The Early Access Playbook v2.0

Written by AI

TLDR: Hiten Shah’s six-play early-access method turns real customer work into evidence for deciding what to build, change, retest, or stop.
Best for: Founders and product teams deciding what to learn before widening access
The tip: Track one Working Bet and Learning Thread, then let real use change the next decision.
What you need: A product idea, feature, or working prototype
People who already do the job
Start here: Write the five-line Working Bet.
Get one piece of real evidence from a recent job.
Record the next decision and what would change it.
Filed note: Hiten Shah’s six-play guide to testing products with real users before faster AI-assisted shipping outruns the learning.

Highlights: The faster you can build, the faster you need to learn.
Fast shipping only helps when the learning loop keeps pace.
You can generate a hundred plausible customer conversations without learning what anyone will do with your product.
Synthetic research cannot replace observed behavior.
Early access works when the evidence changes what you build. It may change which problem matters, how you solve it, or whether you should continue at all.
Evidence should change the decision, not merely decorate it.
Excerpt: Today AI can turn an idea into working software in hours. Teams can ship another version before learning what happened with the last one. I wrote this playbook to help founders and product teams learn while there's still room to change what they're building.
From docs.google.com