AI reception for a service business that actually chases
AI reception for a service business that actually chases
An enquiry comes in on a Tuesday afternoon. Web form, from the ad you're running. You're under a house, or with a patient, or driving to the next job. It sits in your inbox with eleven other things until you get to it that night, if you remember which folder it landed in.
That's reception at most service businesses. Not a phone problem, an attention problem. The enquiry arrived fine. Nobody got to it in time to matter.
Here's what an AI operator does differently for reception specifically, how it's not the same thing as a phone bot or a human on the desk, and what it actually costs to run.
Reception is not "answer the phone"
Say "AI receptionist" to most owners and they picture a voice on the phone. That's one channel, and it's not even where most of your leads come in.
Your enquiries arrive by email, by the form on your website, by WhatsApp because that's what your customers already have open, and by SMS from someone who saw your ad and typed a question. A phone bot handles none of that. It sits on one line, waiting for a call that increasingly isn't how people reach you.
The standard Azlar build doesn't answer live phone calls either, and we're not going to pretend otherwise. What it does is read every enquiry that lands by email, form, WhatsApp or SMS, work out what the person's actually asking for, and reply the way you would if you weren't three jobs behind.
That's the difference. A phone bot answers a channel. Reception, done properly, covers where your customers actually message you and does something with what they send.
What it does with an enquiry
Walk through one. Someone fills out your web form asking for a quote on a job.
The operator reads it, checks the CRM to see if they've enquired before, and replies with the questions you'd ask yourself: where's the job, what do they need, when. It logs the enquiry against their record and notes where it came from.
Once there's enough detail, it drafts the quote from your price list, in your wording, and sends it. It books the job into your calendar if they're ready to go. Then it sets a follow-up date, because most quotes don't get answered the same day, and most owners don't remember to chase them either.
If the enquiry's anything odd, a complaint, a request outside what you do, a price question you'd want to handle personally, it flags you on Telegram before it replies. That's the approval front door. Sensitive or unusual things wait for your yes.
The two-week rollout
Nobody hands this the keys on day one. For the first two weeks it runs supervised: every reply it wants to send comes to you on Telegram first. You approve it, edit it, or knock it back.
This is where it learns your tone, your prices, and your rules for what needs a human. Most owners stop approving the routine replies by the end of week two, because by then it's just repeating what they already signed off on. Approval on quotes and anything client-facing outside the routine stays with you for as long as you want it there.
Worried it'll say something dumb to a customer? That's the entire point of the two weeks. Nothing goes out unseen until you decide it can.
Phone bot vs human receptionist vs Azlar reception
| Phone bot | Human receptionist | Azlar reception | |
|---|---|---|---|
| Channels covered | Phone calls only | Phone, desk, walk-ins | Email, web form, WhatsApp, SMS |
| Hours | 24/7 on that one channel | About 38 hours a week | 24/7 |
| Sends the quote | No | Yes, if trained on your pricing | Yes, from your price list |
| Chases a quote that's gone quiet | No | If they remember | Yes, on schedule, until booked or cold |
| Updates the CRM | Rarely | If it's part of the job | Every time |
| Cost | From $25 to $349 a month for phone-only bots | Roughly $60,000 to $70,000 a year plus super and on-costs | Setup build quoted on the call, then Partner $800/mo or Managed $2,000/mo |
A phone bot is the right buy if your only leak is missed calls and you quote on the spot anyway. A human on the desk still wins for the upset customer, the walk-in, the read on a difficult conversation. Where both of them stop, at the quote that needs sending and the follow-up that needs chasing, is where reception through an operator keeps going.
What it costs
Every plan starts with a one-off setup build. That gets quoted after the 20-minute call, because what you need connected (your CRM, your inbox, WhatsApp, your calendar) is different for a clinic than it is for a plumber, and we're not going to make up a number that doesn't reflect your setup.
After that, it's a monthly plan. Partner is $800 a month for light support and small tweaks once it's running. Managed is $2,000 a month, which is what most clients land on, for ongoing operation, tuning, and new workflows added as your business changes. Full Ops is $3,800 a month, built for teams already past $100,000 a year in admin wages who want deeper hands-on ops.
Put it against a hire. A receptionist costs somewhere around $60,000 to $70,000 a year once you add super and the rest of the on-costs, for about a fifth of the week at the desk. Managed, at $24,000 a year, runs the reception side around the clock and also chases the quote after they've gone home. That's the comparison worth doing before you commit to either one.
What it doesn't do
Worth being straight about this so you're not disappointed later. It doesn't answer the phone in the standard build. If live calls are your main channel and your customers won't message any other way, start with a phone bot or a human, or talk to us about what a phone-capable build would take.
It doesn't handle the customer who needs a human read, the one who's upset or scared or asking for something outside your scope. Those get flagged, not answered blind.
It doesn't set your prices or make judgement calls on discounts. It uses your price list and asks you when something's outside it.
We run this on our own reception
We built this for Intelligent Living Solutions, our own Canberra smart-home company. Two founders, no admin staff. Enquiries come in from Meta ads, get synced into the CRM hourly, and the operator works them the same way described above: reads, replies, logs, chases. As of 17 August 2026 there were 337 leads through the system since December 2025, 267 of them from Meta ads. That's the only case study we tell, because it's the only one we've actually run.
Related reading
- AI receptionist for small business
- AI receptionist vs human vs voicemail
- What an AI operator actually does for a small business
- What a done-for-you AI agent costs in Australia
- 9 questions before hiring an AI automation agency
- Case study: the company we built it for first
- Plans and pricing
- Book the call
Book the 20-minute call and we'll tell you what reception would look like connected to your actual inbox and CRM.
Book the 20-minute call
We look at your setup, name the first job worth handing over, and tell you what it would save you. If it's not worth it, we'll say so.
Book the Call