How Contractors Get Cited by ChatGPT and Google AI Overviews for 'Best [Trade] Near Me'
Ask ChatGPT or Google's AI Overview "who's the best plumber near me" and you get a shortlist of three to five named companies with a sentence of reasoning each. That shortlist is not the map pack. It's assembled differently, from different signals, and most contractors have never checked whether their business shows up in it. We started tracking these answers across our home-service accounts about a year ago, and the gap between who ranks in the map pack and who gets named by AI is wide enough that plenty of businesses winning one are invisible in the other.
This is the part almost no ranking guide touches. Everyone writes about Google Business Profile and reviews for the local pack. Nobody explains how an answer engine decides which HVAC company, roofer, or electrician to name when a homeowner asks it directly. The mechanics are knowable, the signals are testable, and you can measure your own citation share this week. This page walks through exactly how the shortlist gets built and what moves you into it.
The full local-SEO picture, map pack included, is in our plumber local SEO guide, and the broader argument for treating answer engines as a channel is in plumber AI search visibility. The same mechanics apply trade by trade — see how it plays out for electrical companies. A large share of what these models cite is directory data, which is why consistent citations matter more here than they do in the map pack.
How an AI answer engine actually builds a 'best [trade] near me' shortlist
When a homeowner asks ChatGPT or Google's AI Overview for the best plumber in their city, the engine isn't ranking your website the way classic search does. It runs a live retrieval pass, pulls a handful of sources it trusts for that query, and then writes a shortlist by pattern-matching which business names show up repeatedly across those sources with consistent, positive context. The named companies are the ones the model can corroborate from more than one place. A business that only exists on its own website almost never gets named, because there's nothing to corroborate.
What we actually see across the accounts we track: the sources these engines lean on for home-service queries are heavily third-party. Directory pages, 'best plumbers in [city]' listicles, Reddit and Nextdoor threads, Yelp and Angi profiles, and the Google Business Profile itself. Your own site matters for confirming details, but the decision to name you is made off-site. That's why a contractor can rank #1 in the map pack and still get left off the AI shortlist, and why a mid-pack company that's mentioned in three local 'best of' roundups gets named first.
The other thing that trips contractors up: these engines resolve you as an entity, not a URL. If 'Rapid Rooter' shows up as Rapid Rooter LLC on your site, Rapid Rooter Plumbing on Yelp, and Rapid-Rooter on your GBP, the model can't confidently merge those into one business, so your mentions get split and none of them cross the threshold to be named.
The four signals that get a home-service business named
From what we can reproduce on demand, four signal categories decide whether a contractor makes the shortlist. First, review volume and recency read as a bundle. It's less about a perfect 5.0 and more about a steady stream of recent, detailed reviews that mention the specific service and the city. A roofer with 180 reviews averaging 4.7 that mention 'storm damage' and the town name gets named ahead of a 5.0 with 22 reviews from two years ago, every time we test it.
Second, structured data that spells out who you are, what you do, and where. LocalBusiness or the specific subtype (Plumber, HVACBusiness, RoofingContractor) schema with matching name, address, phone, service list, and service area gives the engine machine-readable confirmation instead of forcing it to guess from page copy. Third, entity consistency across every surface. The exact same business name, phone, and address on your site, GBP, Yelp, Angi, BBB, and every citation. This is the single most common thing we fix on new accounts, and it's the cheapest.
Fourth, and the one contractors underrate most: third-party mentions with your name in context. Being listed in a genuine 'best HVAC companies in [city]' article, quoted in a local news piece about the winter freeze, or recommended by name in a Reddit thread. These are the corroborating sources the model reaches for. One good local roundup placement moves the needle more than another month of on-page tweaks, because it directly feeds the retrieval layer the engine is reading from.
Map pack signals vs AI citation signals — where they split
The map pack rewards proximity, GBP completeness, and review velocity inside Google's own ecosystem. AI answer engines don't care about proximity the same way. A homeowner asking ChatGPT 'best electrician near me' with location sharing off gets a citywide answer built from published sources, not a ranked-by-distance list. That single difference explains most of the mismatches we see between a client's map rankings and their AI visibility.
Concretely: the map pack can name you because you're 1.2 miles from the searcher and your GBP is dialed in. The AI shortlist names you because BuildZoom, a local blog, and forty recent reviews all describe you as the panel-upgrade electrician in that city. You can win the first with zero off-site presence. You cannot win the second. That's why we now treat 'get mentioned in the sources AI reads' as a separate work stream from GBP optimization, not a byproduct of it.
This is also why AI citation is more defensible once you have it. Map pack position swings with proximity and competitor review spikes. But once three or four durable third-party sources describe you as the best drain-cleaning company in town, that framing persists across model updates and query variations, because it's baked into the corpus the engine retrieves. The full interplay between these two systems is laid out on the hub guide at our main guide.
How to test your own citation share this week
You don't need a tool to start. Open ChatGPT, Google AI Overviews, and Perplexity, and run the exact queries a homeowner would: 'best plumber in [your city]', 'who should I call for a water heater replacement in [city]', 'top-rated HVAC companies near [neighborhood]'. Run each three or four times, because these answers vary between sessions. Log every business named and count how often your name appears versus your competitors. That ratio is your citation share, and it's the number to move.
When you do this across a few queries, a pattern shows up fast. Note which sources the engine cites or links, because those are the exact pages you need to be on. If Perplexity keeps citing a 'best contractors in [city]' directory you're not listed in, that directory is now your highest-priority task. We keep a simple monthly log of these results per client so we can see citation share climb as the third-party work lands, usually with a four-to-eight-week lag before a new mention starts getting pulled into answers.
One practitioner note that saves wasted effort: don't chase every engine equally. For home services we see Google AI Overviews and ChatGPT drive the real homeowner traffic, while the others are worth monitoring but not optimizing for individually. Fix the underlying signals — entity consistency, reviews, schema, and third-party mentions — and you improve across all of them at once rather than gaming any single one.
The 30-day plan to move into the shortlist
Start with entity cleanup, because it's free and it unblocks everything else. Pick one canonical business name, phone, and address, then make every surface match exactly: website footer, GBP, Yelp, Angi, BBB, Facebook, and every existing citation. Add or correct LocalBusiness schema with your specific trade subtype, service list, and service area. This is the week-one job, and on most accounts it alone lifts how confidently the engine can name you.
Weeks two and three are about feeding the retrieval layer. Get into the real local roundups by pitching the writers of existing 'best [trade] in [city]' articles, claim and complete any directory profiles that AI engines keep citing, and turn recent jobs into detailed reviews that name the service and the city. Ask the customer to mention what you fixed and where, because that specific language is what the model matches against the query. Generic 'great service, five stars' reviews barely register in the shortlist decision.
Week four, re-run your citation-share test and compare it to your week-one baseline. Some mentions won't be indexed yet, so treat the four-to-eight-week window as normal. The businesses that win this are the ones treating AI citation as its own discipline sitting alongside the map pack, not hoping their GBP work quietly covers it. It doesn't, and the contractors who figure that out first in each market get named while their competitors are still optimizing for a search result homeowners increasingly skip.
This is one piece of the bigger picture. For the complete playbook, read our main guide: Contractor Local SEO.
FAQ
Frequently Asked Questions
Is getting cited by AI the same as ranking in the Google map pack?
No, and treating them as the same is the most common mistake we see. The map pack rewards proximity, GBP completeness, and Google reviews. AI answer engines build their shortlist from third-party sources they can corroborate — directories, local 'best of' articles, forum threads, and review platforms. You can dominate the map pack and be completely invisible in ChatGPT's answer, because they read from different places and weight different signals.
How do I actually check if ChatGPT or Google AI recommends my business?
Run the queries a homeowner would use — 'best [your trade] in [your city]', 'who should I call for [specific job] near me' — in ChatGPT, Google AI Overviews, and Perplexity. Run each several times since the answers vary, and log which businesses get named and which sources get cited. The share of times your name appears is your citation share, and the cited sources tell you exactly which pages you need to get onto.
Do reviews affect whether AI names my company?
Yes, heavily, but volume and recency matter more than a perfect average. Across the accounts we track, a business with 150-plus recent reviews that mention the specific service and city gets named ahead of a flawless 5.0 with a handful of stale reviews. The engine reads reviews as corroborating evidence, so reviews that name what you did and where you did it carry far more weight than generic praise.
What structured data do home-service contractors need to get cited?
Use the specific LocalBusiness subtype for your trade — Plumber, HVACBusiness, RoofingContractor, Electrician — rather than generic LocalBusiness, and populate name, address, phone, service list, and service area so they match your other profiles exactly. Schema gives the engine machine-readable confirmation of who you are and what you do, which reduces how much it has to guess from page copy and makes it more confident naming you.
Why does my competitor get named by AI when I outrank them on Google?
Almost always because they're mentioned in more third-party sources than you are. If they're listed in two or three local 'best [trade]' roundups, quoted in a news piece, or recommended by name in a Reddit thread, the engine can corroborate them from multiple places and names them first. Your Google ranking doesn't feed that decision. The fix is getting into the same off-site sources the engine is reading from.
How long until third-party mentions start showing up in AI answers?
Expect a four-to-eight-week lag between a new mention going live and it getting pulled into answers, because the engine has to index the source and start retrieving it for your queries. That's why we baseline citation share before starting the off-site work and re-test monthly — the improvement is real but delayed, and contractors who quit at week two conclude it didn't work when the mentions simply hadn't been picked up yet.
See where you rank right now
Free local SEO audit — GBP, reviews, and your map rankings. No sales call.
Get your free audit