Customer Simulation constructs synthetic buyer personas — the spec-obsessed Reddit researcher, the safety-anxious parent, the deal-hunting comparison shopper — calibrated against real review language and competitive spec benchmarks from the Golden Master Ontology.

It runs your proposed PDP, pricing tier, or new SKU concept through each persona and predicts exactly where they abandon. "Technical Buyer rejects: no oven-safe temperature specified." "Safety Buyer rejects: no third-party PFAS-free certification logo."

Output is a Simulation Scorecard — a per-persona pass/fail matrix with specific failure reasons and recommended fixes ranked by conversion impact. Each simulation runs in minutes, not the 4–6 weeks focus groups require. You review every recommendation before anything changes.