Table of Contents▾
Short answer: AEO and GEO are how a dealership gets named inside AI answers from Google's AI Overviews and AI Mode, ChatGPT, Ask Maps and Siri. Google says it is still SEO, and the evidence says the sources AI trusts most are the ones you already control: your website, your Google Business Profile and your reviews. Rewritten October 21, 2026, with every number linked to its source.
One in five car buyers used an AI site or an AI-generated overview while shopping in 2025, and one in four new-vehicle buyers did, according to Cox Automotive's 2025 Car Buyer Journey Study, a survey of 2,344 buyers fielded from August 6 to September 5, 2025. By mid-2026, 63 percent of in-market shoppers told Cox's AI in Auto Retail Tracker they will definitely or probably use AI on their next vehicle purchase. The systems they are asking are very large. At I/O in May 2026, Google reported AI Overviews at more than 2.5 billion monthly users and AI Mode at more than 1 billion, with AI Mode queries more than doubling every quarter. OpenAI says ChatGPT reaches 1.2 billion people a week, up from 900 million in February.
Two terms, then the work. Answer engine optimization (AEO) is making a page the best self-contained answer to a question a shopper asks, so an engine can quote it. Generative engine optimization (GEO) is being the dealership an AI recommends when the question is 'which dealer', which rests on your listings, reviews and mentions more than on any single page. Google's own guide, published May 15, 2026, says optimizing for generative AI search 'is optimizing for the search experience, and thus still SEO.' We first published this guide on February 21, 2026 and rewrote it on October 21 because enough had changed to make parts of the old version wrong. Where no credible number exists, it says so rather than borrowing one.
What Changed in AI Search Since February 2026
Here is the dated record of what shipped between this guide's first version and this rewrite, with what a dealer should do about each item. Every row links to the primary announcement.
| Date (2026) | What changed | What a dealer does |
|---|---|---|
| Feb 9 | OpenAI began testing ads in ChatGPT for U.S. users on the Free and Go tiers. OpenAI says ads do not influence answers. By October 21 the program covered 89 countries and more than 100,000 advertisers. | Treat ChatGPT as a paid channel as well as an organic one, and budget a small test. |
| Mar 12 | Google Maps added Ask Maps, a Gemini conversational search over 300 million places and reviews from 500 million contributors, in the U.S. and India. | Your Business Profile attributes, hours and reviews are now answers, not just a listing. |
| Apr 14 | Apple Business replaced Apple Business Connect, free, with data migrating automatically. Apple's next foundation models are built with Google's Gemini, and Siri AI, shipped September 14, answers local questions, such as the best coffee spot in a new city, from information across the web. | Claim and complete the Apple Business place card. Siri reads it. |
| Mid-April | Google Merchant Center began flagging vehicle-feed data problems (price, mileage, VIN, condition, landing-page mismatch) as account issues. | Make the feed and the vehicle detail page agree, every day. |
| May 7 | FAQ rich results stopped appearing in Google Search for all sites. The documentation was removed June 15. | Keep FAQ pages for shoppers. Stop expecting a rich result. |
| May 15 | Google published its guide to optimizing for generative AI features: still SEO, no special files or markup, pages must be indexed and snippet-eligible. Spam policies now cover AI responses, and pages built for every query variation count as scaled content abuse. | Delete the page-per-prompt plan. One useful page per real topic. |
| Jun 3 and Aug 31 | Search Console added a Generative AI performance report (impressions, pages, countries, devices), first for a subset of sites, then for every site worldwide on August 31. | Baseline your AI impressions now so you can tell later whether anything you did worked. |
| Aug 19 | Generative UI: Google builds interactive tools inside AI Mode answers, now rolling into AI Overviews. | Generic calculators now live inside the answer. Publish what the engine cannot make up. |
| Oct 15 | Google said it has increased inline links in AI Mode and AI Overviews and is testing a dedicated section of official links at the top of AI answers. | The brand's own site gets a dedicated slot. Make sure yours is the canonical, fast, crawlable one. |
Why AEO and GEO Matter for Dealers in 2026 and Beyond
What shoppers are doing
Start with how people shop. Cox's 2025 study found buyers spent about seven hours researching and shopping online, more than they spent at the dealership where they bought. Third-party sites were used by 75 percent of buyers, dealership sites by 59 percent, search engines by 41 percent, and 'AI sites', tracked for the first time, by 12 percent. By Cox's mid-2026 tracker, AI tools (36 percent) nearly matched automotive websites (38 percent) as a research channel.
Mostly, shoppers use AI to get ready for the visit. In the same tracker, 24 percent said AI helps them feel more prepared when working with a dealership, 26 percent of AI users use it to generate questions to ask dealers, and 17 percent use it to avoid dealership staff. Urban Science and The Harris Poll found 44 percent of buyers spend more time researching before contacting a dealer than they did a year ago. CarGurus found 36 percent of buyers and sellers want AI to summarize dealership reviews for them. And the purchase still ends in the store: in Cox's study, 53 percent completed every step at the dealership and only 7 percent bought completely online.
The local-search habit moved fastest. BrightLocal's 2026 consumer survey found 45 percent of consumers now use ChatGPT or another generative AI tool for local business recommendations, up from 6 percent a year earlier, with ChatGPT used by 31 percent and Google's AI Mode by 23 percent. SOCi's 2026 Local Discovery Index put it at 52 percent who used an AI tool to find a local business in the past 30 days, up from 9 percent, and found 67 percent of those users had been given wrong information about a local business at least once. A wrong answer about your hours, your inventory or your financing is now a measurable risk, and most of the time the cause is a listing nobody has updated.

The gap is the point. Cox's tracker found only 29 percent of dealers actively adapting for AI-powered search in Q2 2026, down from 32 percent in Q1, while the share who know they need to act but have not started rose from 26 to 32 percent. Meanwhile 82 percent of dealers say they use AI somewhere in the store, mostly for task automation, follow-up and content. The store is automating its own work faster than it is showing up where shoppers ask.
What happens to the click
Pew Research Center watched 900 U.S. adults make 68,879 Google searches in March 2025. When an AI summary appeared, users clicked a traditional result on 8 percent of visits, against 15 percent when no summary appeared, and they clicked a source cited inside the summary on just 1 percent of visits. They were also more likely to end their browsing session on that page: 26 percent versus 16 percent.

Industry datasets agree on the direction. Ahrefs found an AI Overview now correlates with a 58 percent lower click-through rate for the top-ranking page, comparing December 2023 with December 2025 across 300,000 keywords. Semrush found 92 to 94 percent of AI Mode sessions ended without a click to any external site in mid-2025 desktop data, and Similarweb puts the zero-click share of all Google searches at 68 percent, with AI Overviews appearing on more than 20 percent of queries. Semrush's October 2026 study found AI Overviews now cover 88 percent of commercial-intent searches in its tracked categories, up from 46 percent a year earlier.
Two things keep this honest. Graphite's analysis of the 40,000 largest U.S. sites found organic search traffic fell only 2.5 percent year over year through late 2025, with the losses concentrated in news, health, cooking and entertainment sites, and Google says total organic click volume has been 'relatively stable year-over-year', with AI features sending billions of clicks to websites every week. Set that beside the forecast dealer-facing posts still quote as if it had happened: in February 2024 Gartner predicted traditional search volume would drop 25 percent by 2026 as chatbots took share. What arrived was a 2.5 percent dip, with AI Mode and ChatGPT growing on top of search. Plan for a landscape that is bigger and more fragmented than the one you had.
What we can and cannot prove
Here is the sentence every vendor deck wants and no study supports: 'dealers that did AEO sold X percent more cars.' We looked. Cox's tracker is the best third-party data and it points the other way: 69 percent of dealers expected AI to drive sales and revenue growth, 22 percent report seeing it, and about one in three either are not measuring AI's impact or cannot say how they measure it. Urban Science and The Harris Poll found 51 percent of dealers saying AI has contributed to higher sales, which is perception, not a measured lift.
What the data does support is a visibility-and-click case. Seer Interactive's analysis of 53 brands and 5.47 million queries found a brand cited inside a Google AI Overview earned about 120 percent more organic clicks per impression than uncited brands on the same queries, and organic click-through on AI Overview queries recovered from 1.3 percent in December 2025 to 2.4 percent in February 2026. Adobe's retail data showed AI-referred visitors converting 42 percent better than other traffic by March 2026 and 60 percent better by July, though that is retail, not automotive. The '9x conversion' figure that circulates in dealer marketing traces to a single Seer client site where ChatGPT visitors converted at 15.9 percent against 1.76 percent for Google organic, on AI traffic that was 0.07 percent of the site's organic volume. One site, a tiny base, and a hint at most.
What a dealer can actually own today is being cited, being clicked from the citation, and being contacted by a shopper who checked the answer against your site. The rest of this guide works backward from those three.
Foundations: Local SEO as the AEO/GEO Bedrock
The most useful number in local AI search comes from Yext's analysis of 6.8 million AI citations about local brands across ChatGPT, Gemini and Perplexity: 86 percent came from sources the brand already controls, 44 percent from its own website and 42 percent from its listings, with reviews and social supplying 8 percent. BrightLocal's study of 1.9 million citations across ChatGPT, Google AI Mode and AI Overviews found the same shape: business websites 42 percent, Google Business Profile 28.5 percent, Yelp 9.5 percent, Facebook 2.2 percent. Most of what an engine says about a dealership, it read on that dealership's own site or listings.

BrightLocal also found the three AI surfaces barely agree on which businesses to name for the same local query: AI Mode and AI Overviews share only 29 percent of the businesses they name, and ChatGPT overlaps just 19 to 20 percent with either Google surface, and Profound's study of 680 million citations shows each engine leans on different sources: Wikipedia for ChatGPT, Reddit for Perplexity and AI Overviews. The same dealership can be ChatGPT's first pick and absent from AI Mode. So the listings have to be complete on every surface; none of them carries the others.
Google Business Profile
Your profile is now an answer source for Ask Maps, which Google says analyzes information from more than 300 million places, including reviews from its contributors, and for AI Overviews and AI Mode, which Google's guide says draw on Business Profile and Merchant Center data. Two things follow. Fill every attribute honestly: departments, hours by department, payment options, languages, accessibility, and the services a shopper might ask about by name, because a question such as 'open Sunday' or 'works with bad credit' is answered from those fields. And stop expecting your inventory to appear on the profile. Google retired organic vehicle listings on Business Profiles and in Search during 2025, phased the Vehicle Listing structured data out of results that summer, and dropped it from Search Console reporting in September 2025. Inventory visibility on Google now means Vehicle ads through Merchant Center plus crawlable vehicle pages on your own site.
Apple, Bing and the listings AI reads
Three listings most dealers ignore now sit upstream of AI answers. Apple Business, which absorbed Business Connect on April 14, 2026, is where Apple says your brand and locations are managed for Maps, Siri and Wallet, and Siri AI, in English beta on current devices since September 2026, pairs that with information from across the web to answer questions such as the best local spot in a new city. Copilot answers from Bing's index, and a verified site in Bing Webmaster Tools (free; the measurement section covers the report) is how you see what it cited. And Yelp now supplies its reviews, ratings, photos and business details to ChatGPT through a content integration with OpenAI, so what ChatGPT says about your store is partly what Yelp says about it. Claim all three and keep them current. Name, address, phone and hours must match character for character across Google, Apple, Bing, Yelp and your own site, because an engine that finds a conflict either picks one at random or skips you.
Hyperlocal pages that are not spam
City pages still work when each one earns its place. Deloitte's 2026 Global Automotive Consumer Study ranks the dealer website among the top three research sources for U.S. shoppers, used by 40 percent, behind only a dealer visit and the manufacturer site. A page for a neighboring city should carry things a shopper there cares about: drive time and directions, the models you stock most for that market, the lenders you work with, service pickup areas, and real photos of the store. What no longer works is the page-per-variation playbook. Google's May 2026 guidance says creating separate pages for every query variation 'primarily to manipulate rankings or generative AI responses' violates its scaled content abuse policy. One genuinely useful page per real topic, and one per real place.
Technical performance
AI retrieval is ordinary crawling, so the ordinary rules apply. Google's Core Web Vitals thresholds are a Largest Contentful Paint of 2.5 seconds or less, an Interaction to Next Paint of 200 milliseconds or less, and a Cumulative Layout Shift of 0.1 or less, measured at the 75th percentile of real visits; the Core Web Vitals report in Search Console tells you where you stand. Preview controls matter more than they did: Google says nosnippet, data-nosnippet and max-snippet limit how your content appears in AI features, so audit your templates for a max-snippet tag a plugin added years ago. Do not block OAI-SearchBot in robots.txt; OpenAI's crawler documentation says it is the bot that surfaces sites in ChatGPT search, while GPTBot is the separate training crawler you may block without losing visibility. And build for phones first: CarGurus found 54 percent of shoppers use their phone more than a computer when shopping. If your current site cannot pass these checks, Get My Auto's dealer websites are built to.

AEO for Car Dealerships: Answer First, Sell Second
Answer engine optimization is about being the quotable answer. The engines quote passages rather than whole pages, so a page earns a citation when one of its passages answers a sub-question completely on its own.
Turn real questions into content
Start from what shoppers actually ask AI. The Cox numbers above say they use it to draft questions for the dealer and to walk in prepared. In CarEdge's 2025 survey, 88 percent of buyers who used AI found it helpful, and the top uses were research, price checks and negotiation prep. Those are your page topics: what a fair price looks like for the cars you stock, what a brake job or a timing belt costs at your store, what documents a buyer needs, how your financing works for a 580 credit score, how your trade-in process works. Pull the questions from your BDC's call notes, your chat transcripts and your service advisors, then write one honest page per question that a real shopper would thank you for. Forty near-duplicate pages chasing forty phrasings is the pattern Google now penalizes, as the city-pages section above explains.
Lead with the answer, as a self-contained passage
The old advice in this guide was 'the first 100 to 150 words matter most.' Retire it. Google's AI Mode uses query fan-out, breaking a question into subtopics and issuing many searches at once, and Google says the average AI Mode search is triple the length of a traditional search query. The engine is looking for the passage that answers one sub-question completely, wherever it sits on the page. Ahrefs' study of ChatGPT citations found cited pages had titles closer to the prompt's wording, and that URLs with natural-language slugs were cited 89.8 percent of the time versus 81.1 percent for slugs without words. In practice, put the question in the heading, answer it in two to four sentences that make sense with nothing else on the page, then give the detail, then the next step.
Take a made-up 60-car independent store, Northgate Motors in Fresno (illustrative, not a client), and its brake service page. Before: 'Northgate Motors offers premium brake services performed by certified technicians using quality parts. Contact us today for a quote!' Nothing an engine can quote, no price, no time. After: a heading that reads 'How much does a brake job cost at Northgate Motors?' followed by three sentences: 'A front brake pad and rotor replacement at Northgate Motors in Fresno runs $320 to $480 for most sedans and small SUVs, parts and labor included, and takes about 90 minutes. Trucks and European models cost more, and we quote them from the VIN. Book online and the price is locked before you arrive.' Then the detail: what is inspected, the warranty, the loaner policy, and the schedule button. The three-sentence block is what a fan-out query like 'brake job cost Fresno' retrieves, and it answers the price, the scope and the time in one breath. The prices are invented for the example; yours should be real.
Build FAQ hubs for shoppers, not for rich results
FAQ rich results stopped appearing in Google Search on May 7, 2026, for every site, and the documentation was removed on June 15. So why keep FAQ pages? Because shoppers read them and engines retrieve from them. The format is still the cleanest way to publish a self-contained answer. Keep the FAQPage markup if you like, it is valid vocabulary, but write the page for a person: group questions by the moment they arise (financing, trade-in, service, delivery), keep each answer complete on its own, and date the page. Engines favor fresher pages, and the local content section below has the numbers.
Use structured data for what it still does
Google's guide says there is 'no special schema.org markup you need to add' for AI features, and warns against over-focusing on structured data. Markup still helps machines read the facts you want them to read, as long as it matches what is visible on the page. Here is where each type stands.
| Type | Status in 2026 | Use it for |
|---|---|---|
| AutoDealer (a LocalBusiness subtype) | Current. Google asks for the most specific subtype. | Address, geo, opening hours by department, phone, sameAs links to your profiles |
| Vehicle or Product on vehicle pages | Vehicle-listing rich results and reporting retired in 2025. | Machine-readable price, mileage, VIN and condition that match the Merchant Center feed |
| FAQPage | Valid vocabulary. The rich result ended May 7, 2026. | Structure only. Expect no rich result. |
| HowTo | Rich result deprecated September 13, 2023. | Nothing. Remove it. |
| QAPage | Only for pages where users submit answers; Google says not for FAQ pages. | Nothing on a dealer site. |
| BlogPosting and Article | Current. | Headline, author, datePublished and dateModified on every guide you publish |
Practical GEO for Dealers: Building Share of AI Voice
Generative engine optimization is about the 'which dealer' question. Semrush's analysis of 126 million U.S. prompts found ChatGPT cites about 15 sources per response and Gemini about 3, and Similarweb found automotive prompts in ChatGPT carry citations about 20 percent of the time, roughly triple the all-category rate of 6.8 percent. Car questions are among the most heavily cited questions people put to AI, which means most of those answers name sources, and a dealership can be one of them.
Optimize for near-me and local intent
Test the engines in the order shoppers use them for local recommendations, ChatGPT first, then Google's AI Mode, and expect a hard ceiling. SOCi's 2026 Local Visibility Index found ChatGPT recommends only 1.2 percent of multi-location brands' locations, against a 35.9 percent appearance rate in Google's local 3-pack, across more than 350,000 locations at 2,751 multi-location brands, with no automotive figure published. The shoppers who do get a recommendation verify it: Yext's 2026 consumer report found 53 percent search Google or Bing next, 49 percent visit the business website, and only 5 percent act without checking, while 28 percent have tried a new local business because of an AI recommendation. The recommendation gets a shopper moving; your site and your profile are what they check before they call.
Trace one prompt through the stack. A shopper in Fresno asks 'which used car dealer near me works with bad credit and is open Sunday' in Ask Maps, in Siri, and in ChatGPT. Ask Maps answers from Google's data on more than 300 million places and their reviews: it needs your Sunday hours filled in, a financing attribute, and reviews that mention credit. Siri answers with information from across the web, in Apple's words, and Apple Business is where your Maps listing is managed: it needs a claimed Apple Business listing with hours and a crawlable financing page. ChatGPT answers overwhelmingly from live search results plus Yelp's reviews and business details through its OpenAI integration: it needs a page whose title says 'bad credit car loans in Fresno' in plain words, a Yelp profile with hours, and reviews. Northgate Motors, our made-up store, wins all three with a single financing page, complete hours on three listings, and a review request that asks customers to say what they came in for.
Earn third-party mentions that are real
Ahrefs found branded web mentions correlate with AI Overview visibility at 0.664, three times the correlation for backlinks (0.218). Where do the citations come from? In Ahrefs' count of the most-cited domains, YouTube supplies 22.9 percent and Reddit 18.5 percent of Google AI Overview citations among the top sources, ahead of Facebook at 10.1 percent. For a dealership that means a YouTube channel with walkaround videos that name the store and the city, a real presence in the local subreddit (answering, not advertising; our Reddit monitoring guide covers how), local news coverage of what you actually do, and the chamber, league and charity pages that list you. Google's guide warns that seeking inauthentic mentions across the web is not as helpful as it might seem; paid 'best dealers in' listicles are the dealership version of that.
Reviews are retrieved, not trained on
The earlier version of this guide called reviews 'AI training signals.' That was wrong, and the correction changes what you do about it. ChatGPT's search feature fetches pages at answer time (the OAI-SearchBot and GPTBot split from the technical section), and Yelp's reviews, ratings and photos reach it through a data partnership, served through Yelp's APIs rather than baked into the model. So a review posted this week can be in an answer next week, and a bad month shows up just as fast. Seer Interactive found, across 804,491 AI responses, that brands with even a basic Trustpilot profile had a median AI citation rate of 53.5 percent against 1 percent for brands with none.
Consumer thresholds tightened in 2026. BrightLocal found 68 percent of consumers will only use a business rated four stars or higher, 31 percent require 4.5, 74 percent want reviews from the last three months, and 47 percent skip a business with fewer than 20 reviews. Widewail's dealership benchmark, a vendor dataset covering about 18,000 franchise rooftops, puts the median store at 12.3 new Google reviews a month and a 4.66 average rating, with volume down about 15 percent year over year and communication the top complaint, appearing in 44 percent of negative reviews. Ask for reviews after service as well as sales, answer every one within a day, and ask customers to say what they came in for, because a review that says 'they got me approved with a 560 score' is the sentence an engine quotes.
Publish local content that survives the AI answer
Google now builds interactive tools inside AI answers, in AI Mode since late 2025 and in AI Overviews since August 2026, so a 'car payment calculator' page now competes with a calculator the engine draws on the spot. Freshness is rewarded: AI assistants cite pages averaging 1,064 days old versus 1,432 for organic results, and Seer found about 65 percent of ChatGPT crawler hits land on content published within the past year. Publish what AI cannot generate: your real inventory with real prices, your actual service menu, your hours, your lenders, your people, the questions your BDC heard this month, and the dated answer. Put a visible 'updated' date on it and mean it.
Paid placement beside organic GEO: ChatGPT ads and ads in AI Mode
OpenAI began testing ads in ChatGPT on February 9, 2026, matched to the conversation topic, added a self-serve Ads Manager beta with cost-per-click bidding on May 5, and by October 21 counted more than 100,000 advertisers 'from local businesses to global brands' in 89 countries. Google, for its part, put ad formats inside AI Mode and launched Business Agent for Leads, which answers a shopper's questions inside the ad instead of serving a form. A dealer's organic citation and a competitor's paid placement can now sit on the same screen. Our ChatGPT ads guide for auto dealers covers targeting, formats and what to test first.
On-Site Content Architecture for AEO and GEO
Pillar and cluster, one page per topic
Organize the site the way an engine understands a business: one pillar page per thing you sell or do (new, used, financing, service, parts, trade-in) and a cluster of specific answer pages under each, linked in both directions. The pillar says what; the cluster pages say how much, how long, and for whom. Give the site a shape a shopper and a crawler can both follow, and resist spawning variations of a cluster page for the engine's benefit.
Treat inventory pages as content assets
Vehicle detail pages are the pages shoppers most want and engines most often get wrong. Since mid-April 2026 Merchant Center flags vehicle feeds whose price, mileage, VIN, condition or landing page disagree with the page itself, and the same mismatch confuses an AI answer. Keep one source of truth for price and status, publish it to the feed and the page from the same record, and write a real description: trim, drivetrain, notable options, condition notes, warranty, and a sentence about who the car suits.
Add textual detail AI can parse, and photos Lens can read
Since September 24, 2026 Search Console reports multimodal searches (Lens, Circle to Search, image uploads) as their own search type, and Google says more than one in six U.S. searches now use voice or images. A shopper can point a camera at a car on your lot, or in a photo, and ask what it is and who sells it. Give every vehicle photo real alt text (year, make, model, trim, color, your store) and keep the facts in text on the page, not baked into the image.
How Does GEO Measure AI Visibility for a Dealership?
Measurement was the weakest part of this guide in February. It is now the most concrete part, because the platforms shipped reports.
Search Console's Generative AI performance report
Since August 31, 2026 every site has a Generative AI performance report under Search in Search Console. It shows impressions for AI Overviews and AI Mode by page, country, device and date, and nothing about clicks or queries. Two things to know when you read it. AI Mode traffic has been counted inside the regular Web totals since June 16, 2025, so a rising impression count with flat clicks is the expected shape, not a bug. And after September 24 the Web type is split into text-based and multimodal, so compare periods on the same type. First check that the Search generative AI control under Settings says Include, since a site must be included to be eligible for AI features at all. Then screenshot the report. That is your baseline.
Bing Webmaster Tools AI Performance
Bing's report, launched February 10, 2026 and expanded on June 16 with Intents, Topics, Citation Share and Compare, is the only one of the two free platform reports that shows the actual grounding queries behind AI answers that cited you, and it classes local intent separately. Because Copilot answers draw on Bing, it is also the closest free view of what Microsoft's assistant says about your store.
GA4 referrals and what they miss
Add a channel group in Google Analytics 4 for referrals from chatgpt.com, gemini.google.com, perplexity.ai and copilot.microsoft.com. The numbers will be small; ChatGPT's outbound referrals tripled in 2025, but more than 30 percent of them went to ten domains. They will be good ones, since Similarweb found AI referrals to transactional sites converting at about 7 percent and Momentic's analysis of Similarweb clickstream data found ChatGPT users click out 1.4 times per visit versus 0.6 for Google. And they will undercount, because an AI Overview impression that sends a shopper to search your name and tap your listing shows up as organic or direct, never as AI.
Manual prompt checks
Once a month, run the same ten prompts in ChatGPT, AI Mode, Gemini and Siri, logged out, from a phone in your market, and record which dealers are named, what facts are stated about you, and whether they are right. Run each prompt twice: BrightLocal found only 20 to 33 percent overlap between repeat runs of the same local query. With ChatGPT citing about 15 sources per answer, note which of your pages and listings it used. And log the errors, because the fix for most of them is a stale listing you can correct in an afternoon.
What no dealer can measure yet
No platform reports clicks, leads or sales from AI Overviews or AI Mode to a specific site. Anyone promising that attribution is estimating. Measure what exists (impressions, citations, referrals, correct facts) and keep your lead-source question honest: 'How did you hear about us?' with 'an AI assistant' as an option is still the best lead-level signal a store has.
Automotive AEO Companies and AEO Tools for Automotive Brands: What to Buy and What to Skip
The vendor market grew faster than the evidence. Use Google's own guidance as the filter. Skip: llms.txt packages, which Google Search ignores; 'chunking' or rewriting your pages for AI, which Google says is unnecessary; page-per-prompt content programs, which now fall under scaled content abuse; FAQ schema sold as an AI lever, since the rich result ended in May; and any pitch built on the '9x conversion' claim, which is one site, or on the '30 percent of car shoppers use AI, 68 percent of them ChatGPT' statistic, which comes from a 627-person survey of powersports, marine and RV shoppers, not car buyers. A vendor that cites those without the caveats is not reading its sources.
Buy, or do in-house: the two free first-party reports, Search Console's Generative AI report and Bing's AI Performance; a prompt tracker that reports run-to-run variance, because a tool that shows one clean number per prompt is hiding the overlap problem; listings management that covers Google, Apple, Bing and Yelp from one source of truth; feed hygiene between Merchant Center and your vehicle pages; and a review program with response-time targets. Dealers who work with an outside AI partner are more likely to report revenue growth from AI, 36 percent versus 21 percent in Cox's tracker, though that figure is self-reported, covers AI in general rather than search visibility, and comes from a company that also sells AI products. The partner worth paying is the one that does the unglamorous work in this list.
Frequently Asked Questions
The Future of Automotive Search Visibility in an AI-Driven Market
Three things to watch. First, agents. Google already calls stores and checks stock on a shopper's behalf for retail and service categories, and nothing has been announced for dealerships. Prepare anyway, by getting the phone answered and the inventory feed accurate, because when it arrives it will run on the same facts. Business Agent, a branded conversational agent inside Search that eligible U.S. retailers activate in Merchant Center, points the same way: the lead form becomes a conversation. Second, interfaces. Generative UI builds tools inside answers, Google is testing a dedicated section at the top of AI answers for links to official pages and says Preferred Sources users grew more than sevenfold this year, and AI Mode is available in more than 200 countries and territories and 35-plus new languages, so Spanish-language pages are in scope. Third, Siri. Apple's assistant now answers local questions from information across the web, and Apple Business is where the Maps listing it may draw on is managed. Every one of these surfaces answers from facts a dealer already owns.
This guide will be updated as the platforms change; the date at the top is the last time every number was checked against its source. If you want a website and marketing stack built for this, see how Get My Auto connects your website, listings and CRM.
Topics
Share this article
Help others discover this content
Continue Reading
View All
ChatGPT Ads for Auto Dealers: What to Know (2026 Guide)

AI Literacy for Auto Dealers: What Your Team Needs to Know

Reddit Monitoring for Auto Dealers: The 2026 Guide

