You're under a house with your hands full and the phone rings again. An AI receptionist promises to pick up every call, chat to the customer and take the details for you. For a sole trader that sounds brilliant. Before you sign up, it's worth knowing what these tools actually do, how customers feel about them, and what to check.
What an AI receptionist actually does
An AI receptionist is a voice system that answers your business number when you can't. Most work the same basic way. Calls you don't pick up get diverted to the service, a synthetic voice greets the caller, and software listens to what they say and replies.
- Answers missed or after-hours calls with a greeting in your business name.
- Asks the caller for their name, number, address and what the job is.
- Answers simple questions you've given it, like your service area or hours.
- Sends you a summary or transcript by text or email.
- Some can offer booking times from a calendar, if you connect one.
Done well, that means fewer enquiries lost to voicemail. Done badly, it means a customer arguing with a robot about a burst pipe. If you're weighing up the cost of the calls you miss, start with what a missed call really costs a tradie.
How customers feel about talking to AI
This is the part the sales pages skip. Plenty of customers are wary of AI on the other end of the line.
In a global Gartner survey of 5,728 customers, run in December 2023, 64% said they would prefer that companies didn't use AI in their customer service, and 53% said they would consider switching to a competitor if they found out a company was going to use it. Gartner found the top worry was that it would get harder to reach a person.
Closer to home, the OAIC found Australians aged 50+ are more likely to feel uncomfortable with customer service chatbots, at 49% against 27% of those aged 18 to 49. Older homeowners are a big share of the people booking renovations, repairs and maintenance.
Be upfront that it's AI
People want to know. As well as the OAIC result above, Salesforce's global survey, which included Australian consumers, found nearly 75% of consumers want to know if they're communicating with an AI agent. Pretending your AI is a person called Sharon is a quick way to annoy a customer once they work it out.
Give callers a way to a human
The same Salesforce research found 45% of consumers are more likely to use an AI agent if there's a clear escalation path. For a tradie, that just means the caller hears when they'll get a real call back, and that it actually happens.
What to check before you switch one on
- 1Ring it yourself, as a cranky customer would. Mumble, talk over it, ask an odd question. See how it copes.
- 2Check the greeting says it's an automated assistant, in plain words.
- 3Make sure urgent jobs get flagged to you straight away, not buried in a daily summary.
- 4Find out where call recordings and transcripts are stored, who can see them and how long they're kept.
- 5Check it can't promise prices or times you haven't agreed to. A receptionist that quotes a figure on the phone has made a commitment for you.
- 6Read the contract for lock-in periods and per-minute charges.
- 7Test it for a fortnight before you put the number on the side of the ute.
If you'd rather keep the phone personal, there are simpler fixes for busy days, like a better voicemail greeting and a quick text back. See handling enquiries when you're flat out on the tools.
The call is only the start
Here's the catch with any receptionist, human or AI. Taking the details doesn't win the job. The customer who rang you has often rung two other tradies too. What decides it is what happens next: who gets a clear quote back first, who turns up when they said, and who follows up.
Plenty of tradies capture the enquiry fine and then lose it in the gap between the site visit and the quote landing in the customer's inbox days later.
Where Agentize fits (and where it doesn't)
To be clear, Agentize is not a phone receptionist. It doesn't answer calls. It handles what happens after the enquiry, the part that usually slips.
- Quote: record a short voice note and a couple of photos on site, and get an itemised, GST-sorted quote priced only from your own rate card. Unpriced lines are left blank and flagged, never guessed.
- Booking: the job goes into your week, and the customer gets a reminder email the day before.
- Invoice: the finished job becomes an invoice that gets chased until it's paid, and it connects to Xero.
- Follow-up: quotes that go quiet get a drafted nudge in your voice, and after a paid job it drafts a referral ask.
When a customer texts back, their message lands in the app with a reply drafted. Nothing reaches a customer until you tap send. So you could use an AI receptionist to catch the call and Agentize to turn it into a quote, or skip the receptionist and focus on getting quotes out fast.
The verdict
An AI receptionist can be a sensible way to stop enquiries dying in voicemail, as long as it tells callers it's AI, gets urgent jobs to you fast and never commits you to a price. Just don't mistake answering the phone for winning the work. The job is won after the call, by the tradie who gets back to the customer quickest with a clear, honest quote.