• JSON-LD is Google's recommended format.
• Markup must describe content that is visible on the page.
• Structured data makes rich results possible. It does not guarantee them and is not a direct ranking factor.
• Google supports a limited and changing list of rich result types, so check current documentation.
• Connect your entities with @id and sameAs to build a consistent picture of who you are.

Search engines are good at reading text and still benefit from being told things explicitly. Structured data does that. It labels a page's content: this is a product, this is its price, this is the author, this is the organisation that published it.

Formats

• JSON-LD: a script block, separate from the visible HTML. Recommended by Google and the easiest to maintain.
• Microdata and RDFa: attributes woven into the HTML. Still supported, harder to manage.

A basic example

 <script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Canonical Tags: The Complete Guide",
  "datePublished": "2026-10-07T09:00:00+01:00",
  "dateModified": "2026-10-07T09:00:00+01:00",
  "author": {
    "@type": "Person",
    "name": "Jane Example",
    "url": "https://www.example.com/about"
  },
  "publisher": {
    "@type": "Organization",
    "@id": "https://www.example.com/#organization",
    "name": "Example Ltd"
  },
  "image": "https://www.example.com/images/canonical-tags.png"
}
</script> 

What structured data does

• Rich results: enhanced listings such as review stars, prices, availability, breadcrumbs, recipes, events and video thumbnails. These can lift click-through rate noticeably.
• Merchant and local features: product markup supports free listings and shopping experiences, and LocalBusiness markup supports local results.
• Entity understanding: Organization and Person markup helps search engines connect your site with the real-world thing it represents.

It is not a ranking factor in itself. Pages do not rank higher because they have schema. They can earn more clicks and be understood more accurately.

Types worth knowing

• Organization and LocalBusiness: name, logo, contact details and sameAs links to official profiles.
• WebSite: site name and alternate names.
• BreadcrumbList: the page's position in the hierarchy.
• Article, NewsArticle, BlogPosting: headline, dates, author.
• Person and ProfilePage: author and expert details.
• Product, Offer, AggregateRating, Review: price, availability and ratings.
• Event, Recipe, JobPosting, VideoObject: each with its own rich result.
• FAQPage: rich results for this type have been heavily restricted by Google since 2023, with HowTo rich results removed altogether. The markup remains valid Schema.org.

Google regularly retires rich result types that see little use, and did so for several in 2025. Always check the Search Gallery in Google's documentation before investing in a type purely for the visual treatment.

Connecting entities

Instead of isolated blocks, link your markup into a graph. Give key entities a stable @id and reference it elsewhere:

 {
  "@context": "https://schema.org",
  "@type": "Organization",
  "@id": "https://www.example.com/#organization",
  "name": "Example Ltd",
  "url": "https://www.example.com/",
  "logo": "https://www.example.com/logo.png",
  "sameAs": [
    "https://www.linkedin.com/company/example",
    "https://www.wikidata.org/wiki/Q00000000"
  ]
} 

Rules to follow

• Only mark up content users can see on that page.
• Do not add review markup for reviews you wrote about yourself. Self-serving reviews for Organization and LocalBusiness are not eligible for stars.
• Use the most specific type that fits.
• Complete required properties and as many recommended ones as you honestly can.
• Keep values in sync with the page, especially price and availability.
• Breaking the guidelines can lead to a structured data manual action that removes rich results.

Common mistakes

• Markup that contradicts the visible page.
• Missing required properties, so the page is not eligible.
• Invalid JSON from a stray comma or unescaped quote.
• The same Organization block with different details on different pages.
• Injecting markup through a tag manager and assuming every crawler will see it. Google usually will, as it renders JavaScript. Most other systems will not.
• Adding every type imaginable in the hope something sticks.

How to test

• Rich Results Test shows which Google rich results a page qualifies for.
• Schema Markup Validator at validator.schema.org checks the syntax of any Schema.org type.
• Search Console enhancement reports show errors and warnings at scale for supported types.
• A site crawler with structured data extraction validates whole templates.

The GEO angle

There is a lot of debate about how much language models use schema directly. What is well supported is this: structured data strengthens the knowledge graphs and indexes that AI search draws on, Microsoft has said Bing uses schema to help its models understand content, and consistent Organization and Person markup reduces ambiguity about who you are. Treat it as supporting evidence that backs up clear on-page content, and keep it in the server-rendered HTML so every crawler can read it.

Add your title here

This is a paragraph. Writing in paragraphs lets visitors find what they are looking for quickly and easily. Make sure the title suits the content of this text.