Docs

How does an online store get its products recommended by AI?

An online store gets its products recommended by AI by making every product readable: crawlable product pages with price, stock and reviews in text, Product structured data, a Merchant Center feed for Google, and buying guides that answer the questions shoppers ask. Assistants recommend products whose facts they can read and check.

Facts

What does an AI assistant need from a product page?

It needs the facts a shopper compares, in text it can read without running scripts:

  • Name, brand and model, exactly as people search them
  • Price and currency, stock and shipping
  • Specifications that answer "is it right for me": size, material, compatibility
  • Real reviews and ratings
  • Return policy

Which structured data should an online store add?

An online store should add Product markup with an Offer on every page where people can buy, and keep it equal to the visible price and stock.

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Trail running shoe X2",
  "image": "https://www.example.com/x2.jpg",
  "offers": {
    "@type": "Offer",
    "price": "89.00",
    "priceCurrency": "USD",
    "availability": "https://schema.org/InStock"
  }
}

Are product pages enough, or does a store need guides?

Product pages answer "what is this and what does it cost"; buying guides answer the questions asked before that, such as which shoe for flat feet or which blender for a small kitchen.

Assistants answer those earlier questions too, so a store with guides that link to its products has more pages that match what shoppers ask.

How does easyIndex help an online store?

If the store's code is in a repository your coding agent can reach, easyIndex gives the agent one page at a time over MCP (buying guides, category answers, markup fixes), audits each one, checks it live, notifies Bing through IndexNow and measures visits from AI assistants.

Step by step

  1. Let AI crawlers in. Allow OAI-SearchBot and the other AI crawlers in robots.txt and check the CDN does not block them.
  2. Render product facts in HTML. Price, stock, specs and reviews must be in the page, not loaded only by script.
  3. Add Product markup. Add Product and Offer structured data matching the visible price and stock.
  4. Send a Merchant Center feed. Upload the catalogue to Merchant Center and opt into free listings.
  5. Publish buying guides. One guide per question shoppers ask, linking to the products that answer it.
  6. Measure AI visits. Track which product and guide pages receive visits from AI assistants.

Common mistakes

  • Markup that shows a different price or stock from the page.
  • Blocking AI crawlers at the CDN while robots.txt allows them.
  • Copying the manufacturer's description, identical on every other store.
  • Loading prices and reviews only with JavaScript.

How easyIndex solves it

easyIndex costs US$ 30 per website a month, with a 7-day free trial: your coding agent builds the guides and fixes the markup over MCP, and easyIndex audits each page, checks it live and measures AI visitors and AI crawler reads.

Frequently asked questions

Can I pay to have my products recommended by ChatGPT?

This guide quotes no paid placement in ChatGPT's answers. What a store controls is whether its pages can be read and what they say.

Is a Merchant Center feed enough for AI?

It helps on Google. Other assistants read web pages, so the facts must also be on the product pages.

Do I need a page for every product variant?

Not always. A variant needs its own page when people search for it by name, such as a specific size or colour.

Does GEO work for marketplace sellers?

Only on pages you control. easyIndex works on a website whose code your coding agent can change.

Continue reading

Last verified on .

GEO for ecommerce: get your products recommended by AI