Glossary · Beginner
Structured data
Structured data is machine-readable markup that describes what a page means, usually expressed as JSON-LD using the schema.org vocabulary.
- Published
- Last checked
Definition
Structured data is markup that states a page's meaning in a form machines can
parse. The common format is JSON-LD, placed in a script element and written
against the schema.org vocabulary.
It describes, it does not decorate
Structured data changes nothing visually. It makes explicit what a person infers from the layout: that this is an article, that this is its publication date, that this trail of links is a breadcrumb. Search engines may use it to understand the page and, in some cases, to display richer results.
The parity rule
Every fact asserted in structured data must be visible on the page. Marking up a rating, a price, or an author that a reader cannot see is a guideline violation and a good way to lose eligibility entirely. Generate the markup from the same data that renders the page, and the two cannot drift apart.
Eligibility is not entitlement
Valid markup makes a page eligible for a rich result. It does not produce one. Search engines decide what to display, and support for any given type changes over time. Treat structured data as clarity for machines, and treat any display change as a possible bonus.
Common failures
- Fields describing content that does not appear on the page.
- Markup copied between templates and never updated.
- Multiple conflicting graphs on one page.
- Breadcrumb data that disagrees with the visible trail.
- Types chosen for their rich result rather than their accuracy.
Review question
Take each assertion in the JSON-LD and point at where a reader sees the same fact on the rendered page. Anything you cannot point at should be removed rather than defended.