- Product schema is a block of JSON-LD on each product page that tells search engines, in a fixed vocabulary, what the product is, what it costs, and whether it is in stock.
- Google uses it to make pages eligible for product rich results and merchant listings, and it works best when markup, visible page and Merchant Center feed all agree.
- Most Shopify themes already output basic Product markup. The common problems are missing identifiers, duplicate blocks from apps, and prices or stock that disagree with the page.
A shopper reads your price, photo and “Ships in 2 days” note in a second. A crawler has to guess. Product schema removes the guessing by stating those facts in a format machines read the same way every time.
What is product schema and why does it matter for an online store?
Structured data is code that describes the content of a page using a shared vocabulary. Schema markup is the most common vocabulary, maintained at schema.org. For eCommerce, two types do most of the work:
- Product describes the item: name, images, description, brand, SKU and global identifiers.
- Offer sits inside Product and describes the deal: price, currency, availability, the URL where it can be bought, and optionally shipping and return terms.
Google recommends JSON-LD, a script block separate from the HTML layout. It is easier to maintain than microdata, which wraps attributes around visible elements.
Two Google search features depend on this markup:
- Product snippets add price, availability or review stars under a regular search result.
- Merchant listings let product pages appear in shopping-style results, including the Shopping tab and product knowledge panels. These require an Offer with a price.
Markup makes a page eligible. It does not guarantee either feature.
Which product schema properties should every product page include?
Use the table below as a working checklist. “Required” here means what Google’s documentation needs for merchant listings to be eligible. “Recommended” properties make the listing more complete and the product easier to match.
| Property | Level | What it holds | Common mistake |
|---|---|---|---|
| name | Required | Product title as shown on the page | Adding promo text (“SALE!”) that the H1 does not show |
| image | Required | One or more crawlable image URLs | Pointing to a tiny thumbnail or a lazy-load placeholder |
| offers.price | Required | Current selling price as a number | Using the compare-at price instead of the sale price |
| offers.priceCurrency | Required | ISO 4217 code, for example USD | Missing on multi-currency stores |
| offers.availability | Recommended | schema.org value such as InStock or OutOfStock | Hardcoded to InStock in the theme |
| offers.url | Recommended | The canonical product or variant URL | Pointing to a collection-scoped URL |
| description | Recommended | A plain-text product description | Pasting raw HTML or a blank string |
| sku | Recommended | Your internal stock-keeping unit | Different SKU than the feed |
| gtin / mpn | Recommended | Barcode (UPC, EAN) or manufacturer part number | Leaving blank when the product has a barcode |
| brand | Recommended | The brand name as a Brand object | Using the store name for third-party brands |
| aggregateRating / review | Optional | Ratings and reviews shown on the page | Marking up reviews that are not visible or not genuine |
| shippingDetails | Optional | Shipping cost, destination and delivery time | Values that contradict the shipping policy page |
| hasMerchantReturnPolicy | Optional | Return window, method and fees | A 30-day window in markup and 14 days on the policy page |
When should you add ratings and reviews?
Only mark up ratings and reviews that are real, collected from customers, and visible on the same page. If the review widget loads stars that are not on the page, or the numbers come from somewhere else, leave aggregateRating out. Misleading review markup can lead to a manual action, and it is never worth the risk.
What about variants?
A product with sizes or colors can be described as a ProductGroup, with each variant as its own Product linked through hasVariant and variesBy. One Product with several Offers can also be valid. Either way, each price and availability value must match what a shopper sees when that variant is selected.
What does valid Product JSON-LD look like?
Here is a short, illustrative example for a single-variant product. Replace every value with real data from your store.
json
{
"@context": "https://schema.org",
"@type": "Product",
"name": "Merino Crew Sock, Charcoal",
"image": ["https://example.com/images/merino-crew-charcoal.jpg"],
"description": "Midweight merino wool crew sock with a reinforced heel and toe.",
"sku": "MCS-CHAR-M",
"gtin13": "0123456789012",
"brand": { "@type": "Brand", "name": "Example Socks Co." },
"offers": {
"@type": "Offer",
"url": "https://example.com/products/merino-crew-sock-charcoal",
"price": "24.00",
"priceCurrency": "USD",
"availability": "https://schema.org/InStock",
"itemCondition": "https://schema.org/NewCondition",
"shippingDetails": {
"@type": "OfferShippingDetails",
"shippingRate": { "@type": "MonetaryAmount", "value": "0", "currency": "USD" },
"shippingDestination": { "@type": "DefinedRegion", "addressCountry": "US" }
},
"hasMerchantReturnPolicy": {
"@type": "MerchantReturnPolicy",
"applicableCountry": "US",
"returnPolicyCategory": "https://schema.org/MerchantReturnFiniteReturnWindow",
"merchantReturnDays": 30
}
}
}
There is no rating block because this example has no reviews on the page. Do not add properties just to fill the template.
How do you keep schema consistent with Merchant Center and the page?
Google sees your product data from three places: the visible page, the structured data, and your Merchant Center product feed. When they disagree, you get warnings, disapproved items, or a listing that shows the wrong price.
A simple consistency rule: the page is the source of truth, and the markup and feed must both mirror it.
- [ ] Price in markup equals the price a shopper pays before checkout, including active sale prices.
- [ ] Availability updates when stock changes, not only when the theme is deployed.
- [ ] GTIN, MPN and brand match the feed, character for character.
- [ ] Offer URL is the canonical URL, consistent with your canonical tag.
- [ ] Shipping and return values match the policy pages and your Merchant Center settings.
- [ ] Currency matches the market the page serves.
Shipping and return terms can be declared per product, as above, or store-wide: Google also accepts a return policy in Organization markup. Pick one approach and update it with Merchant Center whenever policies change.
What do Shopify themes and SEO apps output, and how do you avoid duplicates?
Most current Shopify themes, including Dawn and other Online Store 2.0 themes, output Product JSON-LD on product pages. Many generate it with Shopify’s structured_data Liquid filter, which produces a reasonable baseline. Older or heavily customized themes may still use microdata or hand-written JSON-LD with gaps, such as availability hardcoded to InStock or no GTIN.
Trouble usually starts with an SEO or schema app. The app adds its own Product block, the theme keeps its block, and the page now carries two Product entities. If one shows the sale price and the other the regular price, Google gets conflicting facts.
To fix it:
- View the page source of a product page and search for “@type”: “Product”. Count the blocks.
- Pick one source. Apps are easier to manage at scale; theme code gives full control.
- Turn off the other. Most schema apps can replace theme markup. If not, remove the theme snippet in a duplicated theme first.
- Recheck review apps, which sometimes inject a third block with their own aggregateRating.
WooCommerce: Core outputs Product JSON-LD, and plugins such as Yoast or Rank Math add their own. Confirm the plugin extends the core markup instead of duplicating it.
Magento / Adobe Commerce: The default Luma theme has traditionally used microdata. If you add a JSON-LD extension, remove the microdata so the product is not described twice.
Do you need Organization markup too?
Yes. Product markup describes what you sell. Organization markup, usually on the homepage, describes who sells it: name, URL, logo, contact details and sameAs links to official profiles. Keep one Organization entity across the site, with the brand name spelled the same way everywhere.
Does product schema help with AI answers?
Nobody outside the AI companies can say exactly how each assistant weighs structured data. Be wary of anyone who calls schema a guaranteed ranking factor for AI engines.
What is safe to say: clear, consistent structured data and plain on-page facts make a product easier for machines to understand. When the name, price, materials, dimensions and return terms appear the same way on the page, in the markup and in the feed, any system that reads your page has less to misinterpret. That is the same discipline that supports generative engine optimization: state facts clearly, in text, in one consistent form.
In practice: put key specs in visible text, not only in images; use one product name across the H1, markup, feed and marketplaces; and answer sizing, care and compatibility questions on the page.
How do you validate product schema?
Use three tools, in this order:
- Google Rich Results Test (search.google.com/test/rich-results). It shows which rich results a URL is eligible for, plus errors and warnings. Test one product per template: simple, with variants, on sale, out of stock.
- Schema Markup Validator (validator.schema.org). It checks markup against the full schema.org vocabulary and is good for spotting duplicate entities and typos.
- Google Search Console. The Merchant listings and Product snippets reports show errors and warnings across all indexed product pages. Check them after any theme change, app install or migration.
Warnings on optional properties are not failures. Errors on required properties mean the page is not eligible.
If you want a quick read on schema gaps alongside other revenue leaks, run a Free Profit Audit on your store. For ongoing technical SEO work, including schema, feeds and AI visibility, see our eCommerce SEO service.
FAQ
Is product schema a ranking factor?
Google has not described structured data as a direct ranking factor for regular results. It makes pages eligible for richer displays, such as price and availability in results, which can change how many people click.
Do I need product schema if I already have a Merchant Center feed?
Yes. The feed powers Shopping ads and free listings, while markup describes the page itself to search crawlers. Keeping both consistent reduces mismatches and helps Google trust your data.
Can I add review stars if my reviews come from a third-party platform?
Only if those reviews are shown on the product page and reflect genuine customer feedback about that product. If they are not visible on the page, leave the rating markup out.