AI Visibility · 28 June 2026

By Eddie Eastwood, Founder

How Tradespeople Get Recommended by AI (Plumbers, Electricians, Builders)

It is half past seven on a Sunday evening, a boiler has packed in, and the house is getting cold. Once, the homeowner would have reached for Google, tapped in "emergency plumber near me", and run a thumb down the blue links with the phone already open. Tonight they open ChatGPT instead, or tap Google's AI answer, and ask it straight: "who can fix a broken boiler near me tonight?" Back comes a short paragraph naming two or three firms. They ring the first one that sounds right.

Two heating engineers live in that town. Both are genuinely good at the job; both could be at the door within the hour. One of them gets named in that reply and takes the call-out. The other is never mentioned, and never finds out the job existed.

This guide is about which engineer you want to be, and what decides it: Answer Engine Optimisation, or AEO. It is how you end up inside the AI's reply when a customer asks for someone in your trade, instead of watching the job go to whoever the engine happened to know about. Most of what follows holds for any trade, so wherever you read "boiler", picture your own work. For the underlying mechanics that apply to every local business, the complete guide to AEO for local businesses is the place to start; this piece stays in the van with you and gets specific to the trades.

Two engineers, one Sunday night

Picture the two of them.

The first has been trading for fifteen years and does excellent work. He also has no Google Business Profile, a single-page website that loads slowly and calls what he does "heating solutions", and a handful of reviews from three years ago on one site. When the AI goes looking for someone to name, there is almost nothing about him to find, read or quote, so it reaches straight past him.

The second is no better with a spanner, but her business is legible to a machine. Her Google profile is complete and lists the towns she covers. Her site has separate, plainly written pages for boiler repair and boiler replacement in each of those towns. Her Gas Safe registration is set out in black and white. Recent reviews land on Google and Checkatrade most weeks. When the same question is asked, the AI has a clear, consistent, well-reviewed business to point at, so it points at her.

Nothing about that outcome is about skill. It comes down to what each engineer left behind for the AI to find. The rest of this guide is the second engineer's setup, taken apart piece by piece, and why the gap between the two is smaller and far more winnable than it looks.

Why the invisible engineer loses (and why that is good news for you)

The Sunday-night scene is not a rare one. It is becoming the ordinary way people find a tradesperson, and an AI-generated summary now sits at the top of a large share of everyday Google searches. When that summary appears, the click you used to count on mostly does not happen: Ahrefs studied 300,000 keywords and found that the top-ranking page loses about 58% of its click-through rate once an AI Overview is present (Ahrefs, December 2025)^[1]. People are also handing the local choice to the AI itself.

BrightLocal's Local Consumer Review Survey 2026 found the share of consumers turning to ChatGPT and similar tools for local business recommendations leapt from 6% a year earlier to 45%, now the third most popular way to find a local business; 40% said they trust AI tools to recommend businesses, and 42% trust them as much as traditional reviews (BrightLocal, 2026)^[2]. Add those together and the recommendation that once came from a neighbour or a Google list now happens inside an AI reply you never see. Your competitors' names may well be in it.

Many trades assume that reply is rigged for big national brands with marketing budgets. For trade work the evidence runs the opposite way, for two concrete reasons.

The first is that AI hangs onto location. When you put a question to an AI tool, it rarely hunts for your exact words. Behind the scenes it reshapes and broadens the request before it starts pulling sources. The natural fear is that this reshaping quietly bins "in Leeds" or "near me" and washes out the local edge your whole trade depends on.

It does not. The engine trims the adjectives, sheds the price filters and cuts the brand lists down, and yet the town you cover is the one detail it nearly always leaves standing. That sits squarely with the broader pattern in the BrightLocal evidence, where engines lean hard on location and reviews to settle a "near me" question.

That single fact tilts the field toward you. A national chain has to be the answer everywhere at once; you only have to be the clearest, best-reviewed, best-documented answer to "emergency electrician in Bristol" or "boiler repair in Stockport". That is a much shorter climb, and it is exactly the ground a local trade is built to hold. The shape of the question helps as well. A phrase like "best emergency plumber in Bristol" names a place and an intent, so it survives the reshaping cleanly and points the engine straight at the kind of answer you can be.

The second reason is emptier than you would expect: most of your rivals are not in the data at all. BrightLocal reports that only around 35% of small businesses have claimed a Google Business Profile (BrightLocal, Local Consumer Review Survey 2026)^[2]. That is why the same two or three firms keep surfacing in "near me" answers, because everyone else is invisible to the engine. The first engineer is not losing to a corporation with deep pockets. He is losing to the few local firms who simply turned up in the data.

What the AI did in those three seconds

In the gap between the homeowner's question and the answer that named the second engineer, the AI did a good deal more than match a phrase.

It did not run "who can fix a broken boiler near me tonight?" as a single search. Google calls its own method "query fan-out": it breaks a request into subtopics and fires off "a multitude of queries simultaneously on your behalf" (Google, AI in Search, May 2025)^[3]. One question about a dead boiler quietly became separate look-ups for local heating engineers, emergency call-out availability, reviews, prices, Gas Safe registration and more, then got reassembled into one reply.

We take that mechanism apart in full in the query fan-out guide; here, the trade lesson is what counts. You are not chasing one magic phrase. The second engineer won because her presence answered the whole cluster of small questions at once: what she does, where she works, what it costs, how fast she turns up, what she is registered to do. The more of those small questions you answer plainly across the web, the more ways there are for an engine to arrive at your name.

The AI also read well beyond her own website. To settle a "near me" question it leaned on her Google Business Profile, the review sites, the trade directories, the forums and the local press, which are the very places a homeowner would check before trusting a stranger in their home. BrightLocal found the average customer now consults around six review sites before choosing, and that 97% read reviews at all (BrightLocal, 2026)^[2].

So your visibility is never just a website job: it is built from everything the rest of the web says about you, and whether that picture is consistent. How each engine decides which of those sources to trust is its own subject, covered in how AI engines pick their sources; for now the takeaway is that you have to exist, and agree with yourself, across all of them.

One last thing the homeowner never saw: ask the same engine the same boiler question an hour later and it might name a different firm. That movement is normal, not a glitch and not a penalty. That is why everything below aims at steady, lasting signals instead of chasing one flawless answer on one particular evening.

The visible engineer's setup

Everything that put the second engineer inside the answer comes down to seven things she had in place. Take them roughly in this order; the first few do most of the work.

A Google Business Profile, claimed and filled in

She had claimed hers and completed every field: exact business name, address or service-area settings, phone, hours, and every job she takes on, described the way customers say it ("emergency boiler repair", "fuse board replacement", "blocked drain", "bathroom installation"), with real photos of finished work and her service towns marked one by one. Google's AI answers and most other engines lean on that profile heavily for local questions, and for "near me" trade work it is often the signal that settles who gets named. Only around 35% of small businesses have claimed one (BrightLocal, 2026)^[2], so doing it properly is the highest-return hour most trades will ever spend.

Recent reviews, on more than one site

What convinces a homeowner is what convinces the engine: a run of recent reviews is the firmest trust signal either of them gets for a local trade. People want them fresh: BrightLocal found 74% pay attention only to reviews from the last three months, and 31% will not consider a business rated below 4.5 stars (BrightLocal, 2026)^[2]. The second engineer asked every satisfied customer the moment the job was signed off, with a direct link rather than a vague "leave us a review sometime".

She gathered them on Google first, then the platforms that carry weight in her trade: Checkatrade, Trustpilot, manufacturer approved-installer listings. She replied to all of them, the awkward ones included, because visible replies add to the consistent picture an engine reads. She never bought a single fake one: beyond the deception, planted reviews are now unlawful under UK consumer law and platforms strip them out, so the risk is real and the gain imaginary. Two genuine, recent five-star reviews a month tell a customer and an engine the same thing, which is that this business is busy and reliable right now.

Service and area pages that answer real questions

Her website earned its place by answering specific questions plainly, not by greeting visitors with "welcome to our website" copy. That meant a dedicated page for each service, named the way customers name it ("Emergency plumber", "Boiler repair and replacement", "Rewiring", "EV charger installation", never "thermal and electrical solutions"), and a genuine page for each town she covers, written about that town, instead of one page listing twenty place names.

Each page carried the practical facts a person and an engine both want: call-out charges or price ranges, emergency response times, what a job includes, and what to expect. She wrote the way a customer asks, so "how much does it cost to replace a boiler in Manchester?" became a question her page could simply answer. A page that hides that behind brochure language is one an engine cannot quote.

Accreditations, set out in black and white

This is where trades hold an advantage almost no other local business can match. A trade carries trust marks an engine loves precisely because they are clear, checkable facts it can repeat: Gas Safe registration for gas work, NICEIC or NAPIT for electrical, OFTEC for oil, FENSA for windows, TrustMark, manufacturer approved-installer status, and trade-body membership. The second engineer put hers in plain sight, on her website, on her Google profile, and on every listing, written out as concrete detail.

"Gas Safe registered, registration number 123456" is exactly the kind of verifiable line that makes an engine more confident naming her and reassures the homeowner reading the reply. The rule with accreditations is simple: state only what you genuinely hold, keep the numbers current, and never let a lapsed registration linger on the page. Used properly, they are the cheapest credibility a trade can put in front of a machine, and the one asset a faceless national brand cannot easily borrow.

The directory and "approved installer" listings that matter

Because the AI reads the wider web, the second engineer made sure she existed on it well beyond her own site. For a trade, the listings that earn their keep are usually the trade-specific directories (Checkatrade, Rated People, MyBuilder, TrustMark), kept accurate and consistent; the manufacturer "find a fitter" or "approved installer" pages for the brands she fits, which carry real authority; the trade-body and accreditation registers, which double as proof of her qualifications; and the local press and community sites, which still pull weight even as their reach shifts.

For trade work these specialist listings tend to count for more than generic business directories, because they carry the qualification and review signals an engine weighs most heavily here.

Give the crawler a page it can read

Every signal above counts for nothing if the crawler never reaches your words in the first place. Some builders and heavily designed sites paint their text with JavaScript that only runs in a visitor's browser, and several AI crawlers do not run it, so they land on what looks like a blank page and move on. The second engineer's site rendered its words as plain text.

Two minutes tells you where you stand: on the site, right-click and hit "View page source", then read the raw text for your service names, town names and phone number. If they are missing, whoever built the site can put it right. This sits beneath everything else rather than on top of it, but on the wrong setup it silently undoes all the rest.

The same facts, everywhere

Finally, her business name, address and phone number matched exactly across her website, her Google profile, Checkatrade and every other listing. When those details disagree (an old mobile here, a slightly different trading name there), an engine grows less sure who it is dealing with, and a less certain engine is a less likely recommendation. It is dull housekeeping that quietly decides more than it should, especially for sole traders who have switched numbers or rebranded over the years.

Try it on your own trade tonight

You can see where you stand right now, for nothing, in about ten minutes.

1. Put the same questions a real customer would to ChatGPT, Google's AI Mode and Perplexity: "best [your trade] in [your town]", "[your trade] near me for [the job]", "who should I call for [problem] in [area] tonight?"
2. Write down which firms get named, whether yours is among them, and the sources each answer leans on.
3. Run it a few more times. The replies shift from one go to the next, so you are looking for the pattern rather than a single result.

If your name never comes up, you have just found the gap to close. If the same competitor is named every time, study what they have that you do not: fresher reviews, sharper service pages, a fully completed Google profile, accreditations on display, a wider spread of listings across the web. That difference is your list of jobs to do.

Where the invisible engineers go wrong

The first engineer's problems are common, and every one of them is fixable:

• No Google Business Profile, or one left half-finished. This is the single most common reason a trade never appears in local AI answers.
• Reviews that are old, thin, or stuck on one site. A small pile from years ago reads like a business that may have packed up.
• One page trying to do everything. A lone "Services" page spanning ten jobs across five towns answers none of the precise questions the AI breaks a request into.
• Brochure language where an answer should be. "Bespoke heating solutions tailored to you" gives an engine nothing to quote; "emergency boiler repair in Stockport, same-day call-out" gives it everything.
• Accreditations left off the page instead of written out with their numbers.
• Contact details that disagree across the website, Google and the trade directories.
• A site a crawler cannot read because the words load only in a browser.
• Leaning on a Google ranking to do the work. A first-page Google spot and a mention inside ChatGPT are won in two different arenas now, as the guide on Google rankings versus ChatGPT lays out.

Frequently asked questions

I'm a one-van sole trader. Is this really worth my time?

More than it is for a big firm, in fact. Most of your competitors have skipped the basics (only around 35% of small businesses even have a Google Business Profile), so the bar is low. The first three things here, a complete Google profile, recent reviews, and real service and area pages, cost effort rather than money, and between them they close most of the gap.

Can I pay my way into ChatGPT or an AI answer?

No. A real AI recommendation cannot be bought; it comes from the signals above: reviews, a complete profile, clear pages, real listings. If someone sells a guaranteed AI slot for a set fee, walk away.

Will AI only ever point people to big national chains?

Not for "near me" trade work. Location is the signal an engine almost always keeps, and that works in a local trade's favour. A well-documented, well-reviewed local plumber can be a better answer to "emergency plumber near me" than a national brand with no real presence on the ground.

How soon will I start showing up?

No timeline is guaranteed, and the names in the reply move about from one day to the next. Finishing your Google Business Profile and pulling in recent reviews can start to shift things inside a few weeks, while filling out listings and off-site presence is a longer haul. Patient groundwork wins out over any shortcut.

Is AEO for plumbers any different from AEO for other trades?

The core is the same, but plumbing leans harder on a few things. Emergency and "near me" questions make up a bigger share of plumbing searches, so location-rich service and area pages matter more, and manufacturer "approved installer" listings for the boiler brands you fit carry real weight. Reviews that name the actual job ("fixed a burst pipe on a Sunday") are worth most, because they match the exact questions customers put to AI.

Which directory should I prioritise?

Google Business Profile first, every time. After that, the trade-specific directories and any manufacturer "approved installer" pages for the brands you fit, since those carry the qualification and review signals an engine weighs most for trade work.

How do I get reviews that actually help me get named?

Ask the moment a job is signed off, with a direct link rather than a vague "leave us a review sometime", and prompt the customer to mention the specific job and town. A review that says "replaced our boiler in Stockport, same day" matches the exact question a homeowner puts to an AI, so it carries more weight than a generic "great service". Start on Google, add the one platform that carries weight in your trade, and answer every review that lands.

Do accreditations like Gas Safe or NICEIC actually help with AI answers?

Yes, because they are clear, checkable facts an engine can repeat, and a faceless national brand cannot easily borrow them. Write them out in full ("Gas Safe registered, registration number 123456") on your website, your Google profile and every listing. State only what you genuinely hold and keep the numbers current: a lapsed registration left on the page does more harm than good.

I cover several towns. Should I make a page for each one?

Yes. Because the engine keeps the location when it reshapes a query, a genuine page written about each town beats one "areas we cover" page listing twenty place names. Give each its own page with the same clear facts (call-out charges, response times, what a job includes) and keep your Google Business Profile service area accurate. A specific, well-reviewed page for each town is far more winnable than trying to claim them all at once.

How do I check whether an AI can even read my website?

Open your site, right-click and choose "View page source", then look for your service names, town names and phone number in the text. If they are there as plain words, an AI crawler can read them. If the page looks nearly empty, your site is probably painting its text with JavaScript that several AI crawlers do not run, and whoever built it can fix that. It is a two-minute check that sits beneath everything else.

Where to start

If you do only three things from all of this, do these: claim and complete your Google Business Profile; start gathering recent reviews on Google and the main directory in your trade; and rewrite your key service and area pages so they answer real customer questions plainly, accreditations and all. For most trades, those three close most of the distance between the first engineer and the second.

When you would rather have this checked for you first, a free QBiz Leads AI visibility check scans your website in about thirty seconds and returns a clear pass or fail on the key signals that decide whether AI tools can find and recommend your business, handed to you as a jobs-to-fix list in priority order. It is the cheapest hour you will spend before you commit to any of the work above.

Get your AI Visibility audit →