What is structured data (JSON-LD / Schema.org)?

Structured data is markup embedded in a page's HTML, usually in JSON-LD format, that describes the content to machines in the Schema.org vocabulary. It states what the page is (Article, FAQPage), who wrote it (Person, Organization), when it went live, when it last changed, and how it connects to other entities. For AI crawlers, it is the unambiguous version of the content.

Which schema types matter for AI Visibility?

TYPE WHAT IT DESCRIBES WHERE TO USE IT
Organization / WebSite Who the company is and what the site is Home
Article + Person Editorial content and its author Articles and analysis
FAQPage Question and answer pairs The FAQ at the end of each page
BreadcrumbList The page's position in the site hierarchy Every internal page
DefinedTerm The canonical definition of a term Glossary entries
HowTo A step-by-step process Method and tutorial pages

Source: Schema.org vocabulary; application: ReBo architecture, 2026.

Related entries

Structured data is one of the levers of GEO and a precondition for Level 3 — AI Friendly and Level 4 — AI Reference in the AI Visibility framework. AI crawlers are what read it, and its measured effect on citations appears in the article on what AI Visibility is.

Frequently asked questions

Is structured data visible to the reader?

No. JSON-LD sits in an invisible script block in the HTML. The reader sees the normal page; machines (AI crawlers and search engines) read the structured description. That is why the visible content and the JSON-LD must always say the same thing.

What effect does structured data have on being cited by AI?

2026 GEO factor studies measured up to 73% higher selection for citation on pages that carry Article and FAQPage schema, because the AI extracts question, answer, author and date without ambiguity.

How do I validate a page's JSON-LD?

With Google's Rich Results Test or the Schema.org validator: paste the URL or the code and check that the types and properties are recognized without errors.