Structured Data & Schema Markup

Brass-SEO treats this as the authoritative source for all schema markup implementation. Schema.org is maintained by a W3C Community Group founded by Google, Microsoft, Yahoo, and Yandex, and defines over 900 types and 1,500 properties. The specification is the direct source for which types are eligible for Google rich results and which properties are expected by each search engine's parser. Schema types not defined in schema.org are not processable by major search engines. Brass-SEO's structured data recommendations are derived directly from the schema.org type hierarchy.

Examines:

The schema.org vocabulary specification — 900+ types and 1,500+ properties maintained by the W3C community, covering entities from WebPage to Product to Person.

Brass-SEO draws on:

The type and property definitions — the basis for Brass-SEO's structured data audit and rich result eligibility checks.

Primary source

schema.org/docs/about.html

Brass-SEO treats this as the implementation reference for JSON-LD serialization. JSON-LD 1.1 is a W3C Recommendation — a final specification — published July 2020, defining the syntax for embedding Linked Data in JSON. The @id field creates a dereferenceable entity identifier that both search engines and AI retrieval systems use for entity disambiguation; the @type field maps directly to schema.org types; and @graph enables multi-entity documents with shared context. These features are supported by Google's structured data parser and enable more precise entity relationships in markup.

Examines:

W3C JSON-LD 1.1 Recommendation — syntax, context processing, @id entity identifiers, and Linked Data serialization rules for embedding schema.org markup in web pages.

Brass-SEO draws on:

The @id and @graph features — cited in Brass-SEO's structured data recommendations for entity disambiguation and multi-entity markup.

Primary source

w3.org/TR/json-ld11

Brass-SEO monitors this for structured data adoption benchmarks. The 2024 Web Almanac Structured Data chapter reports JSON-LD present on 41% of analyzed pages, up from 34% in 2022, making it the fastest-growing structured data format. Open Graph markup appears on 64% of pages. JSON-LD is now mainstream for basic types; the competitive differentiation has shifted to less common but high-value types — BreadcrumbList, VideoObject, FAQPage — and to schema accuracy and data freshness. Schema presence alone is no longer a meaningful differentiator.

Examines:

Annual independent measurement of structured data adoption — formats (JSON-LD, RDFa, Microdata), schema types, and Open Graph usage across millions of analyzed pages.

Brass-SEO draws on:

The 41% JSON-LD adoption rate and type-level breakdown — used to contextualize a site's structured data coverage in Brass-SEO's page audit.

Primary source

almanac.httparchive.org/en/2024/structured-data

Brass-SEO draws on this to benchmark schema type prevalence at web scale. The Web Data Commons project at the University of Mannheim extracted 74 billion RDF triples from 2.4 billion URLs across 37 million domains in its October 2024 release, covering JSON-LD, Microdata, RDFa, and Microformat markup. The longitudinal dataset, tracking from 2009 through 2024, shows which schema types have moved from early adoption to baseline expectations. Types that were differentiators in 2015 are now table stakes; practitioners can compare their schema coverage against the full-web corpus to find gaps common among peer sites.

Examines:

Large-scale extraction and analysis of structured data from the Common Crawl — 74 billion RDF triples from 2.4 billion URLs, with longitudinal data from 2009 to 2024.

Brass-SEO draws on:

The 74B-triple corpus and longitudinal type adoption data — used to contextualize which schema types are standard versus still differentiating.

Primary source

webdatacommons.org/structureddata

Brass-SEO draws on this to explain the gap between standard schema SEO and AI retrieval optimization. Volpini et al. found that standard JSON-LD markup alone did not produce the AI retrieval gains practitioners expected. A 29.6% RAG accuracy improvement was achieved specifically for enhanced entity pages that combined schema.org markup with dereferenceable linked data and explicit agent-readable instructions — not for pages with standard homepage JSON-LD alone. Schema markup is a necessary but not sufficient condition for AI engine visibility; dedicated entity pages for brand, product, or person entities with rich linked data signals produced gains that standard schema did not.

Examines:

Research on how structured linked data functions as a memory layer for AI agent retrieval — comparing standard JSON-LD, enhanced entity pages, and linked data configurations.

Brass-SEO draws on:

The 29.6% RAG accuracy finding — cited when explaining why Brass-SEO's AI Citability recommendations extend beyond basic schema markup to entity page strategy.

Primary source

arxiv.org/abs/2603.10700

Brass-SEO draws on this to correct a common misconception about FAQ schema. In August 2023 Google removed HowTo rich results from search entirely and limited FAQ rich results to well-known, authoritative government and health sites. For most sites the visual payoff of FAQPage and HowTo markup — the expandable questions in the result — is gone. Google confirmed the markup itself stays valid and does not need removing. The SEO reason to add FAQ schema mostly disappeared; the GEO reason grew, because question-and-answer markup still hands AI systems a clean, self-contained answer to extract. Keep FAQPage schema and repurpose it for AI, instead of stripping it out.

Examines:

Google's August 2023 announcement removing HowTo rich results and restricting FAQ rich results to authoritative government and health sites, with confirmation that the markup remains valid.

Brass-SEO draws on:

The rich-result reduction paired with the still-valid markup — cited when Brass-SEO explains why FAQPage schema is worth keeping for AI extraction even without a rich result.

Primary source

developers.google.com/search/blog/2023/08/howto-faq-changes