Structured Data Audit (JSON-LD / schema.org)

What do search engines and AI actually understand about this page? Structured data is how a page tells machines what it is — an article, a product, an organization, an FAQ. This Actor extracts a page's JSON-LD, validates it, reports the schema.org types present, and flags the missing recommended properties that quietly disqualify a page from rich results and weaken how AI assistants ground on it.

What you get

Summary plus one row per typed object:

{
  "url": "https://example.com/product/",
  "jsonld_blocks": 2,
  "jsonld_parse_errors": 0,
  "schema_types": ["BreadcrumbList", "Product", "Offer"],
  "incomplete_types": 1,
  "has_open_graph": true,
  "fixes": ["Product is missing recommended property: image"]
}

{ "schema_type": "Product", "complete": false, "missing_recommended": ["image"] }

Why it matters now

Rich results (stars, prices, FAQs in search) depend on complete, valid structured data — and AI assistants increasingly read JSON-LD to understand and cite pages accurately. A single invalid JSON-LD block is skipped silently; a missing recommended property costs a rich result nobody knows they lost. This audit surfaces both.

Use cases

• SEO teams: verify rich-result eligibility across templates; catch a JSON-LD change that broke parsing.
• Agents grounding on pages: check whether a page exposes machine-readable facts before relying on scraped text.
• Publishers optimizing for AI answers: complete structured data is one of the few levers that helps.

Honesty notes

• Audits JSON-LD (validated), and detects the presence of microdata, RDFa, and Open Graph.
• Reads the static HTML — the same view search and AI crawlers get; JSON-LD injected purely client-side is out of scope, as it is for those crawlers too.
• Unreachable pages report cleanly and you are not charged; billing is per completed audit.

Built by 1450 Enterprises, alongside the AI Access Audit and llms.txt Generator.