pick one customer before you chase users
name one customer so specific you could almost name the person. build for them, then widen.
what to do
Write down one customer profile so specific you could almost name the person: what they do and where they already hang out. Build and sell to that one profile before you touch a second type of user.
for example
A founder I spoke with in 2026 kept bouncing between a trading bot, a location app, and a billboard marketplace. Each one stalled at launch because the people who would use them had nothing in common. The fix was to pick a single profile and point everything at it. Adoption only moved once there was one person the product was obviously for.
the failure it prevents
Three half-built products that each reach a handful of strangers. When your users share nothing, word of mouth has nowhere to travel, and you cannot read what to build next.
when this doesn’t apply
If you already have a dense group using one thing and asking for the next, you have earned the right to widen. The rule is for the blank page. Once a product already pulls, widening is the next move.