porbot
1.9 KiB
1.9 KiB
Genre packs
Structural direction (working): three layers when surfacing work to owners — not one giant schema dump.
| Layer | What | Owner hears |
|---|---|---|
| A — Universal | Discovery foundation: NAP + Core 3 claimed + hours + website + description + service-area mode + no duplicates/contradictions | Everyone’s first punch list |
| B — Genre pack | Thin ranked list for that vertical (secondaries, high-value attributes, services/menu shape, booking/links, maybe one vertical directory) | “For salons like yours, next is X” — never “schema” |
| C — Known-for | From the owner’s own words | Extra attributes/photos/proof + later AI-readiness queries |
Surface rule: scorecard ladder first (exist → know what you do → would recommend), then 3–7 actions max, sorted by impact. Genre pack ranks the backlog.
Core 3 coverage rule: GBP is richest; Apple Business Connect mid; Bing Places thinnest. Pack actions: set on GBP; mirror on Apple/Bing where the field exists.
Deep example packs (planning-rev2)
Earlier planned packs (from main / prior docs)
- Auto repair (see UC-01)
- Beauty (overlaps salon — reconcile later)
- Home services
- Tourism activity (UC-02)
- Local retail / general store (UC-03)
Why this matters
Assistants evaluate helpfulness by whether the business answers the right questions for its category. Salon booking ≠ bakery known-for products.
Detail source
Conversation checkpoint on planning-rev2; BUSINESS_PLAN.md; Canonical Strategy §1a.