Growth as Compass

Once you commit to a target growth rate, running the startup simplifies enormously. Pick a rate you think you can hit, then try to hit it every single week. Judge everything by one test: will this get us our growth this week?

This turns the terrifying, open-ended problem of building a company into an optimization problem. Should you go to that conference, hire that person, chase that press hit, build that feature? Whatever gets you the growth you need is, as PG puts it, ipso facto right.

It also imposes honesty. A week where you were busy but the number didn't move was a bad week, no matter how it felt. The compass doesn't care about effort.