What you get: 25 buyer questions × 4 assistants × 3 runs = 300 answers, read by hand. Share of voice against up to 5 competitors, a map of the cited sources, a technical readiness check of what AI crawlers can read, a dated 30-day fix list, the raw data and a 45-minute walkthrough.