Don’t ask users what they want

It’s a common misconception when undertaking user research that we need to survey a bunch of users and find out what they would like in (a new version of) a website. I’ve had clients sometimes get skeptical about doing user research at all for this reason and I sympathise with them. They’ve spoken to their clients before and had requests for all sorts of things that weren’t suitable. By asking people what they want on a website, you’re asking them to think about the problem you are trying to solve. That is not the aim of user research. Instead, what you need to focus on is finding out what the users’ real needs and goals are.

A user may not find it easy to articulate his goals. Often he may say that he would love to see a feature X on the homepage. But when you actually take the time to understand his situation, it is clear that he’d never really use X. Perhaps he saw it somewhere else and thought it looked ‘cool’ and that’s why he’s suggesting it. People usually like to be helpful, after all. And it may well be that understanding his real needs will prompt you to dream up feature Y. Feature Y may not even exist yet, it may be a new invention.

“If I had asked my customers what they wanted,” Henry Ford said, “they would have said a faster horse.” Some people actually use that as an argument against user research: don’t listen too closely to your customers. But Ford was successful because his product addressed people’s underlying needs. And that is what proper user research is all about discovering.

When gathering information on your users, using online surveys is a fairly limited approach. Asking them what they would like is worse. The best thing to do is ask them to talk freely about what they do on a daily basis within your domain of interest. Then ask them why they do that. A little bit of empathy is useful, and there is no harm in collecting information on deeper, longer-term goals. Good design is, after all, about really understanding the user.