Schema MarkupTelling machines what you mean

What structured data does, and what it does not#

A web page is written for people. A search engine has to infer that a number is a price, that a name is the author, that a block of text is a review. Structured data removes the inference by stating the facts explicitly in a format machines read reliably.

Two benefits follow. Pages can become eligible for rich results, such as product prices and ratings, breadcrumbs or event dates in search listings. And search and AI systems get a clearer picture of the entities on your site: your organisation, people, products and places, and how they relate.

What structured data does not do is make a page rank by itself. Google has said structured data is not a general ranking factor, and eligibility for a rich result is not a guarantee of one. Markup describes content; it cannot substitute for it.

Structured data is a label on the jar, not the jam.

The vocabulary and the format#

The vocabulary is schema.org, a shared set of types and properties maintained collaboratively and supported by the major search engines. It is large: there are hundreds of types. Only a fraction are used for search features, and each search engine documents which types and properties it supports.

The format Google recommends is JSON-LD: a block of script in the page that describes the content separately from the HTML. It is easier to generate, maintain and validate than the older inline formats, Microdata and RDFa, which weave attributes through the visible markup.

Which types matter for most businesses#

Resist the urge to mark up everything. Choose types that describe your real content and that search engines actually use. Rich result support changes over time; Google, for example, has narrowed or retired some features, such as FAQ and how-to rich results, in recent years. Check the current search gallery documentation before investing.

How to implement structured data, step by step#

The mapping step is where quality is decided. If price in the markup comes from a different field from the price on the page, they will eventually disagree, and the mismatch can cost rich results or trigger a manual action for misleading markup. Generate structured data from the same source of truth that renders the visible page.

For an Organization, a minimal JSON-LD block names the organisation, its URL, its logo and its official profiles through the sameAs property. On a product page, a Product block nests an Offer with price, priceCurrency and availability. Most CMS platforms and ecommerce systems can output these from templates or through well-maintained plugins; verify what they output rather than trusting defaults.

Validation and monitoring#

Validate in two ways. Google's Rich Results Test shows whether a page is eligible for Google's rich results and flags errors in supported types. The Schema.org validator checks general syntax and vocabulary use beyond Google's features. Use both before release.

After release, Search Console's enhancement reports show valid items, warnings and errors across the site by type. A sudden rise in errors usually means a template or plugin change broke the output. Make checking these reports part of the monthly routine described in the SEO audit checklist.

Mistakes that cost rich results#

• Marking up content users cannot see, such as reviews that do not appear on the page or a price that differs from the visible one.
• Self-serving review markup: a business marking up ratings of itself on its own site is not eligible for review rich results under Google's guidelines.
• Stale data: event markup for past events, out-of-stock products marked in stock.
• Wrong types: marking a category page as a single Product, or every page as an Article.
• Plugin conflicts: two plugins outputting overlapping, contradictory blocks.

Entities: the quiet benefit#

Rich results get the attention, but the longer-lasting value of structured data may be entity clarity. An Organization block that names your business, links your official profiles through sameAs and describes your logo and contact points gives search systems an unambiguous anchor. Person markup for authors, linked to their profiles, does the same for expertise.

Consistency is what makes this work. If your organisation's name, address and profiles differ between your site, your business profiles and directories, markup cannot resolve the confusion; it adds one more version. Fix the facts everywhere, then describe them in markup. This is especially important for local SEO, where name, address and phone consistency matters for every location.

For larger organisations, consider maintaining a single internal record of core entities, with official names, identifiers, logos and profile links, that every template and team draws from. It sounds bureaucratic and saves a great deal of confusion.

Structured data and AI search#

There is reasonable debate about how much generative systems use structured data directly. What is clear is that consistent, explicit facts about your organisation, products and people make you easier to understand for any system that reads the web. Structured data is one way to state those facts; clear visible content and consistent profiles across the web are others.

Our advice is pragmatic: implement the core types properly because they have concrete search benefits today, and treat any AI benefit as a bonus. Read generative engine optimisation and E-E-A-T for how entity clarity supports credibility more broadly.

Published by Fabulous.Media, a network of specialist marketing agencies. Updated 9 October 2026. Platform features change often; check current official documentation before acting on platform-specific detail.