Schema Markup (JSON-LD)

What It Is

Schema markup expresses content as machine-readable structured data using the schema.org vocabulary, emitted as JSON-LD (preferred over Microdata/RDFa because it's a self-contained script block, not interleaved with markup). The useful types for a marketing/content site:

  • Organization / Person / SoftwareApplication — define the brand and its named people/products as entities so they can be retrieved as known concepts.
  • Article with datePublished / dateModified — content identity + the freshness signal recency-weighted engines use.
  • FAQPage and HowTo — structured Q&A and step formats that map directly onto how engines extract answers.

How It Applies to Marketing Factory

Schema is the extractability plumbing in the ~3-hour llm-search technical baseline, and it's fully agent-automatable: the build pipeline emits per-page JSON-LD (entity blocks, Article dates, FAQ) from the same structured data it already holds, plus a sitemap.xml and an IndexNow push for immediate discovery. Entity definitions here are also what make content-authority machine-legible — an undefined entity can't accrue recognized authority. Declare crawler access in robots-txt so the engines that read this markup are allowed in.

  • llm-search — schema is core GEO extractability plumbing
  • content-authority — entity definitions make authority retrievable
  • robots-txt — ensure the crawlers that consume schema are permitted

Referenced from: llm-search-visibility-and-content-metrics