What a home services AI receptionist actually does
Answers an inbound service call, triages it as emergency or routine, and books the job straight into the dispatch board a technician will actually see.
A home services AI receptionist answers calls for HVAC, plumbing, electrical and similar trades the way a dispatcher would: understand whether the call is a no-cooling emergency or a routine maintenance request, check technician availability through a tool call into the dispatch system, and either book the job or route it. The job isn't just answering the phone, it's getting the booking into ServiceTitan, Housecall Pro, Jobber or FieldRoutes correctly the first time, because a job that lands in the wrong queue or with the wrong urgency flag costs a truck roll a tech didn't need to make.
Avoca closed on $125M+ at roughly a $1B valuation in April 2026, co-led by Meritech and General Catalyst, built natively on top of ServiceTitan and reportedly serving 800-1,000+ operators. ServiceTitan itself shipped AI Virtual Agents directly inside Contact Center Pro the same year, a sign the system of record is building its own answer rather than leaving the category entirely to third parties.
Where it still breaks: anything that needs real judgment about whether a job is actually an emergency (a "no heat" call in January reads very differently from a "no heat" call in July, and the agent has to be configured for that rather than assume it), and the fact that the oft-cited missed-call and lost-revenue numbers for this industry trace back to a single vendor's benchmark (Invoca), not an independent study, so treat them as directional rather than exact.
AI capabilities involved
Real-time speech-to-text on the live call
Emergency-vs-routine triage and booking via tool calls
Membership and maintenance-agreement renewal outbound calling
Dispatch ETA and on-the-way follow-up calls
Who uses this
- Single-truck and small multi-truck HVAC, plumbing and electrical operators losing after-hours calls to voicemail
- Businesses already on ServiceTitan, Housecall Pro, Jobber or FieldRoutes wanting phone coverage that writes into the same dispatch board
- Companies running membership or maintenance-agreement renewal outreach that needs a documented consent trail
- Seasonal businesses with call spikes during extreme weather that overwhelm a small office staff
What the home services receptionist handles (and what it shouldn't)
| Task | Agent handles it? | Systems it touches | Human role |
|---|---|---|---|
| Emergency vs routine call triage | Fully | Dispatch board (ServiceTitan, Housecall Pro) | Reviews borderline urgency calls |
| Booking a service call | Fully | Dispatch board | None for standard slots |
| After-hours and weekend answering | Fully | Telephony, dispatch board | None |
| Membership and maintenance-agreement renewal outbound | With approval | Dispatch board, outbound dialer | Reviews opt-outs and consent records |
| Estimate follow-up calls | With approval | CRM | Confirms before quoting a firm price |
| Dispatch ETA and on-the-way calls | Fully | Dispatch board, SMS | None |
| Complex or custom-quote calls | No | None | Office staff or a technician calls back |
| Payment collection over the phone | With approval | Payment link via SMS, not the call itself | Confirms payment landed |
Off-the-shelf AI agents on the market
Real products you can sign up for today — with current 2026 pricing, honest pros and cons.
ServiceTitan AI Virtual Agents (Contact Center Pro)
ServiceTitan shops wanting the lowest-integration-risk option
Not applicable; part of a paid ServiceTitan contract
Included as part of the Contact Center Pro add-on; not priced standalone and scoped inside your broader ServiceTitan contract
Outbound virtual agents available on ServiceTitan Max
Pros
- +Writes directly into the same dispatch board you already use, since it's the same platform
- +No separate integration project, because the vendor running your dispatch also runs the AI
- +Backed by the system of record itself, so it won't disappear out from under your workflow
Cons
- −Only available if you're already a ServiceTitan customer
- −Pricing isn't broken out separately from your broader ServiceTitan contract, making the AI cost hard to isolate
Avoca
ServiceTitan shops wanting a purpose-built voice layer beyond ServiceTitan's own native agents
None; demo only
Contact sales; third-party estimates put it at roughly $1,000-$3,500/mo, unconfirmed on the vendor's own site
Raised $125M+ at a $1B valuation in April 2026, co-led by Meritech and General Catalyst, reportedly serving 800-1,000+ operators
Pros
- +Built natively on top of ServiceTitan rather than as a bolt-on calendar sync
- +Backed by a large, recent raise that lowers the risk of the vendor disappearing mid-contract
- +Vendor-reported traction (on track to book $1B in jobs in 2026) suggests real adoption even as a vendor claim
Cons
- −No published self-serve pricing; every figure in the market is a third-party estimate
- −Built specifically for ServiceTitan shops, a poor fit if you run Housecall Pro or Jobber instead
Sameday
Operators on Housecall Pro, Jobber or FieldRoutes who want published pricing instead of a sales call
None
Launch $449/mo (500 minutes); Scale $789/mo (1,000 minutes); enterprise tiers past 22,500 minutes
Enterprise tiers for high-volume multi-location operators
Pros
- +Integrates with ServiceTitan, Housecall Pro, Jobber, FieldRoutes and Service Fusion, covering most of the trade-software landscape
- +Published, predictable minute-bucket pricing instead of a sales-only quote
- +Minute buckets are simpler to forecast against your call logs than per-call billing
Cons
- −Exceeding your minute bucket requires an upgrade, not a small overage charge
- −Doesn't build natively on any one dispatch platform the way Avoca does for ServiceTitan
Goodcall
Solo operators and very small trades businesses without a dedicated dispatch platform yet
No free tier
Starter $79/mo (100 unique callers, unlimited minutes); Growth $129/mo (500 callers); Scale $299/mo (2,000 callers)
Custom enterprise tier for higher caller volume
Pros
- +Billed per unique caller with unlimited minutes, good for long diagnostic calls
- +Works for solo operators not yet on any dedicated dispatch platform
- +Overage drops as low as $0.15/extra caller on the Scale plan
Cons
- −No native ServiceTitan, Housecall Pro or Jobber integration
- −Bookings land in a generic calendar rather than a real dispatch board
The AI stack
A home services receptionist needs the same voice and telephony layers any phone agent does, plus a working integration into whichever dispatch board actually runs your trucks.
Voice (speech-to-text + text-to-speech)
Converts the caller's speech to text and the agent's reply back to speech
ElevenLabs Agents
$0.08/min on every self-serve plan; $0.16/min burst over the concurrency capCustom builds prioritizing voice quality
Bland AI (bundled)
$0.11-$0.14/min all-in, STT/LLM/TTS included in one rateTeams that want one vendor instead of three
Our pick: ElevenLabs plus a chained model for a custom build, or a bundled vendor like Bland for a single invoice.
Telephony
Carries the call and connects it to your voice pipeline
Vapi's built-in telephony
Free SIP/number hosting; Twilio pass-through at $0.008/min inboundDeveloper builds on Vapi
Twilio direct
$0.008/min inbound, $0.014/min outboundFully custom builds
Our pick: Let your orchestration platform handle telephony pass-through rather than managing Twilio directly.
Reasoning model
Understands the caller's intent, decides urgency, and books through the dispatch board
Claude Sonnet 5.5
$2/M input, $10/M outputThe main call-handling loop
Claude Haiku 5.5
$0.10/M input tokens (up to 100K), $0.50/M outputRoutine confirmation and reminder calls
Our pick: Sonnet 5.5 for triage and booking, with Haiku 5.5 handling routine ETA and confirmation calls to cut cost.
Dispatch integration
Books a real job on the real dispatch board instead of a generic calendar
Direct API (ServiceTitan, Housecall Pro, Jobber, FieldRoutes)
Developer time plus whatever API access the dispatch vendor chargesCustom builds
A vendor's existing integration (Sameday, Avoca)
Included in the vendor's monthly feeBuy-SaaS
Our pick: Use a vendor's existing dispatch integration unless you run a platform neither Sameday nor Avoca supports.
Observability & evals
Reviews call transcripts for triage mistakes and missed bookings
Vendor call logs and transcripts
Included on most SaaS tiersBuy-SaaS
Langfuse
Free up to 50K units/mo, then $29+/moCustom builds
Our pick: Review a weekly sample of real calls, paying extra attention to any call flagged as routine that later turned out to be an emergency.
Reference architecture
The hardest engineering problem is urgency triage in context: the same words ("no heat", "no AC") mean something very different depending on the season and the caller's situation (an elderly resident in a heat wave versus a routine maintenance question), so the triage logic needs to weigh context, not just keywords, and it should err toward flagging as urgent rather than under-triaging.
Call connects
Telephony (Twilio/Telnyx or vendor-native)The caller's phone call routes to the receptionist's number
Speech-to-text
STT engineThe caller's audio is transcribed to text in real time as they speak
Urgency triage
Reasoning modelThe model weighs the stated issue against season and context to flag emergency versus routine
Dispatch tool call
ServiceTitan/Housecall Pro/Jobber integrationA booking, with the correct urgency flag, writes directly into the real dispatch board
Text-to-speech
TTS engineThe agent's reply and an ETA estimate are converted to speech and streamed to the caller
Transfer or escalation
Call-transfer logicGenuinely ambiguous or high-value custom-quote calls get warm-transferred to office staff
Call logging and eval
Observability toolThe recording, transcript and dispatch-write confirmation are logged for a weekly review
Estimated cost per request
~$0.08-$0.15 per minute of call on a self-assembled stack; $449-$789/mo by minute bucket on packaged SaaS tiers
Cost calculator
Drag the sliders to model your actual usage. The numbers update in real time so you can stress-test economics before writing a single line of code.
Estimate your monthly cost under either path: a packaged minute-bucket vendor, or a DIY voice stack.
Estimated monthly cost
$537
≈ $6,444 per year
Calculator notes
- These two lines model two different paths, not costs that stack: the fixed line is Sameday's Launch tier, and the per-call line is a DIY Vapi/ElevenLabs stack.
- Avoca and ServiceTitan's native AI Virtual Agents aren't modeled here since neither publishes standalone per-call or per-minute pricing.
- A custom agency build replaces either line with your own voice-stack bill, typically $100-$400/mo at this volume, plus the upfront build cost.
- None of these figures include the revenue from a missed call that went to a competitor instead of voicemail.
Build it yourself with vibe-coding tools
Vapi lets a technical owner-operator assemble a working home-services receptionist from a phone number, a system prompt and a dispatch-board tool call, without paying a packaged vendor's minute-bucket fee.
Time to MVP
A few days for a basic version, 2-4 weeks to get dispatch write-back and triage rules production-ready
Total cost to MVP
$0-$100 (a Vapi Core plan at $29/mo plus usage, or pay-as-you-go with no monthly fee)
You'll need
Starter prompt
Create a new Assistant in Vapi using Claude Sonnet 5.5 as the model, ElevenLabs for text-to-speech, and Deepgram for transcription. Set the first message to: 'Thanks for calling {Business Name}, this is their AI assistant, how can I help you today?' Give the assistant this system prompt: 'You are the phone receptionist for {Business Name}, an HVAC and home services company in {city}. Determine whether the caller's issue is an emergency (no heat in cold weather, no AC in a heat wave, a water leak, a gas smell, no power) or routine (maintenance, a non-urgent repair, a quote request), weighing the current season and the caller's description rather than just keywords, and err toward emergency if you're unsure. Use the check_availability and book_job tools, which write into our dispatch board, to schedule the right technician with the correct urgency flag. For any custom quote or multi-system job, say: "I'll have someone call you back with a firm quote," and use the transfer_call tool to notify the office instead of estimating a price yourself. Never claim to be human if asked directly. This call is recorded for quality purposes, consistent with the greeting already given.' Add three tools: check_availability and book_job pointed at your dispatch board's API, and transfer_call pointed at your office line. Turn on call recording, and log every call and its urgency classification to a sheet or database for a weekly review.
Paste this into Vapi
Follow-up prompts (run in order)
- 1
Add a membership-renewal outbound flow that calls customers whose maintenance agreement is expiring
- 2
Add an ETA tool that texts the caller when the technician is on the way
- 3
Add a Spanish-language path that detects the caller's language and switches voices
- 4
Add a nightly report listing every call flagged as an emergency and whether it was dispatched same-day
Expected output
A phone number that correctly triages emergency versus routine calls, books real jobs directly into your dispatch board, and routes custom-quote calls to office staff for a callback.
Known gotchas
- !Urgency triage needs seasonal context, not just keyword matching, or the agent will under- or over-triage calls
- !Dispatch-board APIs vary enormously, so the booking integration is often the slowest part of the build, not the voice layer
- !A booking that silently fails to write into the dispatch board is worse than no AI receptionist at all, since nobody notices until a truck doesn't show up
- !Vendor-advertised latency figures are often optimistic; a laggy voice on a noisy job-site call loses customers fast
- !Forgetting the AI-disclosure line at call start is both a compliance problem and a common caller complaint
- !TCPA consent requirements apply to any outbound membership-renewal or follow-up call, not just inbound answering
Compliance & risk reality check
A home services receptionist carries lighter compliance weight than a healthcare or legal receptionist, but outbound calling and payment handling still create real exposure.
TCPA and FCC AI-voice rules for outbound calls
The FCC's February 2024 Declaratory Ruling (24-17) holds that AI-generated voices count as an "artificial voice" under the TCPA, requiring prior consent for outbound calls; this mainly hits membership-renewal and follow-up calls rather than inbound answering.
Mitigation: Capture written consent naming AI-generated calls for any membership-renewal or outbound follow-up campaign, and disclose the AI at the start of every outbound call.
Call recording and wiretap law
Ambriz v. Google and Taylor v. ConverseNow both survived motions to dismiss in 2025 on the theory that an AI vendor capable of using call recordings to train its own models can be an unlawful eavesdropper.
Mitigation: Play a recording-and-AI notice at the start of every call, and put a no-training clause in every vendor contract covering your call data.
AI disclosure on the call
California's BOT Act, Utah's AI Policy Act and the EU's Article 50 transparency rule all push toward telling callers they're talking to AI.
Mitigation: State "this is their AI assistant" in the greeting's first line, and never claim to be human if a caller asks directly.
PCI DSS if the agent takes a deposit by phone
If a caller speaks a card number to pay a deposit for a job, the recording, transcript and underlying model calls all enter cardholder-data scope under PCI DSS 4.0.1, fully mandatory since March 31, 2025.
Mitigation: Use DTMF keypad capture or a pay-by-SMS link so the card number never reaches the AI, the recording or the transcript.
Build vs buy: the real math
3-6 weeks for a RapidDev build; a few days for a basic Vapi version
Custom build time
$13K-$25K for an agency build (up to $35K for multi-system or compliance-heavy work); $0-$100/mo for DIY
One-time investment
Roughly two or more trucks running on a dispatch system a packaged vendor doesn't fully support, where a custom integration's cost gets amortized across more call volume
Breakeven vs buying
A single-truck operator paying $449/mo for Sameday's Launch tier is paying $5,388/year, well under a $13K-$25K custom build's first-year amortized cost of $1,083-$2,083/mo. The math only shifts once you're running multiple trucks or locations on a dispatch platform neither Sameday nor Avoca supports well, or once your triage logic needs rules (a specific service-area map, a specific membership-tier pricing rule) that go beyond what a packaged vendor's workflow builder exposes. Below that point, buying a packaged tool is close to always the right call for a small home services operator.
Skip the DIY — RapidDev builds the production version
A Lovable MVP gets you a demo. Production needs auth that doesn't leak data, AI calls that don't bankrupt you, observability when models drift, and code you can audit. That's what we ship through our AI app development service.
Discovery call (free)
30 minWe map your exact home services AI receptionist use case: who uses it, target volume, AI model choice, integrations, compliance scope. You get a detailed scope document and fixed-price quote within 48 hours.
AI-accelerated build
3-6 weeks for a RapidDev build; a few days for a basic Vapi versionOur engineers use Claude Code, Lovable, and custom tooling to ship 3–5x faster than agencies. You see weekly progress in a staging environment — not a black box.
Launch + handoff
1 weekWe deploy to your infrastructure, transfer the GitHub repo, set up CI/CD and monitoring, and train your team. You own 100% of the source code, prompts, and model configurations.
What you get
Timeline
3-6 weeks for a RapidDev build; a few days for a basic Vapi version
Investment
$13K-$25K for an agency build (up to $35K for multi-system or compliance-heavy work); $0-$100/mo for DIY
vs SaaS
ROI in Roughly two or more trucks running on a dispatch system a packaged vendor doesn't fully support, where a custom integration's cost gets amortized across more call volume
30-min call. Fixed-price quote within 48 hours. No commitment.
Frequently asked questions
How much does an AI receptionist for HVAC or home services cost?
Sameday's published tiers run $449-$789/mo by minute bucket, and ServiceTitan's native AI Virtual Agents are folded into your existing Contact Center Pro add-on. A custom RapidDev build runs $13K-$25K upfront (up to $35K for complex, multi-system work), plus $100-$400/mo in ongoing costs. A DIY build on Vapi costs roughly $0.08-$0.13 per minute of call time.
How long does it take to set up?
A packaged tool like Sameday or ServiceTitan's native agent can go live within about two weeks of connecting your number and dispatch login. A custom agency build takes 3-6 weeks to wire into your dispatch board and triage rules. A basic DIY Vapi version can work within a few days, though production-ready triage and dispatch write-back takes 2-4 weeks.
Can RapidDev build this for us?
Yes. RapidDev has shipped 600+ apps and offers a free 30-minute consultation to scope your dispatch system, call volume and compliance needs before you commit to anything.
Does it work with ServiceTitan, Housecall Pro or Jobber?
ServiceTitan's native AI Virtual Agents only exist inside a ServiceTitan contract, and Sameday integrates with ServiceTitan, Housecall Pro, Jobber, FieldRoutes and Service Fusion. A custom build can connect to any of these through the platform's API.
How does it decide if a call is an emergency?
A well-built agent weighs the stated issue against season and context, not just keywords, since "no heat" means something very different in January than in July, and it should err toward flagging a call as urgent when it's genuinely unsure rather than under-triaging it.
Will callers know they're talking to AI?
They should hear it in the first line of the greeting. Stating "this is their AI assistant" satisfies disclosure expectations under California's BOT Act, Utah's AI Policy Act and the EU's Article 50 transparency rule at once.
Is it true that most missed calls cost $1,200 in lost revenue?
That figure is widely cited but traces back to a single vendor's benchmark (Invoca), not an independent study we could verify against a primary report. Treat it as a directional industry estimate rather than an exact number for your own business.
Want the production version?
- Delivered in 3-6 weeks for a RapidDev build; a few days for a basic Vapi version
- You own 100% of the code
- AI cost monitoring built in
30-min call. No commitment.
