How schema markup works
Schema markup is added to a page as structured data — most commonly in JSON-LD format inside a script tag. It does not change the visible content of the page. It adds a machine-readable layer that describes the page's meaning, context, and relationships.
Search engines use schema markup to power rich results such as star ratings, FAQ dropdowns, breadcrumbs, and event listings. AI systems use it to understand page authorship, content type, dates, and entity relationships without parsing prose.
The most important schema types
Organization — describes a company with name, URL, logo, contact details, and sameAs links to verified external profiles. Essential for entity clarity.
Person — describes an individual author with name, role, employer, and links to external presence. Critical for E-E-A-T signals on content pages.
Article / WebPage — describes a content page with headline, author, datePublished, dateModified, and publisher. Makes authorship and recency machine-readable.
FAQPage — describes a page with question-and-answer pairs. Enables FAQ rich results and gives AI systems self-contained Q&A units to retrieve.
BreadcrumbList — describes the page's position in the site hierarchy. Helps search engines and AI systems understand site structure.
LocalBusiness / ProfessionalService — describes a physical or service business with address, hours, and service area. Important for local SEO and entity verification.
Service — describes a specific service offering with name, description, provider, and area served.
JSON-LD vs Microdata
JSON-LD is the recommended format for schema markup. It is added as a separate script block and does not require modifying HTML elements. Google recommends JSON-LD and it is easier to maintain than inline Microdata.
Schema markup and AI Search
For Google AI Overviews specifically, Google states that no special schema markup is required to appear. Structured data should not be treated as an AI shortcut.
Its role is more fundamental: it makes pages self-describing so that systems can answer who published this, when, and on what topic without inferring from prose. This reduces ambiguity and increases the probability of correct attribution.
How to test schema markup
- Google Rich Results Test: search.google.com/test/rich-results
- Schema Markup Validator: validator.schema.org
- Google Search Console: Enhancements section shows detected structured data and errors
Source
Schema markup vocabulary is maintained at schema.org. Google's implementation guidelines are at developers.google.com/search/docs/appearance/structured-data/intro-structured-data