Choosing a solution
Why an AI receptionist costs more than an answering service
The short answer
Because answering a call is the cheap part. The cost sits in knowing your service area, your catalog and every technician's real availability, then writing a confirmed booking into your calendar. A product that ends the call with a message avoids all of that work, and prices accordingly.
Key takeaways
- Receva is $199 to $599 a month plus a flat per-booked-job fee. That is above the flat-rate answering market, and we are not going to pretend otherwise.
- The gap pays for integration, not for a better voice. Voices are close to commoditised; live availability is not.
- Per-minute billing peaks in your busiest month. A per-booked-job fee peaks when you have the most new work, which is the bill you actually want.
- The per-booked-job fee falls at every tier — $10, then $8, then $6 — so scale makes each booking cheaper rather than dearer.
- If your missed calls cost a few hundred dollars a month, none of this is worth it. Say so and move on.
1. The gap, stated plainly
Receva starts at $199/mo plus $10 for each job it books. Pro is $349/mo plus $8, Growth is from $599/mo plus $6. One rate covers every booked job, urgent or routine. Those are the real numbers and they are published on the pricing section of our homepage rather than gated behind a demo.
Plenty of flat-rate AI answering products sit well below that. If you have been reading around this category, you have seen the range, and arriving at our pricing page after those is a jolt. This post exists because "you get more" is not an argument, and the honest version is more specific than that.
2. What actually costs money
Speech is close to a commodity now. The models that answer a phone and hold a natural conversation are available to everyone building in this space, and the difference between a good voice and a great one is not what separates a $50 product from a $300 one.
The expensive part is everything that has to be true before the system can safely say "I can get someone to you Thursday at 2."
- Knowing where you go. The caller's address has to be resolved and checked against your coverage — postal codes and drawn polygons, with overlap rules if you run more than one branch. Service area is a hard gate: a caller outside it is captured as a lead but never booked, and never billed.
- Knowing what you do. The job has to be matched to a service you actually offer, from a catalog that differs by trade.
- Knowing who is free. Availability has to come from the connected calendar or field service system, filtered by each technician's working hours, and be re-checked at the moment of booking so two calls cannot take the same slot.
- Writing it back. The confirmed booking has to land in the system your team actually looks at, and stay correct if the integration drops and reconnects.
- Being auditable afterwards. Every call keeps a recording, a transcript, the qualification result and the reasoning behind each decision. That is what makes a billing dispute a matter of checking the record rather than arguing about it, and it is not free to build or store.
A message-taking product needs none of that. It is a genuinely simpler thing, and it is reasonable that it costs less. The question is not which is cheaper. It is which one leaves you with work still to do.
3. Why the billing shape matters more than the number
Per-minute and per-call billing has a property that is easy to miss when comparing headline rates: it bills you most in the month you can least predict.
The first hard freeze, the week after a storm, the heatwave — call volume spikes, average call length goes up because callers are anxious, and the invoice lands at the end of the month. You cannot budget for it and you cannot control it, because the alternative to taking those calls is not taking the work.
Receva has no per-minute charge and no per-call charge on any plan, on calls in or calls out. Missed-call recovery calls out to people who rang while you were on a job, and that is included rather than metered — not a separate line item with its own limit.
4. Why there is a per-booked-job fee at all
Because a flat subscription alone would have to be priced for the heaviest user, and most businesses are not the heaviest user.
Splitting the price means the fixed part covers being available and the variable part tracks delivered work. A call is billable only when Receva books a real appointment, in your service area, for a service on your catalog, with a contactable homeowner, into a slot that exists on your calendar. All five have to hold. Out-of-area callers, spam, wrong numbers, existing-customer service queries, calls that never booked, cancelled bookings and duplicates of a job already billed are never billable — the full definition with every exclusion is published.
The fee also falls as you grow, from $10 to $8 to $6. That is deliberate: the more work the system books you, the less each booking costs, which is the opposite of how per-minute pricing behaves. It does not move with urgency either -- an emergency and a filter change cost the same, so your worst week is never also your most expensive one per job.
The obvious objection is that a per-outcome fee gives us an incentive to book marginal work. Two things push against it. Every billed job links to its recording, transcript and qualification result, and every line sits on a weekly statement with a review window before that statement is charged. A line you dispute successfully is removed rather than refunded, so it is never charged at all. We would rather lose the fee than have you distrust the statement, because a statement you distrust ends the relationship anyway.
5. Working out your break-even
Take the monthly figure from the missed-call arithmetic — unanswered calls, times the share that were real enquiries, times your close rate, times your average job value.
Then be pessimistic about recovery. Assume you recover half of what you are currently missing, not all of it, because some callers have already booked elsewhere and some were never going to buy. Subtract the subscription and the per-lead fees on the bookings you would gain.
If that is still comfortably positive, the price is not the issue. If it is marginal, it is marginal, and you should not sign up on the theory that it will improve.
The "recover half" assumption is a deliberately conservative planning figure, not a measured recovery rate. We have not published a recovery rate because we have not established one with a methodology we would stand behind. Treat any vendor quoting a precise recovery percentage — including a future version of us — as owing you the sample size and the method.
6. When it is not worth it
Three situations where the answer is no, and we would rather you read them here than discover them in month two.
- Low volume, low ticket. If the arithmetic says missed calls cost you $400 a month, $199/mo plus fees cannot pay for itself. Re-record your greeting, set up forwarding, and text back everyone you missed. Those are free.
- Mostly existing customers. If your inbound is largely account questions, complaints and invoice chasing, you need a person, not a booking system. That is what answering services are good at.
- You need a system we do not support today. Google Calendar and Outlook are on every plan; Jobber and Housecall Pro from Pro. ServiceTitan support is built and being readied for release, but it is not something you can switch on today. There is no HubSpot integration and none planned.
The reason to put this in writing on our own site is straightforward. Under a self-serve model there is no salesperson to catch a bad fit, so the page has to do it — and a customer who cancels in month two costs everyone more than the one who never signed up.
7. Frequently asked questions
How much does Receva cost?
$199/mo on Starter, $349 on Pro and from $599 on Growth, plus one flat fee for each job it books you: $10, $8 or $6 by tier, whatever the urgency. There is no per-minute charge and no per-call charge on any plan, on calls in or calls out.
Why is that more than other AI answering services?
Because most of them stop at answering. The cost difference is the integration work: reading live availability from your calendar or field service system, filtering it by each technician's working hours, checking the caller's address against your service area, and writing a confirmed booking back. A product that takes a structured message does not carry that cost and should not charge for it.
Why charge per booked job on top of a subscription?
So the part of the bill that varies is tied to work delivered rather than to minutes consumed. A per-minute model bills you most in your busiest month. A per-booked-job model bills you only when an appointment lands on your calendar, and the fee falls at every tier as you grow, from $10 to $8 to $6.
Does the per-booked-job fee mean a surprise bill?
It should not. Fees appear as individual lines on a weekly statement in your dashboard, each traceable to the call that produced it, and every line is held for a review window before the statement is charged. A line you successfully dispute is removed rather than refunded, because it is never charged in the first place.
When is Receva not worth the money?
When the arithmetic does not support it. If your missed calls are costing you a few hundred dollars a month, a $199/mo subscription cannot pay for itself and you should fix your greeting and call forwarding instead. It is also the wrong tool if most of your calls are existing customers rather than new job enquiries.
Try it before you decide it is expensive
30 days or 60 call minutes, whichever runs out first. Lead fees are waived for the whole trial, so anything booked during it is free. If the arithmetic does not work for you, you will know inside a month.
Start free trial