Why we stopped keeping a roadmap

For two years our product had a public roadmap: three columns, Now, Next and Later, with about forty cards between them. Customers voted on cards. We moved them around every quarter.

In January we took it down. Nobody has asked for it back.

What was wrong with it

A roadmap is a list of promises with dates attached, even when the dates are only implied by the column. Every time a card moved from Next back to Later, someone who had voted for it was disappointed, and they were right to be.

It also hid our real decisions. The roadmap showed what we planned to build. It never showed what we had decided not to build, or why, and that was the information customers needed most.

What replaced it

Two short lists, on one page.

What we are working on. At most three items. Each one says what problem it solves and who asked for it. No dates. When something ships, it moves to the changelog.

What we have decided not to do. This list is longer, and it is the one people read. Each item has a sentence or two explaining the decision. “We will not build our own video calls: the tools practices already use do this better, and we would rather connect to them.”

What we learned

Saying no in public is uncomfortable for about a week. After that it saves time, because the same request stops arriving every month. A few customers left over things on the second list. Many more told us that knowing where we stood made it easier to plan their own work.

The votes are gone too. We still listen to every request. We just stopped pretending that counting them was the same as deciding.