Local & Industries · 9 min read · July 15, 2026
Does the AI recommend your dealership? The 12-point checklist for a self-test
More and more car buyers no longer ask Google but ChatGPT or Gemini: "Which dealership near me is reliable?" If your business doesn't appear in that answer, you simply don't exist for these customers. This 12-point checklist shows you in an honest self-test whether the AI knows and recommends your dealership - and what you can concretely do about it.
Why the AI question suddenly matters for car dealerships
The path to buying a car has shifted. Previously, a prospect typed "VW Tiguan used car Regensburg" into Google and clicked through ten blue links. Today the same person asks ChatGPT: "I'm looking for a family-friendly used vehicle up to 25,000 euros around Regensburg, where do I best buy it?" The AI answers with whole sentences, recommendations and, in the best case, concrete dealer names. It is exactly here that it's decided whether your dealership is even in the game at all.
For dealerships this is explosive, because your products are expensive and require explanation. No one buys a 30,000-euro car on impulse. There is a long research phase, and it increasingly runs via AI assistants. Whoever isn't named here as a trustworthy address loses the customer before they even think of a test drive. The tricky part: you don't notice it. There's no lost inquiry, no abandoned cart - the customer was simply never there.
Generative Engine Optimization, GEO for short, is the answer to this. It's no longer only about rankings, but about whether the AI classifies your dealership as recommendable and builds it into its running text. The good news: most of your local competitors do nothing for it so far. Whoever starts now has a real head start.
The self-test: how to check your AI visibility in 15 minutes
Before you optimize anything, you need an inventory. Open ChatGPT, Gemini and Perplexity and ask the same questions your customers would ask. For example: "Which dealership in [your city] is recommendable for buying a used car?" or "Where can I have my [brand] serviced in [region]?" Note down verbatim what comes back.
Pay attention to three things: Are you named at all? Are you described correctly, meaning with the right brand, right location, right services? And in what context are you mentioned - as the first recommendation or only as a footnote next to five others? Do this for at least five typical searches and document the results in a simple table.
Repeat the test with variations: sometimes with the brand name, sometimes without, sometimes with location, sometimes with "near me". AI answers fluctuate, so a single test is worthless. Only the pattern across several queries honestly shows you where you stand. This log is your baseline, against which you measure progress in three months.
Points 1 to 4: the foundation checks
Point 1 - Google Business Profile complete? AI systems draw massively on what Google knows about you. Missing opening hours, no current image, no category "car dealer" plus "car workshop" - all of that weakens you. Point 2 - NAP consistency: name, address, phone number must be exactly identical on the website, Google, mobile.de and business directories. Even "St." instead of "Straße" can cause confusion.
Point 3 - brands and services in plain text on the website. Does it state, spelled out, that you are an authorized dealer for a certain brand, carry new and used cars, run a master workshop, offer roadworthiness/emissions inspection and arrange financing? AI loves unambiguous, written-out facts. Point 4 - name location and catchment area. "Your dealership for Regensburg, Kelheim and Neutraubling" helps the AI assign you to the right region.
These four points sound banal, but this is exactly where most dealerships fail. I constantly see websites where the carried brands appear only as a logo graphic that an AI cannot read at all. Text beats image. If a piece of information is to count for the AI, it must be present as real, readable running text.
Points 5 to 8: trust and evidence
Point 5 - reviews in quantity and quality. AI systems rate reviews as a trust signal. A dealership with 240 Google reviews and 4.6 stars is recommended much more readily than one with twelve reviews. Point 6 - do you reply to reviews? Visible, factual replies, especially to criticism, show the AI as well as people an active, reputable business.
Point 7 - mentions on independent pages. Are you named in local news portals, trade directories, chamber registers or club sponsorship lists? Such third-party sources are often more credible for the AI than your own website. Point 8 - make awards and certificates visible. "Distinguished service business" or master-certificate references belong as text on the page, not just as a seal image in the footer.
The common thread of these four points: the AI looks for evidence that you can be trusted. It cannot experience your sales conversation, so it relies on what others say about you. A dealership sells trust just as much as vehicles - and you have to translate this trust into machine-readable signals.
Points 9 to 12: content the AI quotes
Point 9 - do you answer real customer questions as content? A guide like "Leasing or financing when buying a car - what suits whom?" or "What does a timing belt change cost on a [brand]?" delivers the AI exactly the text modules it builds into its answers. Point 10 - timeliness: a blog whose last post is from 2021 signals a dormant business. Regular, dated content keeps you fresh.
Point 11 - structured data (Schema.org) for AutoDealer, opening hours and offers. This invisible markup in the source code helps machines understand your page unambiguously. Ask your web administrator for it specifically, it's standard craftsmanship. Point 12 - vehicle inventory clean and described. If your used cars stand online with complete textual details instead of only photos, you get found more readily for concrete model queries.
If you honestly tick off these twelve points, a picture quickly emerges. Most dealerships land at four to six fulfilled points. That's no reason to panic, but your map: every open point is a concrete task with a direct effect on your AI visibility.
The most common contradiction: good website, invisible in the AI
Many dealership owners tell me: "But we do have an expensive, modern website." That's often true - and yet the business doesn't appear in AI answers. The reason: beautiful websites are built for people, not for machines. Large image sliders, animated effects and brands as graphics look good but deliver the AI hardly any readable content. Looks and machine readability are two different pairs of shoes.
The second contradiction: reach on vehicle marketplaces like mobile.de doesn't mean AI visibility. There you sell individual cars, but the AI recommends you as a business, as a brand, as an address. You build this business reputation on your own domain, in reviews and in third-party sources - not in the listing form of a platform that wants to own the customer contact itself.
Resolve the contradiction by thinking of your website in two layers: the visible one for the person and the content one for the machine. Both must fit together. A single, clearly written paragraph "Who we are, which brands we carry, which service we offer" often gets you further than the next design update.
From self-test to roadmap: the next 90 days
Take your filled-in checklist and prioritize. In the first two weeks you handle the foundation checks: complete the Google profile, align NAP data everywhere, put brands and services as real text on the website. That's work of a few hours with disproportionate effect, because it lays the basis for everything else.
In the second month you take care of trust: a small campaign to actively ask satisfied customers for reviews, plus consistently replying to existing reviews. In the third month you start with content: two to three honest guide texts on the questions your customers ask you daily in the showroom anyway. Write them the way you speak.
After 90 days you repeat the self-test from the second section with exactly the same questions. Compare the answers with your baseline log. Are you now named where you were missing before? Are you described more correctly? This measurement is decisive, because GEO is not a one-off project, but a cycle of checking, improving and checking again.
Honest conclusion: what you can realistically expect
AI visibility is not a switch you flip. No serious service provider can guarantee you that ChatGPT will name you first tomorrow, because the systems are a black box and change constantly. Whoever promises you that sells you hot air. What you can influence are the signals from which the AI draws its recommendations - and those are largely in your hands.
The effort is worth it above all because your competitors are asleep. In most regions not a single dealership has systematically worked on its AI visibility. Whoever works through the twelve points now occupies the recommendation position before the competition even notices it exists. This first-mover advantage disappears in one to two years.
Start small. You don't have to manage all twelve points in one week. Take a single one this week - for example your carried brands as clear text on the homepage. Then the next. Dealerships are used to thinking in long-term relationships instead of quick shots. It's exactly this attitude that is your biggest advantage with AI visibility.
Common questions
Isn't it enough if my dealership is well visible on mobile.de and AutoScout24?
No. On these marketplaces you sell individual vehicles, but AI assistants recommend you as a business and address. When someone asks "Which dealership near me is recommendable?", the AI draws on your own website, on Google reviews and on independent mentions - not on your listings. Both channels are important, but they serve completely different purposes.
We are an authorized dealer of a large brand. Doesn't the AI recommend us automatically?
Unfortunately not reliably. The brand gives you awareness, but for local queries like "Where do I get service for my [brand] in [city]?" the AI must be able to assign you concretely to the place. That only succeeds if location, catchment area and services stand as clear text on your page and your local trust signals like reviews are right. The brand name alone is not enough.
How often should I repeat the AI self-test?
At the start quarterly, so every three months. That way you see whether your measures are working, without letting the natural fluctuations of AI answers drive you crazy. Each time use exactly the same five to seven questions and document the results. After about a year, once your foundations are solid, a semi-annual check is enough to stay on the ball.
Read on