Playbook — Account Managers

You manage a book of live clients. The platform's job for you: kill the 15-minute lookup, and walk into every QBR/renewal already prepared.

Your front doors

  • Dashboard — your AM Portfolio, account pages, and the in-account advisor (enrichment already loaded).
  • HubSpot card — the Prospect Advisor in the deal sidebar; opens with a precomputed brief for the deal on screen.
  • Roam — the same advisor in chat, on the move (name the account).
  • Claude Code skills/check-account, /deep-dive, /portfolio, /am-review for power use.

Daily flows

Want How
Next bill / MRR / product mix / suspended lines / open tickets /check-account or just ask the advisor — the data's already on screen
"What's this deal about?" in HubSpot Open the card → the deal brief answers in ~30ms (CRM + GBP + website + similar clients + recent conversation)
QBR prep (day before) /deep-dive or "Deep dive on [business]/[sub id]" — program, MRR, renewals, DEA periods, ads/GA4/GSC/GBP, reviews, risks, opportunities, health score
Performance read Ads/SEO with the caveats baked in — pre-calculated CPA/CTR, Search broken out, PMax parity caveat before any budget-shift rec
Your whole book AM Portfolio — accounts, MRR, setup/DEA, health, tickets, deals, reviews

What's new since the v1 deck

  • The HubSpot card (was "exploring") is shipped — deal-anchored advisor in the sidebar, with voice dictation.
  • Precomputed deal briefs — opening a deal no longer waits on live lookups; it's instant, and refreshes in the background.
  • Deal-pattern mining (extract_deal_patterns) — recurring win/loss themes for your category.

The discipline that makes it trustworthy

  • It won't calculate metrics itself — it quotes pre-calculated values or says "not available."
  • Ads spend is retail (what the client pays FCR), never "media spend."
  • It won't claim a prospect "has no Google Ads" — our data is ~3–5% of the market.

See ../../AI_ADVISOR_GUIDE.md for the gold-pattern prompts.

FCR Dashboard documentation · generated from docs/ · keep counts verified, not guessed.

Ask the docsRAG over this site
Ask anything about the FCR Dashboard platform — architecture, BigQuery, the worker routes, billing rules, the LRC stack, scoring… Answers are grounded in this documentation, with source links.
How does the deal-brief refresh work? Which routes are Worker vs n8n? How is account health scored?