GEO for car dealers: getting cited by AI search
Ask a younger buyer how they found their last car and you'll increasingly hear a version of: "I asked ChatGPT," or "I just used the AI thing at the top of Google." They didn't scroll a page of ten blue links and weigh their options. They asked a question and got an answer — a short, confident, sourced answer that named a few stores and moved on. If your dealership wasn't in that answer, you didn't lose the ranking battle. You weren't in the room.
That shift has a name now: Generative Engine Optimization, or GEO. It's the discipline of getting your business cited inside AI-generated answers — Google's AI Overviews, ChatGPT, Perplexity, Gemini — rather than just ranking in the traditional results below them. For local businesses like dealerships and repair shops, it's quickly becoming the most important visibility work there is. Here's how it actually works, without the hype.
First, understand the index cut
Google quietly changed the rules of the game. For years, a weak page was simply demoted — it ranked on page four instead of page one, but it still existed in the index. Now Google increasingly removes low-value and redundant pages at the index layer entirely. The bar moved from "rank above the competition" to "convince Google this page is worth keeping at all."
This matters enormously for AI search, because the AI engines pull from those same indexes. Google's AI Overview draws on Google's index; tools built on Bing draw on Bing's. A page that gets cut from the index is a page the AI can never cite. So the foundational GEO move isn't some clever AI-specific trick. It's making sure every page on your site clears the keep-it bar: real, distinct content, a clean mobile structure, and genuine inbound links. Survive the cut and you become eligible to be the answer. Get cut and you're invisible to both Google and the AI.
Optimizing a page to survive Google's index cut is the same work that gets it cited in AI answers. You build once, for both.
Skip the gimmicks — Google already told us they don't work
Before the to-do list, a warning, because the AI-SEO space is full of snake oil right now. You do not need to buy an "llms.txt" file, bolt on weird AI-specific markup, over-optimize your schema with gimmicks, or rewrite everything into rigid question-then-answer-per-sentence formatting "for the AI." Google has said plainly that these don't matter. The engines reward the same fundamentals they always have — useful, trustworthy, well-structured content tied to a real business. Anyone selling you a magic AI-formatting hack is selling you a distraction.
The dealership GEO playbook
What does work is unglamorous and specific. Here's the structure we build for automotive clients.
1. Mirror your Google Business Profile with real pages
Your GBP lists categories and services — new car dealer, used car dealer, auto repair, oil change, brake service, and so on. Most dealership sites try to support all of that with a handful of thin pages. That's backwards. Give every meaningful category and service its own dedicated, substantial page, and link them together in the same hierarchy your GBP uses: homepage to secondary categories, secondary categories down to individual service pages. When your site mirrors your profile, Google maps every surface to one coherent real-world entity — and that coherence is exactly what AI engines trust when deciding whom to cite.
2. Use LocalBusiness schema, the specific type
Mark up your local pages with LocalBusiness schema — and the most specific recognized type, like AutomotiveBusiness or AutoDealer, not the generic catch-all. Give each location page its own distinct schema block with its own service type and service area; don't copy one block across pages and swap the city, because Google filters that as duplicate. Populate areaServed by listing each city you actually serve, individually and spelled exactly as in your GBP — never a vague "Southeast region." And match your name, address, and phone to the profile character-for-character. "St" vs "Street," a swapped tracking number — small mismatches quietly break the website-to-profile trust loop the AI relies on.
3. Write content a competitor can't reproduce in ten minutes
This is where most dealer sites fail. Generic, AI-average pages — "Top 5 Reasons to Buy a Used SUV" — are precisely the redundant content Google now cuts and the AI never cites, because they're the consensus the engine already knows. Before you write a page, look at what the top results actually say and then write what they don't. Mine the real questions: People Also Ask, the threads on Reddit where someone in your city asks about your service, the reviews on competitors' profiles, the actual questions your phone gets all day. Ground each page in operator-level experience — real financing scenarios, real trade situations, real local conditions — so it adds something not already on the internet.
4. Be genuinely, specifically local
Reference real neighborhoods, real roads, real local conditions. "We serve the greater metro area" tells the AI nothing. "Drivers coming off the salt-treated roads through the north side every winter" tells it you're a real business that operates in a real place. Build location pages around actual landmarks near the spots where you rank just outside the top three, to prove you operate there. The goal is for search and AI systems to believe — because it's true — that you are genuinely local.
5. Earn a few real links, and keep reviews flowing
Every page you publish should pick up at least one external link from a domain you don't control within a few days of going live — its job is simply to signal "this is real, someone linked to it on purpose, don't cut it." For the domain as a whole, a handful of hyper-local trust links — the chamber of commerce, a sponsored youth team, a local charity — lift everything. And keep reviews coming with the situational detail AI loves: ask customers to "tell us what happened," not just "leave a review," so the review carries the specifics ("financed with a 590 score," "same-day alternator") that an AI matches to a shopper's exact prompt.
6. Make every surface tell one story
The single biggest reason a real, established dealership underperforms in AI search isn't a missing tactic — it's contradiction. The website says one set of hours, the Google profile says another. The site lists a service the profile doesn't mention. The phone number on the contact page is a tracking number that doesn't match the listing. Each mismatch is a small reason for the engine to distrust that all these surfaces describe the same real business, and AI engines are conservative — they cite the entity they're confident about. Your job is to make the website, the Google Business Profile, the reviews, the photos, and the service area all map cleanly to one entity, with nothing contradicting anything else. It's tedious, unglamorous reconciliation work, and it's often worth more than any amount of new content.
How AI agents and GEO reinforce each other
There's a compounding loop here that's easy to miss. Your GEO work makes your store the answer an AI gives a shopper. The shopper then lands on your page — where your AI sales agent is waiting to answer the follow-up and book the appointment. Visibility without an agent wastes the traffic; an agent without visibility has no traffic to work. Run both and each makes the other more valuable: more shoppers find you, and more of the ones who do convert.
Measure the right thing
Don't accept an impressions report. The metric that matters is whether the work produces visible results you can verify: a monthly rank map that gets greener, calls and booked appointments that trace back to AI and search, and a clean Google Search Console "Pages" report with no spike in "crawled — currently not indexed." That last one is your early-warning system; a rising "not indexed" count means pages are being purged, and it's the signal to fix structure before you lose visibility. Real outcomes, every month, that the owner can see — that's the standard.
Is GEO different from SEO?
It's the next layer of it. The fundamentals overlap heavily — surviving the index cut is the same work that earns AI citations — but GEO adds emphasis on entity coherence, schema specificity, and content that's genuinely non-redundant.
Do I need an llms.txt file or special AI markup?
No. Google has said these don't matter. Do the real work: distinct local content, correct LocalBusiness schema, a clean mobile structure, and real links.
How long does it take to see results?
Foundational fixes can help within weeks as pages get re-indexed; entity and link authority compound over months. We report a monthly rank map so you can watch it move. See our GEO service.
Want your store cited by the AI? Book a demo and we'll audit where you stand today — or read how the on-page agent converts the traffic GEO sends you in our piece on AI agents for dealerships.