Technical SEO
Schema (Structured Data Markup)
Last reviewed
Schema markup is structured data added to a webpage using the Schema.org vocabulary — a shared standard maintained by Google, Microsoft, Yahoo, and Yandex. Rather than asking crawlers to infer meaning from prose, schema labels content directly: this is the Organization, this is its founder, this is an Article published on a specific date, this is a FAQPage with these Question / Answer pairs.
The recommended serialization format is JSON-LD, embedded in a `<script type="application/ld+json">` tag in the page head or body. Microdata and RDFa exist as alternatives but are largely deprecated in modern implementations because JSON-LD decouples markup from rendered HTML, simplifies maintenance, and reduces the risk of cloaking violations.
For schema to function, it must be present in the initial server-rendered HTML — not injected after hydration via a framework component. Crawlers like Common Crawl, GPTBot, and PerplexityBot do not execute JavaScript, so a schema script that lives only in the React hydration stream is invisible to them. The same `<script>` tag must appear when the page is fetched with `curl` and when it loads in a browser.
Schema implementation is validated using Google's Rich Results Test and the Schema.org Validator. Performance of schema-driven rich results is monitored in Google Search Console's Enhancement reports — separate dashboards for Products, FAQs, Events, Articles, and other entity types. Crawlers such as Screaming Frog can audit structured data coverage across an entire site.
Common entity types in B2B and editorial sites include Organization, Person, WebSite, BreadcrumbList, Article, FAQPage, Service, DefinedTerm, and Product. The most defensible implementations cross-reference entities using `@id` URIs (e.g., `https://example.com/#organization`), allowing Google's knowledge graph and AI engines to resolve relationships unambiguously.
Why it matters in GEO / AI search
In traditional SEO, schema unlocks rich results — star ratings, FAQ accordions, breadcrumb crumbs, and sitelinks search boxes — which expand a result's SERP real estate and lift click-through rate. But in GEO / AI search, the role is more fundamental: schema is the primary signal AI engines use to disambiguate entities when deciding whether to cite a source.
When ChatGPT, Perplexity, Claude, or Gemini retrieve a page, the model has to answer a brittle question: "Is this page about the same thing as the user's query?" Plain HTML forces the model to infer from prose, which is noisy. JSON-LD answers the question definitively: `@type: Organization`, `@id: https://example.com/#organization`, `sameAs: [linkedin, x, crunchbase]` — three facts that resolve the entity in a single parse.
For new domains without an established Common Crawl footprint or Wikidata entry, structured data is often the only reliable entity-disambiguation signal available. Sites that ship Organization + Person + WebSite + per-page Article/Service schemas measurably out-cite identical-content sites that don't.
Examples
Organization + Person cross-link
Emit Organization with `founder: { @id: #founder }` and a separate Person with `worksFor: { @id: #organization }`. The bidirectional `@id` resolution is what Knowledge Graph systems look for.
FAQPage tied to visible Q&A
Wrap your visible FAQ section in FAQPage schema. Each Question.name must match the visible H3 text verbatim, and Answer.text must match the visible paragraph — Google demotes schema with hidden content.
Article with publisher reference
Article schema with `author: { @id: #organization }` and `publisher: { @id: #organization }` is the schema.org-compliant pattern for branded editorial content (Stripe blog, Notion blog).
BreadcrumbList per leaf
On a glossary or category leaf page, BreadcrumbList declares Home › Section › Term. Google uses this to render breadcrumb crumbs in mobile SERPs and AI engines use it for hierarchy inference.
Authority Links
Google Search Docs — Introduction to Structured Data
Official Google guidance on schema implementation, formats, and rich-result eligibility.
Schema.org
The canonical vocabulary registry maintained by Google, Microsoft, Yahoo, and Yandex.
Google Rich Results Test
Validates schema markup against Google's rich-result eligibility rules.
Related Terms
Technical SEO
Breadcrumb
Breadcrumbs are a navigation aid that helps users easily understand the relation between their location on a page and higher-level pages.
Content
Featured Snippets
Boxed answer-extractions Google displays above the normal blue-link results — paragraphs, lists, tables, or steps lifted directly from a top-ranking page to answer a query without requiring a click.
Technical SEO
Rich Snippets (Rich Results)
Enhanced Google search-result listings that include structured-data-driven visual elements — star ratings, prices, breadcrumbs, FAQ accordions, event dates, product availability — beyond the standard title, URL, and description.
Technical SEO
SERP (Search Engine Results Page)
The page a search engine returns in response to a query — increasingly less a list of "10 blue links" and more a composed surface combining AI Overviews, featured snippets, People Also Ask, local pack, ads, and organic results.