AEO for car dealers: answer engine optimization that a GM can verify

Answer engine optimization is the work of getting your store into the answer instead of the results. SEO earned you a blue link. AEO earns you a sentence in the response. Here is what the work actually consists of for a dealership, in the order it pays off.

First, the checks that take a day

Unblock the crawlers. Confirm robots.txt does not disallow GPTBot, ClaudeBot, PerplexityBot, or Google-Extended. Fewer than 1% of franchise stores block them in robots.txt, but more than a third block automated requests at the edge, which is worse because it is invisible to the store.

Fix the entity. One name, one phone, one address, everywhere: homepage HTML, schema, Google Business Profile, DealerRater, Yelp. Engines resolve your store to an entity. If the signals conflict, you lose the match.

Serve an llms.txt. A plain-text file at /llms.txt describing the store and pointing to the pages that matter. 84% of Dealer.com stores now have one because the platform shipped it. Most other platforms have not. Template and rules in llms.txt and JSON-LD for dealers.

Then, the content that engines quote

AI answers are built from sentences the engine can lift. Dealer sites are built from hero images and sliders. The fix is a small set of pages written to be quoted:

• A store page that says, in plain sentences, who you are, where you are, what you sell, when you are open, and which towns you serve.
• A service page that does the same for the service department, including the towns. Service prompts are the biggest visibility gap in every market we have measured.
• FAQ blocks with schema for the questions shoppers actually ask: financing with bad credit, trade-in process, EV charging, loaner policy.
• Inventory pages that expose price, mileage, and key options as text, not only inside a widget.

Then, the agent layer

This is what separates a store in 2026. A machine-readable inventory feed at a known path. An AI discovery file. A per-rooftop MCP server so an agent can query inventory and hours directly. See MCP for car dealers. Across the stores we have scored, 0.4% have this. It carries more than half the readiness score because it is the difference between being described and being usable.

Then, the sources you do not own

Engines cite DealerRater, Carfax, Cars.com, KBB, and Yelp more often than dealer domains. Claim every profile. Respond to reviews. Push your happy customers to the profiles the engines read, not only to Google.

How to know it worked

Measure before and after with the same prompts, engines, and repetitions. If a vendor cannot show you the prompts and the raw answers, you are buying a screenshot. Our methodology is public so you can hold us to it.

Is AEO different from GEO?

Mostly vocabulary. GEO, generative engine optimization, usually emphasizes content the model can cite. AEO usually emphasizes the structured answer. The work overlaps almost completely for a dealership.

Can my website vendor do this?

Some of it. Platforms control schema and llms.txt and are rolling them out at different speeds. None of the major platforms ships a per-store MCP server today.