Launch and find your first users
Chapter 4 — one channel, ten users, honest telemetry.
A launch is not a milestone; it's the start of the evidence stream.
One channel, not five
Pick one primary channel and commit to it for a defined period. Where do your ICP members already gather — a subreddit, a newsletter ecosystem, marketplaces, communities, cold outbound? Go where the audience exists instead of building an audience from zero while also building a product. Five channels done at 20% each reads as noise in all five.
The first-10-users plan is concrete: name the ten places or people you'll pitch this week. Not "post on Twitter and see" — names, threads, groups.
What launch is actually for
The launch produces the raw material of your next three decisions:
- Usage — does anyone complete the one workflow you built?
- Objections — the exact words people use when they don't sign up. Write them down verbatim; they're your positioning rewrite.
- Pull — does anyone ask for it when you stop pushing?
Treat these as experiments with targets, not anecdotes to sample on the days you feel good.
Set the review date now
Before you launch, set a date 4–6 weeks out. That's when you decide — formally — whether the launch produced evidence worth building on or a lesson worth recording. Undated launches drift into indefinite "traction" that is actually just ambient activity.
The unglamorous checklist
A working onboarding path, a way to contact users, error tracking, and a feedback channel. Everything else is post-evidence. If you launched from a scaffolded workspace with connections registered, this list was half-done before launch day.
Next: Pricing and the first dollar.