Structured data for an online store: Product, Offer, Review
Structured data (markup) is extra code on a page that explains to a search engine what exactly is on it: here is a product, here is its price, here is availability, here is the rating. It is most often written in JSON-LD format using the schema.org vocabulary. It helps Google show price, availability and stars right in the results, but it does not guarantee display and does not by itself raise a page in search.
In plain words
A person on the page sees: "Coffee maker X, 8,990 UAH, in stock". For a robot this is just text. Markup tells it explicitly: this is the product name, this is the price in hryvnias, this is the "in stock" status. Google uses such data for rich results and, together with the Merchant Center feed, for product cards.
Two different markups for a product
Google tells them apart by whether the product can be bought directly on the page (Google, Product):
- Product snippets are for pages where you can't buy (reviews, comparisons). More options for reviews, including pros and cons.
- Merchant listings are for pages where the product can be bought. They support price, availability, shipping, returns, sizes.
A store page is usually a merchant listing. A category page does not qualify: each page must be about one product or its variants (Google, Merchant listing).
What is required and what is recommended
According to Google's documentation for merchant listings (Google):
Product, required: `name`, `image`, `offers` (specifically `Offer`, not `AggregateOffer`). Product, recommended: `aggregateRating`, `review`, `brand`, `description`, `sku`, `mpn`, GTIN codes, `category`, color, size, `inProductGroupWithID` for variants.
Offer, required: `price` (greater than zero) and `priceCurrency` in ISO 4217 format (`UAH` for the hryvnia). Offer, recommended: `availability`, `itemCondition`, `url`, `shippingDetails`, `hasMerchantReturnPolicy`; for promotions also `validFrom`, `validThrough` or `priceValidUntil`.
Example (illustrative product and price):
```json { "@context": "schema.org", "@type": "Product", "name": "Coffee maker Example X", "image": ["example.ua/img/x-1x1.jpg"], "sku": "EX-100", "brand": {"@type": "Brand", "name": "Example"}, "offers": { "@type": "Offer", "url": "example.ua/kavovarka-x/", "price": "8990", "priceCurrency": "UAH", "availability": "schema.org/InStock" } } ```
Shipping and returns
For `shippingDetails`, Google requires delivery time (handling and transit time in days), the destination country in ISO 3166-1 code (`UA` for Ukraine) and the cost; for free shipping the value is `0`. For returns, the country and the policy category are needed, and for a limited period also the number of days. Google advises setting shipping and return policies at the organization level rather than in each offer (Google); the return policy, by the same advice, is nested in `Organization` (Google, Product).
Reviews and rating
Google's rules for `Review` and `AggregateRating` (Google, Review snippet):
- A review must be about a specific product, not a category or a list.
- If you specify several individual reviews, an aggregate rating is needed too.
- The aggregate rating must be visible on the page and cannot collect reviews from other sites.
- An average value and at least one of `ratingCount` or `reviewCount` are needed.
- Decimal values are written with a dot (4.4, not 4,4).
- The review author must be a valid name of a person or a team.
The documentation changelog has an entry from 24 July 2026 about a new guideline on fake and unlabeled incentivized reviews (Google, changelog). The changelog doesn't give the guideline's details; read it in the documentation. On trust in reviews in general: Social proof.
Markup and AI
- Google in its 2026 guide: structured data is not required for generative search, and there is no special schema.org markup for it; but markup is useful for rich results and must match the visible text (Google, 2026). Google wrote the same in May 2025 (Google, 2025).
- Ahrefs (May 2026): 1,885 pages that added JSON-LD were compared with control pages. In ChatGPT (+2.2%) and AI Mode (+2.4%) there were no changes different from zero, and in AI Overviews there was a small decline (−4.6%). Limitations: the pages were already often cited, other things changed too, the period is short. Pages that AI doesn't see at all were not studied (Ahrefs, 2026).
Takeaway: Product and Offer markup is added for Google's rich results and product cards, not for AI answers.
What has changed recently
- Since 7 May 2026, FAQ rich results are no longer shown in Google; the documentation was removed on 15 June (Google, changelog, 2026). If the site has FAQ markup, pages won't break, but there will be no stars or questions in the results.
- In 2025 Google began removing seven rarely used markup types; Product is not among them (Google, 2025).
- In July 2026 Google added to the documentation a section on promotion duration (`validFrom`, `validThrough`, `priceValidUntil`) and explained the use of `Product.category` (Google, changelog).
Rules that are broken most often
- Markup doesn't match the visible text (a different price, availability). Google requires consistency (Google, 2025).
- An expired `priceValidUntil`: Google warns that the listing may not be shown (Google, Merchant listing).
- Markup added by a script after load. Google advises putting it in the initial HTML, because markup generated by JavaScript is crawled less reliably (ibid.).
- Markup on a category page instead of a product page.
- Several currencies at one URL: Google advises a separate URL for each currency (ibid.).
- The feed and the page diverge. Google advises turning on automatic updates in Merchant Center so data gets corrected when the price or availability on the site is different (Google, Share product data). See Google Merchant Center.
FAQ
What is structured data? Code on a page that describes the product, price, availability and reviews in machine-readable form.
What markup does an online store need? At minimum Product with `name`, `image` and `offers`, where there are `price`, `priceCurrency` and preferably `availability`; then shipping, returns, rating.
Is markup needed for AI Overviews or ChatGPT? Google says directly that it is not. In its 2026 study, Ahrefs found no noticeable increase in citations after adding JSON-LD.
JSON-LD or Microdata? Google's documentation on reviews has examples in JSON-LD, Microdata and RDFa. In its test, Ahrefs checked only JSON-LD.
Related terms
- Google Merchant Center and free product listings
- Product feed
- Product page
- SEO audit of an online store
- GEO (generative engine optimization)
Sources
- Google Search Central, Product structured data — developers.google.com/search/docs/appearance/structured-data/product
- Google Search Central, Merchant listing structured data — developers.google.com/search/docs/appearance/structured-data/merchant-listing
- Google Search Central, Review snippet structured data — developers.google.com/search/docs/appearance/structured-data/review-snippet
- Google Search Central, Share your product data with Google — developers.google.com/search/docs/specialty/ecommerce/share-your-product-data-with-google
- Google Search Central, documentation changelog — developers.google.com/search/updates
- Google Search Central, Optimizing your website for generative AI features (May 2026) — developers.google.com/search/docs/fundamentals/ai-optimization-guide
- Google Search Central Blog, Succeeding in AI search (21.05.2025) — developers.google.com/search/blog/2025/05/succeeding-in-ai-search
- Google Search Central Blog, Simplifying the search results page (12.06.2025) — developers.google.com/search/blog/2025/06/simplifying-search-results
- Ahrefs, We Tracked 1,885 Pages Adding Schema (11.05.2026) — ahrefs.com/blog/schema-ai-citations
Related
Google Merchant Center and free product listings
Google Merchant Center is the account where a store uploads data about its products (name, price, availability, photos, link) so that Google can show them in Shopping, search, images and other services. Free product listings are the display of such products…
Technical site audit
A technical site audit is a check of how the site works "under the hood": speed and Core Web Vitals, server response codes (404, 5xx), redirects, robots.txt and sitemap, HTTPS, the mobile version. Its goal is to find errors that stop buyers from using the…
SEO audit of an online store
An SEO audit of an online store is a check of the site for everything that stops search engines from finding, crawling, indexing and showing the store's pages for the right queries: technical condition, catalog structure, duplicates, meta tags, content and…
SEO trends for an online store
SEO trends are changes in how people search and how search shows results. Over 2025–2026 the picture changed most because of AI-based answers right in Google search (AI Overviews, AI Mode), the growth of zero-click search and new reporting tools. The basic…