Structured data

Structured data is information in a page’s source code that describes its content in a fixed, machine-readable form. Instead of a search engine having to guess from running text what a product’s price is, it states explicitly: this is a product, this is its price, it has this many stars and it is in stock. Search engines use this information to understand pages better and to display them in search results with additional information. For online stores, it is one of the most effective technical measures of all.

Schema.org: the shared vocabulary

So that all search engines speak the same language, Google, Microsoft, Yahoo and Yandex jointly founded schema.org in 2011. The vocabulary describes hundreds of types of things, such as Product, Offer, Organization, BreadcrumbList or Review, each with defined properties such as name, price or availability.

The three formats

• JSON-LD: a separate script block, independent of the visible HTML. Google recommends this format because it is the easiest to maintain.

• Microdata: attributes directly in the page’s HTML elements.

• RDFa: similar to Microdata, mainly used in other contexts.

A simplified example for a product in JSON-LD:

<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Product",
"name": "Metal raised garden bed, 200 cm",
"image": "https://www.your-store.com/images/raised-bed-200.jpg",
"brand": { "@type": "Brand", "name": "Gartenwerk" },
"gtin13": "4012345678901",
"offers": {
"@type": "Offer",
"price": "189.00",
"priceCurrency": "EUR",
"availability": "https://schema.org/InStock"
},
"aggregateRating": {
"@type": "AggregateRating",
"ratingValue": "4.7",
"reviewCount": "38"
}
}
</script>

Rich results: what becomes visible in search results

From structured data, Google can generate so-called rich results, i.e. search results with additional elements. For stores, these are the most relevant:

Rich results stand out in the search results and can significantly increase the click-through rate. There is no entitlement to them: Google decides itself whether and when to show them.

Google has restricted or switched off some displays in recent years. Since 2023, Google only shows FAQ results for selected government and health websites, HowTo results no longer at all, and the search box in sitelinks was discontinued in 2024. Anyone using FAQ markup purely for the display should be aware of this.

Is structured data a ranking factor?

Not directly. Structured data doesn’t improve the position, but the presentation and understanding of a page. It still has an indirect effect: a result with price, stars and availability is clicked more often than one without. And Google identifies products, brands and companies more reliably through structured data, which also plays a role in entity SEO.

Structured data and AI search

According to Google, structured data isn’t a prerequisite for Google’s AI Overviews and AI Mode. For buying queries, however, structured product data plays a central role: price, availability and shipping in product answers come from structured sources, above all from Merchant Center and the markup of product pages. Other AI assistants and shopping agents also read structured data because it is unambiguous. More on this in the article on Google’s guide to AI search and in the entry on Generative Engine Optimization. Whether a store is readable for AI agents is checked by the Agentic Commerce Scanner.

Rules Google expects

• Only mark up what is visible: what is in the markup must be visible on the page. A price in the markup that differs from the displayed price violates the guidelines.

• Reviews only from your own page: stars may only be marked up if the reviews appear on the page itself.

• Keep it up to date: availability and price must be correct. “In stock” in the markup for a sold-out product damages trust.

• No abuse: anyone who deliberately uses misleading markup risks a manual action. Rich results then disappear for the whole website.

Structured data in Shopware and Shopify

Shopware 6 outputs structured data for products, offers and breadcrumbs as Microdata directly in the HTML of the standard storefront. Extended information such as shipping, returns, GTIN or reviews from external services requires theme adjustments or extensions.

Shopify outputs structured data via the theme, with current themes usually as JSON-LD for products and the company. Many apps for reviews or SEO add their own markup. A product can then quickly appear twice in the markup, with different information.

More on the peculiarities of the systems under Shopware SEO and Shopify SEO.

Typical mistakes in stores

• Duplicate markup: theme and app describe the same product, often with different prices or ratings.

• Missing required information: for example an offer without a price or currency. Google then shows no rich result.

• Variants mapped incorrectly: all sizes and colours share one offer with one price. For products with variants there is dedicated markup with which each variant gets its own price and availability.

• Missing product identifiers: without GTIN and brand, Google has a harder time identifying a product, including for Google Images and product comparisons.

• Reviews that aren’t on the page: for example stars from an external review service without visible reviews.

How to check structured data

• Google’s Rich Results Test: shows for a URL which rich results are possible and which information is missing.

• Schema Markup Validator: checks the markup against the schema.org standard, independently of Google.

• Google Search Console: under “Enhancements” and in the product snippet and merchant listing reports, Google shows errors and warnings for the whole website.

How product data in the feed and in the markup work together is described in the article on price comparison sites.

Frequently asked questions

Which format should I use?

JSON-LD. Google recommends it, and it can be maintained independently of the visible HTML.

Does structured data guarantee stars in search results?

No. It makes a page eligible. Whether Google shows rich results is up to Google, depending on the search query and the quality of the page.

Is FAQ markup still worthwhile?

For display in Google Search hardly, because Google now shows FAQ results for only a few websites. Frequently asked questions on the page remain valuable, for visitors and for AI assistants.

Do I need structured data for ChatGPT and co.?

It isn’t a prerequisite, but it helps. Unambiguous product data makes it easier for AI assistants and shopping agents to reproduce price, availability and attributes correctly.

How much effort is this for a store?

Shopware and Shopify provide the basics. The effort lies in adding to them, avoiding duplicates and keeping the information up to date permanently.