Skip to the page

Plainly Put

One sentence about building things, taken apart

HomeProduct

Ask customers what they want, then watch what they do

BlogHer began with a question about who was missing. Asking works best when the questions are about what people already do.

Wonder what your customer really wants? Ask. Don’t tell.

Lisa Stone

BlogHer started in 2005 with a question rather than a product. Its founders, Lisa Stone, Elisa Camahort Page and Jory Des Jardins, kept hearing that women were scarce on the speaker lists of blogging conferences, while plenty of women were writing blogs. They asked the women bloggers they knew what they wanted, and the answer was a place to meet. The first BlogHer conference that summer drew hundreds of people. The company that grew out of it now runs an advertising network that places ads on thousands of women’s blogs.

The line’s force comes from how rarely companies do this. Most product teams tell: they decide what the customer needs, build it and then market it. Asking first saves time and money, and it tends to turn up needs the team would never have guessed. It also changes the relationship: customers who were asked early tend to feel some ownership of what gets built, and they tell others about it. A conference built on what its audience asked for has an easy time filling the next one.

Asking has a known weakness. People are poor at predicting what they will do and often answer questions about the future with what sounds reasonable. Asked whether they would pay for a service, many say yes and never do. The questions that hold up are about the past: what did you do last time this problem came up, what did you try, what did it cost you.

There is also a limit to how far asking can reach. People can describe a problem well but rarely invent a solution to it. Shown something new, they judge it against what they already know. For products that change a habit, a team sometimes has to tell, meaning build a version and put it in front of people, because asking about an unfamiliar thing produces blank looks.

The sensible order is to ask about problems, tell with prototypes, and then ask again about what people did with the prototype.