Building an MVP: a guide from 7 real products

In seven products I've led, the first release (a launch, a beta or, once, a full prototype) came 5 to 16 weeks after kickoff. Scope decided where each one landed in that range: how much had to be built, how much an existing tool could do, and how much had to be right on day one, more than how big the team was.

The timelines run from SportsOnTV, a sports channel and news site that went live in 5 weeks against an 8-week plan, to SmartPeople, an HR and payroll platform that held a fixed 16-week plan to get ready for launch. In between are Varsee's full prototype in 6 weeks, LeanDeck's beta with 6 agencies in 8 weeks, Mark Andrews Creative live in 12 weeks, and Find Your Calm and KpiClock at about three months each.

The fast ones integrated before they built, kept the first release to one job and put money on fixed rules. The long ones had money and access rules that had to work at launch, replaced a system a business used every day, or had to be several products at once.

This guide puts the posts in the order I'd use them on a new product. It starts with the timelines, then the scope: what goes into the first release and what moves after it. Next comes discovery, which turns a one-sentence brief into goals, users and rules, and the PRD that writes those rules down for developers. Then the build itself: the integrate-first approach that put SportsOnTV live in 5 weeks, the ten checks I run before a launch, and the switch-over plan for an MVP that replaces a system customers already use. It ends with the people question: whether to bring in a product manager, a product designer, or one person who does both.

Each post stands on its own, and each comes from a real product with a case study on this site. The PRD template and the MVP scope worksheet are free at the end.