Feedback collection.
Rings the customer the day after a completed job — a 30-second courtesy check-in that captures sentiment and surfaces reviews from happy customers.
Goal
Confirm the work was OK. Ask if anything needs follow-up. If they're clearly happy, ask once if they'd leave a review. If they're unhappy, capture the issue concisely so the team can follow up. Either way, the call stays brief — under a minute when they're happy.
Triggers
- ServiceM8 reports a job completed yesterday (within tenant-configured eligibility window)
- Feedback collection workflow is enabled for the tenant
What the AI does
- Outbound disclosure + the reason ('just checking in after Bill came out yesterday')
- Listens to their reaction and categorises sentiment (positive / neutral / negative) in their own words
- Positive + enthusiastic: if a Google review link is set up, makes one soft review ask and texts the link if they agree — no link set up means just a warm thanks, no review ask
- Negative or escalating: captures the issue without debate; offers a staff follow-up
- Closes warmly with name + thanks (no robotic 'Goodbye')
What the AI won't do
- Push for a review more than once
- Ask for a review at all when no Google review link is configured
- Read a review URL out loud — it texts the link instead
- Argue about complaints — it captures, you handle
- Drag the call past a minute when the customer's clearly satisfied
Tools available to the AI
send_review_linkschedule_callbacktransfer_to_staffTools are typed and bounded — the AI calls them, it doesn't define them. Each tool call is logged with an idempotency key, so retries never produce duplicates.
Success looks like
Toggle
Dashboard → Workflows → Feedback collection · on/off. Add your Google review link under Team → Google review link to turn the review ask on; leave it blank and the call stays a courtesy check-in.
Related
- All workflows
- Configuring the AI — what each workflow reads from your tenant config.
- Approval mode & graduation — how this workflow behaves in Approval Required vs. Monitored.
- Compliance & privacy — disclosure, DNCR, kill switches.