Glossary
AI Dispatcher
AI Dispatcher — An AI dispatcher is a voice agent specialized in routing inbound service requests — triaging emergencies, booking routine jobs, and paging on-call technicians based on the call content.
What an AI dispatcher does (vs a regular AI receptionist)
A receptionist answers the phone. A dispatcher decides what happens next. The distinction matters in trades: a plumber's after-hours line gets calls ranging from "my faucet drips" to "I think I have a gas leak", and the response — book in 3 days, page the on-call tech in 30 seconds — has to be made on the call.
A trades-specific AI dispatcher applies a triage tree to each inbound call. The typical taxonomy:
| Category | Examples | Action |
|---|---|---|
| Emergency | Gas smell, flooded basement, no heat below 32°F, no AC above 95°F | Page on-call tech; SMS + voice page within 30 seconds; tell caller "tech will call back in 15 min" |
| Urgent | Water heater out, breaker tripping, drain backup | Book next available slot, same day if possible; flag for dispatcher review |
| Routine | Tune-up, filter change, leaky faucet | Book into next available slot; standard SMS confirmation |
| Quote / install | New system, replacement, addition | Book a 30-min assessment; flag as higher-value lead |
| Off-topic | Wrong number, vendor cold call, spam | Politely end call; tag for review |
Why generic voice AI usually fails at dispatch
Generic voice agents — the ones built on Vapi, Bland or Retell without trades-specific training — make three predictable mistakes:
- They miss emergencies. A caller saying "my upstairs is wet" might be a routine leak or a burst supply line. Generic agents book the slot. A dispatcher recognizes the ambiguity and asks the qualifying question ("Is water actively flowing?").
- They over-promise. Generic agents commit to "next-day service" because the prompt told them to be helpful. Dispatchers know the truck schedule and book accurately.
- They mishandle insurance claims. Roofing storm-claim calls need a different intake than a leak repair. A dispatcher knows to ask for policy number, date of loss, and to flag the call for an insurance-trained estimator.
The on-call paging integration
For emergencies, an AI dispatcher needs to do three things within 30 seconds:
- Capture the address, contact number, and 1-line problem summary.
- End the call with the caller (don't keep them on hold).
- Page the on-call technician via SMS + voice with the full call summary attached.
The on-call tech typically gets a Twilio-routed page like: URGENT: Mike Johnson, 412-555-0188, 14 Oak St, "smells gas in basement". Hi Agent ended call at 11:47pm. ETA target: 12:02am.
Why this matters financially
The average residential plumbing emergency call is worth $850 in revenue (ServiceTitan 2026 data, US average). The average residential HVAC after-hours emergency is worth $1,400. A contractor who misses 5 emergencies a month is leaving $50–$70K in annual revenue on the table.
The math is why AI dispatch — not just AI receptionist — became the dominant 2026 voice-AI use case in home services.
What an AI dispatcher is not
- It's not a Twilio Studio flow. Those are decision trees; an AI dispatcher is a model trained on real call transcripts.
- It's not a chatbot. The dispatch decisions need to be made on a live call, in seconds.
- It's not a scheduling app. It calls your scheduling app — ServiceTitan, Housecall Pro, Jobber — via the same API your dispatcher would.
Related terms