QBR Prep
Pull account usage and health metrics from connected SaaS systems and assemble them into a structured quarterly business review (QBR) deck outline file for the customer success team.
When to use
- A QBR is scheduled for a named account and metrics need to be compiled beforehand.
- A customer success manager asks for a fresh outline of account health, usage trends, and renewal risk.
- Recurring end-of-quarter prep cycle for a book of accounts.
Tools
http-get— read account usage, billing, and support-ticket metrics from connected SaaS systems (CRM, product analytics, support desk).shell-execute— compute period-over-period deltas, aggregate metrics, and format tables from the pulled data.write-file— save the finished QBR deck outline as a file artifact.
Playbook
- Confirm the target account name and the quarter/date range to cover.
- Call the CRM REST API with
http-get(e.g. GET https://api.hubapi.com/crm/v3/objects/companies/{id}) to pull account details, contract value, and renewal date. Authentication is injected by the org's Integration row for this host — never ask for, echo, or hardcode credentials. If the call returns 401/403 or no Integration row exists, stop and message a human via send-message asking them to create the Integration, naming the service and required scopes. - Call the product analytics API with
http-getto pull usage metrics (active users, feature adoption, login frequency) for the account over the target period, using the same Integration-row authentication pattern. - Call the support desk API with
http-getto pull open ticket count, average resolution time, and any escalations tied to the account. - Use
shell-executeto run a small script that merges the pulled data, computes quarter-over-quarter deltas (usage growth/decline, ticket trends), and flags risk signals (e.g. declining usage, unresolved escalations). - Structure the results into standard QBR sections: account overview, usage trends, support health, risks, and recommended next steps.
- Use
write-fileto save the outline as a deck-ready markdown or text file, named clearly with the account and quarter. - If any metric needs a correction pushed back into the CRM or another system (e.g. updating a health score field), that is a PATCH/PUT operation and is NOT POSSIBLE with this skill's tools — compile the change into the outline file and hand off to a human via send-message rather than attempting the update.
Failure modes
- 401/403 response from any API -> missing Integration for that host; stop and escalate via send-message naming the service and required scopes.
- Empty or zero-value metrics returned -> verify the endpoint and query parameters (account ID, date range) before treating the result as real.
- Rate limit (429) response -> back off and retry later; note the limiting service in memory via update-memory so future runs pace requests accordingly.
- Source page or dashboard structure changed and expected fields are missing -> fall back to browser-search to confirm current field names or endpoint paths before retrying.
Done when
- A QBR outline file exists on disk, written via
write-file, containing account overview, usage trends, support health, and risk sections populated with real pulled metrics. - All metrics in the file are sourced from live
http-getcalls made during this run, not stale or placeholder values. - Any required system-of-record updates are listed as a human handoff rather than silently skipped or falsely marked complete.