How homeowners find contractors now: ranking in ChatGPT, Gemini, and AI Overviews

A homeowner in your town just typed "who's the best lawn care company near me" into ChatGPT. It answered with three company names and never showed a list of links. If yours wasn't one of them, you didn't lose a ranking. You were never in the room.

What actually changed

For twenty years, local search meant one thing: rank in the map pack and page one, win the click. That still matters, a lot. But a growing slice of homeowners now start (and often finish) their contractor search inside an AI answer: ChatGPT, Gemini, Perplexity, or the AI Overview parked at the top of Google itself.

These engines don't return ten links. They return a recommendation: two or three names, a sentence about each, maybe a price range. Getting into that recommendation is what people are calling AEO: Answer Engine Optimization. Strip the jargon and it's this: make your business easy for a machine to understand, verify, and confidently repeat.

The good news for owners: AEO isn't a replacement for SEO: it's built on the same foundation. The companies winning AI mentions are mostly the ones who did local SEO properly and then added a machine-readable layer on top.

What AI engines actually cite

When an AI assembles an answer about local contractors, it's synthesizing from sources it can read and trust. Four things show up again and again in what gets cited:

1. Structured data (schema)

Schema markup is a machine-readable label set on your pages: this is a LocalBusiness, here are its services, its areaServed, its aggregateRating, its FAQs. Humans never see it; machines rely on it. A page that says "we do aeration in Springdale for $250–$400" in clean schema is quotable. The same fact buried in a paragraph of marketing copy inside page-builder div soup is a guess. (This is a big reason we build Node-rendered sites with full schema control instead of WordPress, more on that in our WordPress article.)

2. Entity consistency

AI engines cross-reference. Your name, address, phone, hours, and service list on your site, Google Business Profile, Yelp, Facebook, BBB, and industry directories need to agree. Every mismatch (old address on Yelp, different phone on a directory) lowers the machine's confidence that it knows who you are. Low confidence means you get skipped, because the AI would rather recommend the company it's sure about.

3. Genuine service-area content

Engines quote pages that actually answer questions: what does aeration cost here, when should you treat for grubs in this region, what's included in a mosquito program. Content with real local specifics (your climate, your grass types, your pricing logic) gets cited. Which brings us to what doesn't...

4. Reviews: volume, recency, and detail

Ask an AI about a local company and watch how often it paraphrases review themes: "customers mention reliable communication and fair pricing." Review volume and recency function as the trust layer of AI answers. A steady drumbeat of detailed reviews (the kind a post-job automation asks for) beats a wall of five-star "great job!" ratings from 2023.

llms.txt, in one minute

llms.txt is an emerging convention: a plain-text file at yoursite.com/llms.txt that hands AI crawlers a curated map of your business: who you are, what you do, where you serve, and links to your most important pages, without making the machine dig through navigation and scripts. Think of it as a robots.txt for meaning instead of permission. It costs almost nothing to add, and it removes ambiguity, which, as we covered above, is the thing that gets you skipped. Every M180 site ships with one.

How to measure AI visibility

You can't manage what you don't check. The manual version:

1. Write down the 10–15 questions your customers actually ask: "best pest control in [city]," "lawn treatment cost [city]," "who installs Christmas lights near [suburb]."
2. Once a month, run them through ChatGPT, Gemini, and Perplexity.
3. Log three things per prompt: Were you mentioned? What did it say? Was it accurate (prices, services, area)?
4. Track the trend, not any single answer: AI responses vary run to run.

The automated version: our SEO & Content Engine includes AI-visibility (AEO) reporting that runs those checks across engines on a schedule, alongside your Maps rank grid, so "are the AIs recommending us?" becomes a line on a dashboard instead of a mystery.

What to stop doing

Honesty section. Some habits from the 2015 SEO playbook now actively hurt:

• Thin city-page spam. Fifty near-identical pages with the city name swapped ("Lawn Care in Rogers! Lawn Care in Bentonville!") read as filler to both Google and AI engines. Ten pages with genuinely local substance beat fifty clones. Consolidate the rest.
• Keyword-stuffed FAQ walls written for crawlers, answering nothing. AI engines quote pages that answer clearly; they route around word salad.
• Buying junk backlinks. Machine trust is built on consistency and citations from real sources (suppliers, local news, industry associations), not directory farms.
• Ignoring your review responses. Unanswered complaints become quotable "customers report" material. Respond to everything.

The takeaway: AI engines recommend businesses they can understand and verify. Clean structure, consistent facts everywhere, real local content, and fresh reviews: that's the whole playbook. It's not magic; it's tidiness at scale.

The AI-visibility starter checklist

1. Ask ChatGPT and Gemini about your service in your city. Note what they say about you, and about competitors.
2. Audit name/address/phone/hours across your top 10 listings. Fix every mismatch.
3. Add LocalBusiness, Service, and FAQ schema to your key pages.
4. Publish llms.txt.
5. Replace your thinnest city pages with fewer, genuinely local ones.
6. Get your review request automated on every completed job.