Saturday, February 06, 2010

Why Default Settings Matter

In a fascinating talk at the TED conference, Dan Ariely asks the following question, "What determines whether or not someone decides to donate their organs?" When you look at the data, there is a striking difference between two types of countries in the world. Some countries have an organ donation rate of close to 100% while others hover much lower -- not reaching beyond 30%. What could possibly account for this difference especially among countries with similar cultural and ethnic heritage? Why would Austria be so much higher than Germany? Or Sweden be so much higher than Denmark?

Arielly points us to the paper "Do Defaults Save Lives?" by Eric J. Johnson and Daniel Goldstein. A default choice is the choice that a customer makes by doing nothing -- e.g., always using Internet Explorer to browse the web because that's the way things worked when you bought your computer (in fact, the Microsoft antitrust case in the late 90's revolved to a large extent around these default choices.)

This startling difference in participation in organ donations is tied to how the question is worded at the DMV. In some countries, the question is worded "Check this box if you want to be an organ donor." In others, the question is worded "Check this box if you do not want to be an organ donor." If the question presumes that you will donate an organ, you will. As Arielly points out, it's not because people are lazy, it's because, the decision is so important that it is paralyzing for people to make for themselves. They shouldn't have to choose what to do -- someone should make the choice for them and "the sysem should just work."


When the people at the DMV created these forms, did they realize that they were going to have the single most important effect on the state of organ donations in their country -- certainly not. But that's the point. To a large extent, the default behaviors that the designer created are those that will stick.

Certain choices, especially those with high stakes and low information are very hard for people to make. Let's look at one product in specific that people need -- 401(k) programs. Logically you would think that the more choices you gave someone on their retirement plans, the more likely they would be to join one of the programs. However, as a company adds additional plan choices the rate of participation actually drops because consumers don't know what to do with all of these extra choices. Richard Thaler and Cass Sunstein talk about this paradox at length in their book Nudges. They discuss an innovative solution to the 401(k) problem with the "Save More Tomorrow Program." The idea is that most people don't change their 401(k) options once they join a company. Therefore, the plan should, by default, change your options for you. So this plan increases your 401(k) contribution at the beginning of each year, right at the time you get a raise. This is around the same time most raises are given out so people don't feel like they are losing any money. The company does the "right" thing for you if you decide to do nothing. This is an example of nudging the customer in the right direction or in technical speak "Libertarian Paternalism." It allows people to make their own decisions, but presents them with the best option for themselves as the default choice. It's the equivalent of saying "You can have whatever you want with dinner but I'm serving skinless chicken with fresh vegetables."

As product designers, we often expect customers to make the best choices for themselves given enough options. However, as we have seen, in many areas, people don't choose what is best -- they choose what is easiest. There are very few products that customers will engage with and learn how to use the myriad of features. If you still don't believe me, take a look at this New York Times article on the relatively few applications downloaded on the iPhone. Remember that for the vast majority of consumers they won't be changing their default settings, so make sure that the product works great right out of the box.

No comments: