Notes
- AI receptionist cost for UK dental practices falls into two models: flat monthly (£397–£1,297/mo, fully predictable) or pay-as-you-go per-minute (advertised from £10–£49/mo, but real invoices hit £300–£700+ once call minutes scale).
- The headline "£19/month" rates are PAYG traps. A single 4-minute implant enquiry can cost £2–£6 in raw API and telephony fees; a busy practice burns through bundled minutes by week two.
- One missed new-patient call costs a private practice £500–£1,200 in lifetime value. 68% of patients who can't get through on the first try call a competitor instead.
- A flat-fee AI receptionist like Ava (£397–£1,297/mo, no per-call charges, no setup fee) pays for itself by recovering one to two missed consultations a month.
- The real cost question isn't "how cheap is the bot" — it's "what is the practice losing every morning the phone rings out."
What an AI Receptionist Actually Costs a UK Dental Practice in 2026
A UK dental AI receptionist costs either a flat £397–£1,297 per month (predictable, all-inclusive) or runs on pay-as-you-go per-minute billing that starts cheap and scales unpredictably to £300–£700+ once real call volume hits. The advertised "£10–£49/month" figures almost always belong to the second model. They are not the price you pay.
This is the single biggest source of confusion for practice owners, and the ranking pages make it worse. Most articles list a tidy table of three plans starting at an unrealistic minimum and call it a day. They never explain that the cheap tier is a per-minute meter dressed up as a subscription. A dental triage call is not 40 seconds. A patient spelling out "Invisalign," describing a cracked crown, or asking about implant consultation costs runs three to five minutes. On a PAYG stack stitched together from Twilio trunking, OpenAI Realtime API tokens, and a Make.com or n8n automation layer, those minutes meter out at roughly £0.50–£1.50 each before you've booked a single appointment.
Run the maths on a three-surgery practice taking 600 inbound calls a month at an average of three minutes. That's 1,800 billable minutes. Even at the low end, you're looking at a variable invoice that swings between £200 and £600 depending on how chatty your patients are that month. You cannot forecast it, you cannot budget it, and you find out the real number after the fact. That unpredictability is precisely what practice managers dread on the forums.
The flat-fee model exists to kill that anxiety. Ava runs at a fixed monthly rate across three tiers — Essential at £397/mo, Growth at £697/mo, and Premium at £1,297/mo — with no setup fee, no per-call charge, and no per-minute meter. A 90-second appointment booking and a six-minute anxious-patient emergency triage cost exactly the same: nothing extra. The price you see is the price you pay, every month, regardless of how busy the morning surge gets.
What One Missed New-Patient Call Costs a Practice
A single missed new-patient call costs a UK private dental practice £500–£1,200 in lost lifetime value for routine and cosmetic enquiries, and considerably more for implant or full-mouth cases. With 68% of patients dialling a competitor after one failed attempt, the cost of a ringing phone compounds fast — most practices simply never see the invoice.
Here's the part the cheap-bot comparison articles never put in front of you. The relevant number isn't the monthly software fee. It's the revenue walking out the door every morning between 8:30 and 10:30, when the phones light up and the front desk is physically checking patients in.
The cost-of-missed-call table
| Scenario | Typical patient LTV | Calls missed/month (peak surge) | Monthly LTV at risk |
|---|---|---|---|
| Routine new patient (hygiene + check-ups) | £500–£700 | 8 | £4,000–£5,600 |
| Cosmetic consult (whitening, veneers) | £900–£1,200 | 4 | £3,600–£4,800 |
| Implant / full-mouth enquiry | £3,000–£12,000 | 2 | £6,000–£24,000 |
Figures are modelled scenarios based on UK private dental LTV ranges and typical morning-surge abandonment rates (~32%). Actual numbers vary by practice, location, and case mix.
Even the conservative row dwarfs every pricing tier on the market. A practice losing four cosmetic consults a month to voicemail is leaking £3,600–£4,800 in lifetime value to recover a £697/month receptionist that answers every one of those calls. That is the entire argument. The bot doesn't compete on being cheap. It competes on no longer setting fire to four-figure consultations every week.
This is also why dental is the right vertical for AI front-desk in the first place. As one practitioner put it bluntly in a community thread: clinic retention is sticky. Patients come back to the same dentist for years, so the receptionist pays for itself on repeat visits, not just the first booking. Restaurants are low-margin and high-churn. A dental list is the opposite — high-value, high-retention, and brutally unforgiving of a missed first impression.
AI Receptionist vs Human Answering Service: The Real Comparison
A human answering service for a UK dental practice typically runs £3,000–£3,500/month for full coverage and bills per call on top, while a flat-fee AI receptionist runs £397–£1,297/month with unlimited calls. The AI also handles concurrent calls — the answering service cannot — which is the entire problem during the morning surge.
The per-call answering service has a structural flaw that no amount of staffing fixes. When 14 calls land at 8:32 on a Monday after a weekend closure, a human service answers them in a queue. Patients eleven through fourteen hear hold music or hang up. An AI receptionist answers all 14 simultaneously, books the routine appointments straight into the calendar, and routes the two genuine clinical emergencies directly to the practice principal's mobile. Same minute. No queue.
| Human answering service | Flat-fee AI receptionist (Ava) | |
|---|---|---|
| Monthly cost | £3,000–£3,500+ | £397–£1,297 |
| Per-call charges | Yes, on top | None |
| Concurrent calls | No (queued) | Yes (unlimited) |
| Out-of-hours | Limited / extra cost | 24/7, always on |
| Books into your calendar | Sometimes | Yes, real-time |
| Identifies as AI | N/A | Always |
The Build-vs-Buy Cost Bridge Nobody Explains
The genuine cost decision for a UK practice isn't "which bot is cheapest." It's build your own stack (£47–£100/mo plus engineering time and breakage risk) versus buy a managed verticalised service (£397–£1,297/mo, done for you). The break-even point is set by one metric almost no practice tracks: average monthly inbound call minutes.
You can build a sovereign AI receptionist yourself. The community templates are real: Twilio for the phone number and trunking, OpenAI Realtime API for the voice, and Make.com or n8n to orchestrate the booking logic. On paper it costs £47–£100/month. That number is honest about the software and dishonest about everything else.
What the £47/month build doesn't price in:
- Engineering setup. Days of configuration, prompt tuning, and calendar API plumbing — before it answers a single call correctly.
- The double-booking bug. Calendar and practice-management-software APIs sync with a delay. Two patients calling at once get offered the identical slot unless you build a state lock into the orchestration layer. The community's fix is a 30-second structural lock. You have to build and maintain it.
- UK compliance, by hand. The mandatory AI-disclosure line at second zero ("This call is handled by an AI assistant and may be recorded…"), call-recording consent, UK GDPR data residency, PECR. US-focused tools do not configure this for you.
- Breakage. When OpenAI ships a model update or Twilio changes a rate, your stack breaks at 8:30 on a Monday and you are the on-call engineer.
So the build-vs-buy line isn't really about money. It's about whether the founder's time is better spent maintaining telephony middleware or growing the practice. For a single-chair clinic with low call volume and an in-house technical hand, a self-built stack can work. For a busy multi-surgery practice where a broken phone line on Monday morning costs four cosmetic consults, a managed flat-fee service that ships compliant in 48 hours is not the expensive option. It's the cheap one.
That's where Ava sits deliberately: UK-only, ICO registered, UK GDPR and PECR compliant, with a signed DPA for every practice and the AI-disclosure handled at second zero by default. No per-minute meter, no engineering rota, no compliance homework. The double-booking problem and the morning surge are solved before go-live, not discovered after.
Frequently Asked Questions
How much does an AI receptionist cost for a UK dental practice?
It depends entirely on the billing model. Flat-fee managed services run £397–£1,297 per month with unlimited calls and no per-call charges. Pay-as-you-go services advertise from £10–£49/month but bill per minute, and real invoices for a busy practice land between £300 and £700+. The advertised low figures are almost always the PAYG model.
Are there hidden costs or per-minute charges?
On PAYG and self-built stacks, yes — minutes meter at roughly £0.50–£1.50 each, and longer triage calls run the bill up unpredictably. A flat-fee service like Ava has no per-call or per-minute charges and no setup fee. The monthly price is fixed regardless of call volume or call length.
Does an AI receptionist handle out-of-hours emergency triage?
Yes. A properly configured AI receptionist answers 24/7, identifies genuine clinical emergencies, and routes them to the on-call principal's mobile while logging routine callbacks. It does not provide clinical advice — Ava always identifies as AI and gives no clinical, legal, or financial advice — but it ensures no emergency call rings out overnight.
Is an AI receptionist GDPR compliant for patient calls?
It is only compliant if it's built for UK rules, and many US-focused tools are not. Compliance requires an AI-disclosure line at the start of the call, recording consent, UK data residency, and PECR adherence. Ava is ICO registered, UK GDPR and PECR compliant, and signs a DPA with every practice. The disclosure plays at second zero by default.
Can it integrate with dental practice management software like SOE or Dentrix?
Standard calendar integration (Google, Outlook) is near-universal. Deep integration with closed UK practice-management software such as Software of Excellence Exact or DenGro is more involved and often needs custom API work — this is exactly where cheap generic bots quietly fail. A managed service should confirm your specific PMS integration before go-live, not after.
Is it cheaper than a human answering service?
Substantially. A human answering service runs £3,000–£3,500+ per month and bills per call on top, and it can't answer concurrent calls — patients queue during the morning surge. A flat-fee AI receptionist at £397–£1,297/month answers every call simultaneously, with no per-call fees.
How quickly can it go live?
A managed UK service can typically go live within 48 hours, including the compliance layer and calendar integration. A self-built Twilio-plus-OpenAI stack takes considerably longer once you account for engineering setup, the double-booking state lock, and configuring UK compliance by hand.
The Ultimatum
Stop comparing AI receptionists on the headline monthly fee. That number is the least important one on the page, and on PAYG plans it's actively misleading. The two figures that decide the maths for a private dental practice are your average monthly inbound call minutes (which sets whether a per-minute model ruins you) and your lost lifetime value from missed morning-surge calls (which is almost always larger than any pricing tier).
Track your peak-hour abandoned-call rate for one week. If you're losing even two cosmetic consults a month to voicemail, a flat-fee AI receptionist is already cheaper than the status quo — you just weren't being sent the invoice for the calls you missed.
