AI search visibility for ecommerce

A shopping answer can only repeat a price, a name, and an availability that a crawler could read. Google documents that data as Product structured data and, for pages where the shopper can buy, as merchant listings. Google does not publish a rate for how often an AI answer cites a product page. CiteScore does not query Google, an AI Mode shopping result, or any other engine.

AI Overviews and AI Mode are still Search. The eligibility rules are on Google AI Overviews optimization. Markup for pages that are not a product is on schema markup for AI search. The crawler tokens are on AI crawlers and how to get cited by AI answer engines.

Two Product markups

Google says adding the merchant-listing properties can also make the page eligible for product snippets, and that the two features overlap. Provide the fields you actually sell with. Search experiences that show them — a product snippet, a shopping knowledge panel, annotated images — are listed as examples, and Google says the list is not exhaustive.

Fields a buying page has to include

For a merchant listing, Google requires name, image, and offers on the product. The offer needs an active price greater than zero and a priceCurrency in ISO 4217. A product snippet can omit a price; a merchant listing cannot. availability is recommended. When you include it, use one ItemAvailability value, such as InStock or OutOfStock, not several.

Google’s technical guidelines add limits that change what you mark up:

• Only a page where a shopper can purchase the product is eligible for merchant listings. A page of links to other stores is not.
• The page focuses on one product, or variants of that product. “Shoes in our shop” is not a product. Prefer the product URL over the category URL.
• One currency per URL. A price in USD and a price in CAD are two URLs.
• Image URLs have to be crawlable, and the pictures have to show the product.
• JavaScript-rendered Product markup can make Shopping crawls less frequent. Price and stock change quickly. If the HTML response already contains the JSON-LD, the crawler does not have to execute the script to read them.

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Example ceramic mug",
  "image": ["https://example.com/photos/mug-1x1.jpg"],
  "description": "A 12 oz ceramic mug sold on this page. Example markup only.",
  "offers": {
    "@type": "Offer",
    "url": "https://example.com/products/ceramic-mug",
    "price": "18.00",
    "priceCurrency": "USD",
    "availability": "https://schema.org/InStock"
  }
}

That block is an example. Do not add aggregateRating or review unless the page shows a rating you can support. Shipping and return policies can live once on the Organization, and a single product overrides them only when that product differs. Google recommends a Merchant Center feed as well as the on-page markup. Some experiences combine the two, and a snippet may use a feed price when the page has none.

What an answer can quote

The AI features guide says important content should be available as text, structured data should match that text, and Merchant Center should stay current if you use it. The optimization guide says there is no special schema that makes a page appear in Google’s generative Search features. Product markup is still how Google documents price and availability for shopping experiences in Search. A page that cannot be indexed, or cannot show a snippet, is not eligible as a supporting link in AI Overviews or AI Mode.

Google-Extended does not control those Search features. It controls Gemini training and grounding in Gemini Apps. That split is on how to get cited by Gemini. A store that also wants Copilot has to be crawlable by Bingbot, which is a different token, covered on how to get cited by Microsoft Copilot. Answer engine optimization is the name of the work, not a second feed.

Check the public files

The free check reads JSON-LD on the homepage. A product URL at another path is outside that fetch. Submit the product URL when that page is the one you need read.

Run the free check on the host. It reads the homepage, robots.txt for GPTBot, PerplexityBot, and Google-Extended, /llms.txt, a sitemap, and JSON-LD on the homepage. It does not log into an answer engine. One URL continues at checkout.

Sources

• Introduction to Product structured data. Snippets versus merchant listings, and Merchant Center.
• Merchant listing structured data. Required name, image, offer, price, and the single-product rules.
• AI features and your website and optimizing your website for generative AI features. Text, matching structured data, and Merchant Center.