I never really ran the hard numbers. Once I get an idea in my head, often my first step is to start building it. Code first, think of the business model later. I’ve learned that starting to test the idea with people first is the key to any good project. But testing with users doesn’t give us revenue projections, it doesn’t give us churn rates, or burndown rates.
It simply tells us, if the thing we want to create has any value to a person. I still skip the revenue projections, I’m focused on building good things that people will love.