How it works
- Press Run panel. Simulated buyers walk up to your stall and read your pitch.
- An amber ring means a buyer is thinking.
- Each buyer walks to BUY or PASS; the reason appears in Live decisions.
- Results then show who buys by segment, why the rest pass, and the next test to run.
Quick sim = instant rule model, free, not AI. AI buyers = one isolated model call per buyer. Replay = re-watch a saved panel. All are simulations, not customers.
Price answers appear here as bots decide (live preview), then the skill's van_westendorp.py result.
Offer B
Same buyers (same seed) read a different pitch. The only fair way to see if an offer change helps.
Run panel A first, then B.
Review loop
Five reviewer bots apply the board lenses to your pitch and the last run. An editor applies their fixes, the same buyers judge the new version, and the loop keeps a version only when more of them buy. Your pitch is only replaced if you press Use as my pitch.
- Baseline: the buyers read your current pitch.
- Review: the Offers, Product and Monopoly lenses, a clarity check and the CFO lens walk up to the stall, read the pitch and the results, and post FIX, OK or YOUR CALL.
- Fix: an editor applies the FIX items. YOUR CALL items (facts, prices) stay with you.
- Judge: the same buyers (same seed) read the new version. More buyers: kept. Fewer or equal: reverted.
- Repeats until your target is met, the rounds run out, or two rounds bring no gain.
The verdict only comes from the skill's script: every unit earns money, year 1 profits, break-even fits capacity, enough of the panel buys.