Libertarian Paternalist Product Design
Product designers can borrow an idea from economics to set a threshold for how to direct users in a helpful direction
If you’ve ever paid for a cab using a credit card, you’ve seen an example of the tremendous amount of power product designers have when it comes to shaping our behavior. Even though you probably consider a standard tip for a cab ride to fall somewhere in the 10-20% range, the single-tap choices offered in a New York City cab range from 20-30%. In the absence of a conveniently accessible, lower alternative, I always tip 20%— the high end of what I consider a normal range. I’m sure I’m in the majority in that sense, and I’m sure that’s the intended consequence of the design.
There are a number of incentives for designing choice selections in this way. The cab interface’s is obvious: to get higher tips. But there are many other types of scenarios when tactics like this are used.
Sometimes the motivation is paternalistic, which is to say that the design is meant to generate behavior that is for the user’s own good. To make another New York reference, think of that city’s soda ban as an example of such paternalism.
But there are other times when the motivation for design nudges is mixed, like when they are used in (or as) advertising products. When Mint suggests you open up a new account offered by a partner to meet one of your financial goals, you may be seeing this in action.
These may be two great financial products for me to use to start saving towards the goal of buying a home. But you could argue that Mint is both trying to help me and trying to sell me at the same time.
So it can get fuzzy. How do we decide when behavioral nudges are ethically OK to employ in product design?
I think the idea of libertarian paternalism can offer a strong guiding principle to designers here, despite it being an idea from the world of economics rather than product design. The idea— which is invoked in strategies for shifting behaviors in spaces from retirement savings to weight loss—is often summarized in this way.
Influence choices in a way that will make choosers better off, as judged by themselves.
It’s that last part that’s the kicker, and the one that product designers— and even the researchers who advise them— can easily forget. (I know I sometimes have.) And that is nothing malicious; it’s often simply a byproduct of your incentives being overfitted toward one specific objective. The client asked for a product that produces higher tips? Easy!
But users’ objectives, of course, may be different. As a cab rider, I want an easy way to appropriately tip my driver and get on with my life. And you know what? I’m even open to being nudged toward a higher tip than I may have thought to offer on my own. It might make me feel good about myself.
Tip selection design as I described would fail this test because it places a tax on making an alternative choice. Not only do I not see an easy way to tip a lower amount, I know that I will have to tap that nonresponsive glass screen at least five more times to give it, and I’m in a rush. I choose the 20% option, but I do not leave the cab feeling as though I am better off. If I had had the clear option to give 15% but was subtly nudged toward 20% and did so, I would feel much better about it.
Keeping true to the principle of libertarian paternalism in product design really boils down to one very simple thing: choice design should be clear and obvious, and the costs associated with going in a non-nudged route should be trivial.
As a user, I may get value from a design suggesting I move in a certain direction. But in order for me to judge that I am better off than I would have been otherwise, I need to feel as though I had a choice to begin with.