The short version: AI receptionists for small business typically cost between £50 and £600 a month depending on call volume and whether you want text answers or a proper conversational voice agent. They're brilliant for capturing leads outside office hours and terrible at handling anyone who's upset, confused, or needs a real decision made on the spot. Test one for 30 days before you sign anything longer.
What an AI receptionist does (and doesn't)
An AI receptionist answers your business phone, books appointments into your calendar, texts back missed calls, and takes messages that get emailed or Slacked to you in real time. The good ones (Smith.ai, Ruby, Rosie by Squarespace, and the newer voice-agent players like Synthflow and Bland) use a mix of speech recognition and a scripted or GPT-powered conversation flow. Some sound close to human. Most don't, especially the moment a caller says something that isn't in the script.
What they don't do is replace judgement. If someone rings your boiler repair business at 11pm because their heating's died in a house with a newborn, an AI can log the details and text you. It cannot decide whether that's urgent enough to wake you up. That decision still sits with you or a human on-call rota.
Real pricing, not vague ranges
Prices change, so treat these as a snapshot for 2026, not gospel:
- Rosie by Squarespace: starts around £59 a month for a basic call-answering and message-taking plan, aimed squarely at solo traders and very small teams.
- Smith.ai: their live-agent-plus-AI hybrid plans start around £230 a month for roughly 30 calls, then charge per call after that. It gets expensive fast if your call volume is unpredictable.
- Ruby Receptionists: sits at the premium end, often £350 to £600 a month depending on minutes, and leans more human-in-the-loop than pure AI.
- Pure voice-agent platforms (Synthflow, Bland, Vapi-built agents): often priced per minute of call time, somewhere between 8p and 25p a minute once you add the underlying AI model costs, which suits high-volume callers better than low-volume ones.
Here's the bit reviews rarely mention: almost every one of these has a setup cost buried in onboarding, whether that's a "custom script build" fee or a chargeable hour with an implementation specialist. Ask for that number before you sign anything. I've seen it range from nothing to £400.
The client story that changed how I think about this
A client of mine, a small kitchen fitting company in Kent with three staff and a van full of tools, put an AI receptionist on their main line last year to catch calls while the owner was up ladders all day. Missed calls had been costing them jobs. Within the first month, bookings from inbound calls went up because nothing was going to voicemail anymore.
But three separate customers, when I later asked the owner how it was going, had mentioned they'd hung up and called a competitor because the AI voice "felt off" when they were trying to explain a specific problem with their existing kitchen units. That's the trade-off nobody puts in the sales deck. You gain volume, you lose a slice of the emotionally complicated calls, and for a trade business, complicated calls are often the highest-value ones.
The fix wasn't ditching the tool. It was routing anything that sounded even slightly complex straight to a human, and using the AI purely as a net for after-hours and simple booking calls. That split is the actual skill here, not the tool itself.
Do the maths before you commit to a contract
Work out your break-even point before you sign anything longer than a month. If an AI receptionist costs £250 a month and one missed call that turns into a job is worth £400 to you, you only need to capture one extra job a month to justify it. If your average job value is £40, the maths looks completely different. This is exactly the kind of number you should be running the same way you'd use an interactive calculator to work out whether a marketing spend pays for itself, don't guess, build the spreadsheet.
Also check what happens to unused minutes or calls. Most providers don't roll them over, which means a quiet month is money gone, not money banked.
Where an AI receptionist is worth it
- You get repeat, predictable questions: opening hours, pricing ranges, appointment availability, order status. Scripts handle these well.
- You lose business specifically because calls go to voicemail after hours or during busy periods.
- You run a single-person business and answering the phone physically stops you doing billable work.
- Your booking system (Calendly, Acuity, or similar) can plug straight into the AI so appointments land in your calendar without you touching anything.
Where it's a waste of money
- Your calls are mostly high-emotion or high-complexity: bereavement services, urgent trades, anything where the caller needs to feel heard before they'll book.
- Your call volume is low (under 20 a month) and doesn't justify a monthly fee, an answering service with a real human on a pay-per-call basis will be cheaper.
- You haven't got a clean FAQ or script written down. AI receptionists are only as good as the information you feed them, and building that from scratch eats the time you were trying to save.
How to trial one in 30 days
- Pick your single busiest missed-call window (evenings, weekends, lunch) and route only those calls to the AI first, rather than switching your whole line over on day one.
- Write down the ten most common questions your business gets asked by phone and hand those directly to the provider as your script.
- Set a clear escalation rule: anything mentioning a complaint, an emergency, or a price negotiation goes to a human, full stop.
- Track every call it handles for a month against actual bookings made, not calls answered, bookings made.
- Cancel or renegotiate before the trial rolls into an annual contract if the numbers don't stack up.
This is the same discipline that runs through good brand building generally, think about how deliberately Away built a consistent customer experience around their product before scaling it. An AI receptionist that's badly briefed is a worse first impression than no receptionist at all, because it answers instantly and gets it wrong instantly too.
The bit nobody wants to say out loud
A lot of small business owners buy an AI receptionist because they don't want to answer the phone, not because it's the best move for the business. That's a completely human, completely understandable reason. It's also worth naming honestly rather than dressing up as a growth strategy. If the real problem is that you hate cold calls interrupting deep work, an AI receptionist might just be an expensive way to avoid a boundary-setting conversation you could have for free, like turning your phone to voicemail during work blocks and calling back on your own schedule.
Alex Hormozi talks a lot about the value equation, what the customer gets versus what it costs them in time, effort, and risk. The same logic from his business lessons applies here in reverse: work out what an AI receptionist removes from your day, in hours, not vibes, before you decide it's worth the monthly fee.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
What good briefing looks like
The businesses getting real value from these tools have three things in common. First, a written script covering their actual most-asked questions, not a generic template. Second, a clear handoff rule for anything outside that script. Third, someone checking the call logs weekly for at least the first two months, not setting it up and forgetting it. That third point is the one people skip, and it's the one that determines whether the tool earns its keep.
Napoleon Hill's writing on definiteness of purpose gets quoted a lot in business circles, and it applies directly here. Decide exactly what you want the AI receptionist to achieve, more booked jobs, fewer missed enquiries, cover during specific hours, before you buy, rather than buying the tool and hoping a use case appears. His business lessons hold up surprisingly well against modern AI tool decisions for exactly that reason.
If working through this feels like more than you want to figure out alone, it's the kind of decision an AI consultant for small business can walk through with you in an hour, mapping your actual call patterns against the right tool rather than guessing from a pricing page.
Free resource: The AI Agent Handoff-to-Human Mini-Guide.
Frequently asked questions
Is an AI receptionist better than an answering service with real humans?
Neither is universally better. AI is cheaper and faster for simple, repeatable questions and works around the clock. A human answering service handles emotional or complicated calls better and costs more per call. Many small businesses get the best result using AI for after-hours and overflow, with a human answering service or the owner covering peak business hours.
How much does an AI receptionist cost for a small business in 2026?
Expect roughly £50 to £250 a month for basic call answering and message taking, and £250 to £600 a month for full appointment booking with higher call volumes, plus a possible one-off setup fee of £0 to £400. Per-minute voice agent platforms can work out cheaper if your call volume is low and steady.
Will customers know they're talking to an AI?
Often, yes, especially the moment they ask an unscripted question or the conversation needs any real judgement. Most jurisdictions, including parts of the US, are also moving toward requiring disclosure that a caller is speaking to an AI system, so build that expectation into your script rather than trying to hide it.
What's the biggest mistake small businesses make with AI receptionists?
Switching the entire phone line over on day one without a script built from real, common questions, and without a clear rule for when a call should go straight to a human. That combination is what causes lost jobs and frustrated callers, not the technology itself.
Related reading: I Called My Own Business Pretending to Be a Customer to Test an AI Receptionist and AI Phone Answering for Small Business: What It Really Costs and Where It Falls Down.
