Schema.org markup · JSON-LD
Structured data is machine-readable markup that describes a page’s content, preferably in JSON-LD. It helps engines understand entities and offer rich results, as long as it faithfully reflects the visible content. Google states that no special markup is required for its generative search.
Standalone definition · 43 words
How does structured data work?
The Schema.org vocabulary defines types (Article, Product, Organization, FAQPage) and properties. The markup is inserted in the page, often as JSON-LD, and states explicitly what the text says implicitly. Google recommends JSON-LD in most cases.
Does structured data matter for AI engines?
Google says no specific markup is needed to appear in its AI features, and its May 15, 2026 guide lists over-focusing on structured data among the myths. It remains recommended within an overall SEO strategy because it helps with eligibility for rich results. Markup remains useful to disambiguate entities, connect pages and feed rich results. It must match the visible content, otherwise it loses its value.
Which types should you prioritize?
Organization and Person for identity, Article for editorial content, Product and Offer for e-commerce, FAQPage and DefinedTerm for answer and glossary pages. Validate each page template with the rich results test and the Schema.org validator.
Frequently asked questions
JSON-LD or microdata?
Google recommends JSON-LD. It is easier to maintain because it is separate from the visible HTML.
Does structured data improve rankings?
Google does not present it as a ranking factor. It serves rich result eligibility and content understanding.
Should all content be marked up?
No. Mark up what is visible and useful, with relevant types. Excessive or misleading markup can be ignored.
Related terms
Further reading
- Search Central Live Deep Dive Barcelona 2026: what Google said over the 3 days
- Google Search Central 2025 Zurich
- Google Search Central Live Paris recap: Google Is Betting big on Agentic Commerce
- ← Back to the GEO glossary (114 terms)