this case are still using technology behind the scenes.


Financial advisors are essentially the interface to the technology.

What is the reverse Wizard of Oz analogy that you like to make here?

I think of this new division of labor as the reverse Wizard of Oz. In The Wizard of Oz, the wizard has this big machine that generates this spooky wizard, and behind the curtain there’s a human being running that thing. The reverse is that with a lot of financial advice, there’s a human being there who’s the interface, who’s talking to you, who seems like they’re the one giving you the advice—but they’re always looking at this computer for answers. So there’s a human being with a computer behind them, rather than a computer with a human being behind it.

Today we have fewer financial advisors who are focused on the investments — their returns, their characteristics — and a lot more of them who are focused on the clients. They say, “We have to have a real conversation about your financial goals. We have to be realistic: How much can you save? How much will you really earn?” Such that the data that we put into the machine and the planning that it creates is valid and good.

But ultimately, the computer is going to be better at the actual math and portfolio optimization and ongoing management. A person is not going to do that. A person is going to help you set up the technology correctly.

How do brokerage services’ incentives inform their app and website designs?

There’s a post about this on my website called “Incentives and behavioral design: brokerage UI edition”. The incentives of an organization or a company really inform its design, subtly, maybe even unconsciously. I don’t know how much the designers are aware of it. If you go to a brokerage website, where you buy and sell stocks and bonds, and you look at it from a behavioral scientist’s perspective, you’ll see that the site is optimized to make you do two things: trade your positions and hold cash.

That first incentive exists because that’s how the brokerage makes money. They like it when you trade, so a “buy and hold; don’t do too much” perspective doesn’t help them. As a result, they tend to emphasize recent performance, today’s or yesterday’s performance — was it up, was it down. They want you to trade as much as possible. They also foster a little bit of fear. Most people are risk-averse and loss-averse, so holding cash often feels like the safest thing to do.

What brokerage services won’t do is automatically tell you things that might halt or delay a trade, like, “Hey, if you sell out of this position now, you’re going to owe a significant amount of tax.” Because you might not think it’s worth it to sell after considering that tax.


That first incentive exists because that’s how the brokerage makes money.

How does Betterment’s design differ from most brokerage services’?

One of the things that’s different is that we ask you to set up goals. We ask you to think about what you’re investing for, what is the purpose of this, what is the future state that you hope to have that’s going to make it worth saving? Maybe retirement, having a safety net, having a kid’s college fund. When you log in and look at your financial situation, we want you to think about what you’re saving and investing for in the future, rather than how the market performed yesterday, which can’t inform any useful decisions.

It also lets us think about your financials in terms of on-track or off-track with regard to that plan. “Am I going to have enough money to buy a new car in three or four years, when I expect this one to break down? Am I going to have enough money to send my kids to college?” The planning aspect of it — thinking five, 10, 20 years down the road — is the valuable thing you can do for Future You. (Editor’s note: See also, this conversation about rethinking time.)

Our website is very much focused on your future plans and how we can achieve them. It greatly reduces the emphasis on short-term performance, performance of individual funds inside the portfolio, and any kind of loss-aversion triggering or risk-aversion triggering. Our experience is dramatically different from most brokerages because when you log in, you’re not treated to an “Am I up or down?” experience. You are treated to a “How am I going to be in the future?” experience. “What can I do today that’s going to improve my status in the future?”

How does Betterment “make safe options sexy and dangerous options boring”?

I’ll start with the latter. When you are processing a trade or an allocation change, we’re going to tell you about the taxes that will be due if you go through with it. We’re going to tell you about the implications of that for your future plans, whether or not it’s going to increase your chance of success or decrease it. It’s not like, “Do you want to sell this position and realize a 40 percent gain?!”, which would make a lot of people feel really good. It’s: “Do you want to sell this position and realize $2,000 worth of taxes?” Which, most people will say no, they don’t. Usually that’s good, because generally the longer you hold, the more things go up.

We also don’t show the performance of individual funds. We create a diversified portfolio because of the low correlations between all the assets in it. A brokerage site wants you to look at all of those things in isolation. They want people to say, “Oh, this is up and this is down,” and not think about the fact that that’s the whole point of diversification. That’s why you hold different things, so that when one thing is down, another is up. They want you to see yesterday’s numbers and bail on the thing that just went down and double-down on the thing that just went up. On our site, looking at yesterday’s performance to see what’s up and what’s down is very boring and barely even possible.


A lot of what we do is trying to make staying on track and focusing on the future sexy, rather than focusing on past outcomes that you can’t change.

On the other hand, we will show you that you have a 52 percent chance of achieving your goal—but if you increase your auto-deposit by 12 dollars, you’ll have a 65 percent chance of achieving your goal. If you decrease your risk a little bit today, there’s less of a chance that you’re going to underperform your goal five years from now. So a lot of what we do is trying to make staying on track and focusing on the future sexy, rather than focusing on past outcomes that you can’t change.

We also run a lot of randomized controlled trials on how our designs influence client behavior, and that’s been very helpful in ascertaining and learning and building up a knowledge base of what works and what doesn’t in the real world. As a practitioner in the field, I have a lot more confidence that we’re actually helping clients, because we can run these controlled trials and see what the impact is. The scale of the technology and the ability to do data analysis is a big advantage in knowing that you’re doing the right thing.



Get the latest behavioral
science insights.