Nobody starts a trade business because they enjoy quoting at nine at night.
The work itself is fine. It’s everything around it that piles up: the calls you missed on site, the quotes you owe people, the invoices nobody’s paid, the reminders you meant to send. Most of that is repetitive and rules-based, which is exactly what automates well.
Here are the four worth doing first, in the order I’d do them.
1. The calls you can’t answer
This is first because it’s the only one that costs you revenue directly rather than just time.
When you’re under a house or up a ladder, the phone rings out. Most callers don’t leave a message. They ring the next business on the list, and you never know it happened.
An AI phone system answers, works out what they need, and either books them in or takes proper details: what’s wrong, the address, how urgent, whether they own the place. You come back to something you can act on rather than a missed call notification.
Worth counting before you decide anything else. Five missed calls a week, at your average job value, and a third of them turning into work, is usually a bigger number than anything else on this list.
There’s more detail in the AI receptionist post, including what to test before buying one, and a separate breakdown of what one of these actually costs.
2. Turning site notes into a quote
The gap between looking at a job and sending the quote is where work gets lost. Not because you can’t price it, but because the writing-up happens at night, and by the time it goes out the customer has two other quotes.
This one automates well because the inputs are consistent. Your notes, your rates, your standard inclusions and exclusions. A system grounded in your previous quotes and your price list can produce a draft from a few lines of notes or a voice memo from the van, in your wording, with your terms attached.
You still check it and send it. But checking a draft takes two minutes and writing one from scratch takes twenty, and more importantly the two minutes can happen in the car rather than at nine at night.
Same-day quotes win work that next-week quotes don’t. That’s the whole return.
3. Chasing money
Every trade business has invoices sitting past due, and almost nobody chases them on a schedule, because chasing is uncomfortable and easy to postpone.
Automation is genuinely good at this, partly because it’s consistent and partly because it removes the awkwardness. A polite reminder at seven days, a firmer one at fourteen, an escalation at thirty. It goes out whether or not you felt like sending it.
Set the tone yourself and keep it courteous. The value isn’t in being aggressive, it’s in being reliable. Most late payment is disorganisation rather than refusal, and a well-timed reminder resolves the majority of it without anyone feeling chased.
This one usually needs no AI at all, incidentally. It’s ordinary automation connecting your accounting software to a calendar and a message template.
4. Job reminders and follow-ups
Two jobs here, both small, both worth doing once the first three are handled.
Before the job: a reminder the day before cuts no-shows and wasted trips. If you’ve ever driven forty minutes to an empty house, you know the number.
After the job: a follow-up a few days later asking if everything’s working, and a review request if it is. Reviews are the single most valuable marketing asset a trade business has, and almost nobody asks for them consistently because it feels awkward and it’s easy to forget once you’re on the next job.
Automating the ask means it actually happens. That alone tends to move a business from a handful of reviews to a steady trickle, which changes how you look to the next person searching for a plumber in your suburb.
What not to automate
The relationship. Don’t automate a personal response to a complaint, or a conversation about a job that’s gone wrong. Customers can tell, and those are the moments where being a real person is the whole product.
Anything you can’t describe. If you can’t explain how you price an unusual job, you can’t automate it. That’s not a failure, it just means it stays with you, which is fine.
Everything at once. The most common expensive mistake is trying to automate the whole business in one go. Pick the missed calls, get that working, live with it for a month, then look at the next one.
What it takes to start
Less than most people think, and the honest answer is that the hard part isn’t technical.
The hard part is writing down how you actually work. What makes a job urgent. What you include as standard and what’s extra. When you’d decline work. Most of that lives in your head and has never been said out loud, and getting it out is most of the job.
The technology after that is straightforward. Most trade software, accounting packages and calendars connect to automation tools without needing to be replaced, which matters because replacing the system you already know is expensive and disruptive for no gain.
The realistic picture
None of this replaces anyone. If you’re a two-person operation, you stay a two-person operation. What changes is how much of your evening belongs to you.
For most trade businesses the four above account for several hours a week and a meaningful number of lost jobs. That’s worth having, and it’s a smaller and more boring promise than most of what gets sold under the AI banner.
If your admin is eating your evenings, tell me where it goes and I’ll tell you which of these four would give you the most back. Usually it’s the phone.
Frequently asked questions
How can AI help a plumbing or electrical business?
Mostly by handling the admin around the work rather than the work itself. Answering the phone while you're on a job, turning site notes into a quote, chasing unpaid invoices and sending job reminders are all repetitive, rules-based tasks that automate cleanly and typically eat several hours a week.
What should a tradie automate first?
Missed calls, almost always. Every unanswered call is a job that goes to the next business on the list, and unlike most admin problems it costs you revenue directly rather than just time.
Do I need to change the software I already use?
Usually not. Most trade software, accounting packages and calendars can be connected to rather than replaced. Replacing the system you know is expensive and disruptive, so the better first move is almost always to automate the gaps between what you already run.
Isn't this expensive for a small operation?
It depends entirely on what you automate. A single workflow handling one repetitive job is a modest fixed-scope piece of work with low running costs. It gets expensive when people try to automate everything at once, or automate a process nobody has written down.
Will this make my business feel impersonal to customers?
It shouldn't, if you automate the admin rather than the relationship. Customers don't care who generated the reminder text. They care that they got one, that the quote arrived the same day, and that the phone got answered.
Wondering what this would look like in your business? A short chat is usually enough to tell.
Let’s chat