When an assistant answers a shopping question, the answer can come from two very different places. Understanding which is which tells you why simply having a good website is not always enough to appear inside AI shopping.
Two ways your car can show up
- Live retrieval. The assistant reads web pages in the moment a shopper asks, and quotes what it can parse. This rewards readable, structured listing pages. It is the subject of most of this guide.
- Catalog ingestion. The assistant answers from a product index it built ahead of time by ingesting structured feeds from sellers. When a shopper browses cars inside the assistant, the results often come from this catalog, not a live crawl.
To be everywhere a shopper might look, you want both: pages the live crawlers can read, and a feed in the catalogs.
The major AI shopping surfaces
These programs are evolving quickly and rolling out on their own timelines, so treat specifics as directional. The shape, though, is consistent: each wants a structured inventory feed in its own format.
- Microsoft Copilot has been building shopping experiences around a structured commerce feed, the lane that puts products into Copilot answers.
- OpenAI / ChatGPT has introduced shopping and product surfaces that draw on merchant data, with feed-based programs expanding over time.
- Google connects vehicle and product feeds (through its merchant and vehicle listing programs) into search and AI experiences.
What a vehicle feed needs
Each surface has its own format, but the underlying requirements rhyme. A catalog needs, per vehicle: a stable identifier (the VIN), clear title and specs, an accurate price and currency, a real availability state, photos, a canonical link to your listing, and an unambiguous seller identity. The price in the feed must match the price on your page; mismatches get listings rejected or distrusted.
The one honest caveat for dealers
Shopping catalogs are shared by every seller. This is the single surface where your vehicle can sit near another dealer’s, because that is how a shared index works. Even there, a correctly built feed carries your own link and seller attribution, so the inquiry still routes to you. Your own listing pages and any agent endpoint stay free of competitors; the trade-off is specific to the shared catalogs, and it is usually worth making to be in the index at all.
Why this is operationally hard to do yourself
Maintaining a separate, correctly-formatted, always-fresh feed for each surface, in three different schemas, refreshed as prices and availability change, and re-published as each program opens or changes its spec, is real ongoing work. It is exactly the kind of plumbing that gets out of date the moment it is set up manually.
VIN Index generates and maintains these outbound feeds from the one inventory feed you already send, formats each the way the surface requires, keeps your attribution attached, and publishes to each program as it opens to feeds, refreshed roughly every four hours. To see where your inventory stands today across the readable and structured checks, run the free Analyzer.