The first 90 days of a new product

Most new products spend their first months building things users will never notice. The work that matters early is narrow: one clear job, done well, in front of real people as soon as possible.

Days 1 to 30: the core path

Pick the single thing someone should be able to do with the product, and build the path to it end to end: sign up, do the thing, see the result. Make it plain but complete. A rough path that works beats a polished half of one.

Days 31 to 60: real hands

Put it in front of a small group of real users, ideally people who have the problem and will tell you the truth. Watch them use it. The questions they ask in the first five minutes are your roadmap.

Days 61 to 90: the second loop

Fix what confused people, cut what nobody used, and add the one or two things they kept asking for. Then decide, with evidence, whether to charge, scale or change direction.

“The best feature in the first quarter is a user who came back on their own.”

What to skip

• Admin dashboards beyond the minimum you need to support users.
• Settings and customisation nobody has asked for.
• Scaling for traffic you don't have yet.
• Integrations, unless the core value depends on one.