TL;DR
Homeowners increasingly ask ChatGPT, Perplexity, and Google AI Overviews "who's the best HVAC company near me" instead of scrolling ten blue links. AI assistants recommend specific businesses, and those recommendations come from signals you can influence: structured data, review mass, consistent citations, answer-first content, and brand mentions across the sources LLMs read.
This is called GEO (Generative Engine Optimization), and for contractors it's the biggest positioning shift since the Google local pack. This guide covers how AI assistants pick businesses to recommend, the 7 factors that drive it, a DIY checklist, and how to measure whether you're showing up.
Shortcut: our AI Search Optimization service does this for you. Start with the $999 AI Visibility Audit (credited toward your first month of any SEO plan) or book a strategy call. New to this space? Read GEO vs SEO explained first for the strategic overview.
Why this matters now (the discovery shift)
For twenty years, "how do I get more customers online" meant one thing: rank in Google's ten blue links. That answer is quietly breaking. Homeowners are now typing full questions into ChatGPT, Perplexity, Claude, and Google's own AI Overviews, and they are getting one answer instead of ten links. The question is no longer "am I on page one," it is "am I the business the AI recommends."
The behavior shift is not evenly distributed, but it is real. Younger homeowners already default to ChatGPT for research decisions. Older homeowners still start on Google, but Google itself is now placing AI Overviews above the classic ten links on more than half of informational service queries. Either way, the click that used to reach a contractor website is being intercepted by an AI answer.
The dynamic that matters: when a homeowner asks ChatGPT "who's the best plumber in Charlotte," ChatGPT does not return ten options in a random order. It returns two or three specific business names, sometimes with a citation, sometimes without. Being one of those named businesses is what "AI visibility" actually means, and it is the difference between growing in 2026 and quietly losing share to whichever competitor got there first.
The early-mover window is unusually wide here. Almost no contractors have done anything intentional about AI visibility yet. Most agencies are not even offering it as a service. The market has not priced this in. That combination (real behavior change plus a mostly empty playing field) is why we treat GEO as the single highest-leverage positioning move for contractors in the next 24 months.
What GEO actually is (and the terms people use)
GEO stands for Generative Engine Optimization. It is the practice of making your business visible and recommendable inside AI-generated answers, whether that answer comes from ChatGPT, Perplexity, Google's AI Overviews, Gemini, Claude, or Microsoft Copilot. The goal is not to rank on a page of blue links. The goal is to be one of the businesses the AI names when a homeowner asks a buying question.
You will also see this called AEO (Answer Engine Optimization), AI search optimization, LLM SEO, or ChatGPT SEO. Same practice, different labels. GEO is winning as the industry term because it captures the mechanic more precisely: LLMs generate answers rather than retrieve pages.
GEO overlaps heavily with classic SEO. Google still feeds a huge share of the sources that LLMs use, and AI Overviews mostly summarize the current top-ten organic results. So classic search rankings are not going away, they are becoming an input to AI answers. What GEO adds on top is a set of new surfaces: model training data, live retrieval sources (Bing for ChatGPT search, Google for AI Overviews, curated citations for Perplexity), and citation-heavy sources like Reddit, Quora, and directories that LLMs disproportionately trust.
How ChatGPT actually decides who to recommend
Under the hood, "how to show up in ChatGPT" comes down to four mechanisms. Understanding each one is how you decide where to spend effort.
1. Training data brand mentions. Foundation models are trained on massive web crawls. A business that has been mentioned consistently across the internet (blog posts, directories, press, forums, trade publications) is baked into the model's default associations for its category. When you ask ChatGPT "who is a good HVAC company in Phoenix" without browsing, you are querying that baked-in knowledge. Getting into training data is slow (weighted heavily toward older, established brand presence) but it is the most durable form of AI visibility.
2. Live web retrieval. When browsing is turned on (which is the default for logged-in ChatGPT users on service queries), the model runs a real-time search and reads the top results. ChatGPT search leans heavily on the Bing index and weights structured, crawlable content. If your site is well-structured, fast, and answers the question in the first two paragraphs, you are the source ChatGPT quotes. If your site is a Flash-era brochure, you are invisible even when your competitors are ranking below you in Google.
3. Review platforms and aggregators. LLMs and their retrieval layers weight third-party review platforms heavily because they are structured, consistent, and hard to fake at scale. Google Business Profile, Angi, Yelp, BBB, and Nextdoor all feed into what ChatGPT sees. High review count and high star rating are among the strongest signals that a business is real and worth recommending.
4. Consistency of NAP and entity data. AI models cannot recommend what they cannot cleanly identify. If your business name, address, and phone number differ across the twenty places they appear on the web, the model has trouble deciding you are one entity. Consistent NAP, matching schema markup, and a complete Google Knowledge Panel collapse the ambiguity and make you recommendable.
One important nuance for Google AI Overviews specifically: Google's own AI answers pull disproportionately from pages already ranking in the top ten organic results for the query. So classic SEO still matters enormously. If you rank number three organically, you are in the answer pool. If you rank number thirty, you are not. Perplexity is different again, it cites sources directly, so structured, citable content wins there.
The 7 factors that make AI recommend a business
Below are the seven signals we look at first on every AI Visibility Audit. They roughly correspond to the mechanisms above, translated into things a contractor can actually do.
- Entity clarity. A complete Google Knowledge Panel, consistent name-address-phone across the web, and schema markup on your website (LocalBusiness, Service, FAQPage, at minimum). AI cannot recommend what it cannot identify.
- Review mass and rating. LLMs and their retrieval layers weight review count and sentiment heavily. 100+ Google reviews at 4.7+ stars makes you recommendable. Under 20 reviews and you are almost never chosen. See How to Get 100 Google Reviews for Contractors.
- Answer-first content. Pages that directly answer specific questions ("how much does AC replacement cost in Phoenix," "what does a panel upgrade cost in a 1950s home") get cited by Perplexity and AI Overviews. Wall-of-text sales pages do not. Every important service should have a page that opens with a plain-English answer, then explains.
- Third-party citations. Reddit threads, local directories, trade publication mentions, "best of" city lists, and podcast guest spots. LLMs trust what other sites say about you more than what you say about yourself, and Reddit in particular has become an outsized source for LLM training data.
- Ranking in classic search. AI Overviews mostly summarize the current top ten. If you rank, you are in the answer pool. This is why we sell GEO and classic SEO as one integrated plan. Also see the Google Guaranteed for Contractors guide, because LSA visibility now feeds AI answers too.
- Wikipedia and Wikidata presence (for larger operators). If your business is big enough to legitimately support a Wikipedia entry or a well-formed Wikidata record, that becomes canonical entity data for basically every LLM on earth. Not a fit for a one-truck operation. Very high leverage for regional multi-location operators.
- Freshness and consistency. Stale sites, conflicting hours, dead links, and abandoned social profiles all get downweighted. A recently-updated site with matching signals across GBP, Facebook, Angi, and the site itself outranks a competitor whose data has drifted, even when that competitor has been in business longer.
The DIY GEO checklist for contractors
If you want to do this yourself, here is the exact order to work through it. The steps below compound, so work top down.
- Full schema implementation. LocalBusiness, Service, and FAQPage schema on every service page. HowTo schema where appropriate. Validate with Google's Rich Results Test. This is the single highest-leverage technical fix.
- Google Business Profile completeness. Every field filled, every service listed, every category correct, hours accurate, photos updated in the last 90 days, and Q&A section pre-seeded with real questions and answers.
- Review velocity system. Text-based review ask within 2 hours of job completion. Aim for 15+ new reviews per month until you are past 100 total. Full playbook: How to Get 100 Google Reviews.
- FAQ blocks that answer real homeowner questions. Every service page should include 8 to 12 FAQs that mirror the exact phrasing homeowners use. These are the snippets AI Overviews and Perplexity quote directly.
- Answer-first blog content. Publish one deep post per month that directly answers a buying question in the first paragraph. Cost questions and "best of" comparisons perform disproportionately well.
- Reddit and Quora presence, done non-spammily. Answer real questions in your city's subreddit and in trade Q&A threads on Quora. Do not shill. Genuine helpfulness gets your brand into training data. Bragging gets you banned.
- Directory consistency audit. Every directory listing (Angi, Yelp, BBB, Nextdoor, chamber of commerce, industry associations) must show identical NAP. Fix drift before adding new listings.
- Get into "best [trade] in [city]" roundups. Local blogs, city magazines, and community sites frequently publish these lists. Reach out. Being named in three of these is worth more than a hundred backlinks from random directories.
- Monitor monthly. Once a month, open ChatGPT, Perplexity, Gemini, and Google (looking for AI Overviews) and ask the exact queries a homeowner would ask in your city. Screenshot who gets recommended. This is your GEO scoreboard.
- Keep classic SEO strong. AI Overviews pull from the current top ten, so classic ranking work is the cheapest way to increase AI visibility on Google. Do not treat GEO as a replacement, treat it as the layer on top.
How to measure AI visibility
The honest answer: there is no dashboard that gives you a clean "AI ranking" the way Google Search Console gives you organic rank data. Measurement is manual and directional, and that is okay because the signal is loud when you know what to look for.
The manual method takes about an hour a month. Build a list of 15 to 25 target queries a real homeowner in your service area would ask. Include commercial-intent queries ("best HVAC company in Charlotte," "who should I hire for panel upgrade in Denver") and problem queries ("my AC is blowing hot air, who do I call in Phoenix"). Run each query in ChatGPT, Perplexity, Gemini, and a Google search that shows AI Overviews. For each query, note: are you mentioned, cited (with a source link), or actively recommended? Compare to a fixed list of local competitors so you can see share of voice change over time.
The trend is what matters, not any single query. If your citation rate goes from 5% in month one to 20% in month four, your GEO stack is working, even if any specific query still shows a competitor.
Our AI Reporting add-on (+$499/mo, requires an active SEO plan) does this automatically across ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews, with monthly screenshots, share of voice tracking, and side-by-side competitor comparison. If you would rather not spend an hour a month running queries by hand, that is what it is for.
What SkillMammoth does (if you want it done for you)
We treat GEO as one integrated program with classic SEO because the two feed each other. Doing GEO without SEO is fragile. Doing SEO without GEO leaves 30 to 40% of future customer discovery on the table. Here is how it stacks:
AI Visibility Audit: $999 one-time. Manual brand search across ChatGPT, Perplexity, Gemini, and Google AI Overviews on your top 20 target queries. Full schema audit against LocalBusiness, Service, and FAQPage best practices. Citation gap analysis across the directories and platforms LLMs weight most. Prioritized 12-week action plan. Credited toward your first month of any SEO plan.
SEO + AI Search plans: Local $699/mo, Authority $1,499/mo, Pro $2,999/mo. Every plan includes classic Local SEO and GEO work on the same schedule, because the two share the same underlying signals. See pricing.
Add-ons for operators who want to go deeper (both require an active SEO plan):
- AI Reporting +$499/mo. Monthly AI visibility tracking across ChatGPT, Perplexity, Gemini, Claude, and AI Overviews, with screenshots and competitor share-of-voice comparison across 20+ queries.
- Off-Page AI Optimization +$999/mo. Active citation seeding on Reddit, Quora, niche forums, local directories, and digital PR placements to sources that feed LLM training data. This is the hardest layer to DIY and the layer that separates AI-visible operators from AI-invisible ones over a 12-month horizon.
Book a strategy call to see whether GEO is the right first move for your business, or start with the free SEO Audit and the AI Stack Picker if you want to poke around first.
Related reading
- How to Get 100 Google Reviews for Contractors
- Google Guaranteed for Contractors
- Marketing Attribution for Home Service Contractors
- AI Search Optimization service
- Local SEO service
- Local Service Ads management
- HVAC marketing, plumbing marketing, roofing marketing, electrician marketing, lawn care marketing
FAQ
What is GEO (Generative Engine Optimization)? GEO is the practice of making your business visible and recommendable inside AI-generated answers from tools like ChatGPT, Perplexity, Gemini, and Google AI Overviews. You will also see it called AEO (Answer Engine Optimization), AI search optimization, LLM SEO, or ChatGPT SEO. GEO is the term winning inside the industry.
Does ChatGPT really recommend specific local businesses? Yes, especially when browsing is enabled and the query is local. Ask ChatGPT "who's the best HVAC company in Phoenix" or "recommend a plumber in Charlotte" and it will return two or three specific business names. Which businesses depends on training data mentions, live web retrieval, and review platform signals.
How is GEO different from SEO? They overlap heavily. Classic SEO ranks pages in Google's ten blue links. GEO makes you visible inside AI-generated answers. Because Google AI Overviews and ChatGPT search both draw from currently-ranking content, classic SEO still feeds GEO. GEO adds new surfaces on top: LLM training data, Reddit and Quora presence, and structured citable content.
How long does it take to show up in ChatGPT recommendations? 60 to 180 days depending on your baseline. Review mass and directory consistency move fastest (30 to 60 days). Getting into LLM training data via Reddit, press, and roundups takes 90 to 180 days. Wikipedia-grade entity presence, when applicable, takes longer. Anyone promising two-week AI rankings is guessing.
Can I do GEO myself? Yes. The DIY checklist above is the whole playbook. The hardest layer to DIY is off-page citation work (Reddit, Quora, digital PR), because it requires steady non-spammy participation and outreach relationships. Everything else is achievable with time and discipline.
How do I check if AI recommends my business? Manually run your top 15 to 25 buying queries in ChatGPT, Perplexity, Gemini, and Google (watching for AI Overviews). Note whether you are mentioned, cited, or actively recommended. Compare to a fixed list of competitors so you can see change over time. Our AI Reporting add-on automates this at +$499/mo with an active SEO plan.
What does AI search optimization cost? Our AI Visibility Audit is $999 one-time and credits toward your first month of any SEO plan. Integrated SEO + AI Search plans run $699/mo (Local), $1,499/mo (Authority), and $2,999/mo (Pro). Deeper add-ons are AI Reporting (+$499/mo) and Off-Page AI Optimization (+$999/mo), both requiring an active SEO plan.
Do reviews affect ChatGPT recommendations? Heavily. LLMs and their retrieval layers weight review count and sentiment as one of the strongest signals a business is real and worth recommending. Under 20 reviews and you are almost never chosen. Past 100 reviews at 4.7+ stars, you become part of the default recommendation set. Full playbook: How to Get 100 Google Reviews.
What is LLM SEO or ChatGPT SEO? Same practice as GEO, different names. LLM SEO frames it around the underlying technology (large language models). ChatGPT SEO frames it around the single most-used product. GEO (Generative Engine Optimization) is the umbrella term the industry has settled on because it covers ChatGPT, Perplexity, Gemini, Claude, AI Overviews, and whatever comes next.
Will AI search replace Google? No, at least not in any timeframe you should plan around. The discovery mix is shifting, not flipping. Google still dominates the top of the funnel and now feeds its own AI Overviews from the same organic ranking system. You need both classic search visibility and AI visibility, which is exactly why we sell them as one integrated plan.
Want to implement these strategies?
Book a free strategy call and learn how we can help grow your contractor business.
Book Your Free Call
