Structured Data (Schema) for Ecommerce and AI Search

What is structured data and what is it for in 2026?

A spec sheet for your page written for machines: it doesn't lift rankings on its own, but it unlocks richer results.

Structured data is a standardized description of a page's content, written so a search engine can read it without guessing. A product page says in plain text “Single-origin coffee, 500 g, $12”; the markup tells Google which part is the name, which is the price, in which currency and whether units are available. The most widely used vocabulary is schema.org, and the format Google recommends is JSON-LD, a code block that sits in the page without changing what visitors see.

What it is for in 2026, according to Google: understanding the page's content and making it eligible for rich results, such as price and availability under the title or star ratings. Eligible isn't guaranteed: Google warns that markup enables a feature but doesn't ensure it appears.

What it doesn't do is boost rankings by itself. Its effect is indirect: a result with visible price and stock stands out on the page and attracts better-qualified clicks, because the visitor already knows what it costs.

Which schema types does an online store need?

Product with Offer on every product page, BreadcrumbList across the catalog and Organization for the site. The rest is optional.

An online store needs, in this order: Product with an Offer on every page where something can be bought, BreadcrumbList for navigation and Organization for the company's identity. Google separates two uses of product markup: product snippets, for pages where people can't buy directly, such as editorial reviews, and merchant listings, for pages where customers purchase. Meeting the merchant listing requirements generally makes a page eligible for product snippets too.

The table sums up which types matter for ecommerce, whether they produce a Google rich result today and what they add for AI search:

Are FAQ and HowTo still worth it after Google's changes?

As Google rich results, no. As well-written visible content, yes.

Not as rich results. HowTo stopped appearing in 2023, first on mobile and from September on desktop too, and Google removed its documentation. FAQ was restricted in 2023 to well-known government and health sites, and on May 7, 2026 Google stopped showing it for everyone; in June 2026 it removed the documentation and Rich Results Test support, and scheduled the end of Search Console API support for August 2026.

It isn't the only cut. In June 2025 Google announced it would phase out several types to simplify the results page, including book actions, course info, estimated salary, ClaimReview fact checks, learning videos, special announcements and vehicle listings.

Should you delete FAQ markup? There's no rush: it causes no penalty, it simply produces nothing visible on Google anymore. What is worth keeping is the FAQ section itself, written as direct question and answer, because it is still useful for readers and easy to quote for any system that reads the page.

Does structured data help you appear in ChatGPT and AI Overviews?

Google says it isn't required for its AI features; other engines haven't officially confirmed how they use it.

For Google, the official answer is clear: structured data isn't required to appear in AI Overviews or AI Mode, and there is no special schema.org markup for those features. Google does recommend keeping it as part of your SEO strategy because it makes pages eligible for rich results, and it asks that markup match the visible text on the page.

Where Google gets specific is ecommerce: its AI optimization guide says generative answers can include product listings and product information, and that Merchant Center and its feeds help your products show up in both AI responses and the rest of Search. For a store, the Merchant Center feed matters as much as on-page markup, and both must say the same thing.

For ChatGPT, Perplexity and other assistants, there is no official documentation on how they use schema. The sensible approach is to treat it as what it is: an unambiguous statement of price, currency, stock and brand that any system reading your source code can use. Useful and cheap, but not a guaranteed lever. For the rest of the signals, see our guide to GEO and AEO.

How do you implement it on WooCommerce and Shopify?

Both platforms already output basic product schema: your job is to complete it and remove duplicates.

WooCommerce has generated JSON-LD product markup in its core since version 3.0, through the WC_Structured_Data class. Shopify offers the structured_data Liquid filter, which turns a product into a Product object, or a ProductGroup when it has variants; whether your theme uses it depends on the theme. In both cases the defaults usually fall short: brand, GTIN or shipping and return policies are missing.

The most common mistake we find is duplication: the theme outputs one Product, an SEO plugin outputs another and a reviews app adds a third, each with different data. Check what's there before adding anything. Then make sure the final markup includes these properties, the ones Google requires or recommends for merchant listings:

• name and image: required. Crawlable, high-resolution images showing the actual product.
• offers with a price above zero and priceCurrency as an ISO 4217 code, such as USD or COP.
• availability with a single value, such as InStock or OutOfStock, kept in sync with your inventory.
• description, brand.name and sku; a GTIN such as gtin13 if the product has a barcode.
• aggregateRating and review only if the reviews come from real customers and are visible on the page.
• Shipping and return policies: Google recommends declaring them once under Organization, with hasMerchantReturnPolicy and a shipping policy, rather than repeating them on every product.
• ProductGroup with its variants if you sell different sizes, colors or formats.

How do you validate your structured data?

Rich Results Test for Google, Schema Markup Validator for schema.org and Search Console reports for the whole site.

To validate a page's structured data, paste the URL into Google's Rich Results Test: it shows which types it detects, whether the page is eligible for rich results and which properties are missing or broken. The Schema Markup Validator checks markup against the full schema.org vocabulary, including types Google doesn't use for rich results.

For the whole site, use Search Console. The enhancement reports show, per type, how many pages have valid markup, warnings or errors, and the URL Inspection tool shows the HTML Googlebot actually received. If valid items drop without a matching rise in errors, a theme or plugin update has most likely removed the markup.

Finally, the rule most often broken: markup must match what is visible. A price that differs from the page, made-up reviews or stock that doesn't exist can lead to a manual action. According to Google, that action removes rich result eligibility but doesn't affect how the page ranks in web search.