Structured Data.

Every syntax, one response shape

Most "schema extraction" tools only handle JSON-LD. The reality is that production sites still use Microdata (legacy ecommerce), RDFa (older CMS templates), OpenGraph (social previews), Dublin Core (publishing), and Microformats (older blog templates). A complete audit needs all six. structured-data extracts all of them in one call with a consistent output shape.

curl -X POST https://api.crawlcrawl.com/v1/actors/structured-data \
  -H "Authorization: Bearer crk_..." \
  -d '{"url":"https://your-client.com/product/123"}'

# → 200
{
  "actor": "structured-data",
  "elapsed_ms": 114,
  "data": {
    "counts": {
      "json-ld": 2, "microdata": 0, "rdfa": 1,
      "opengraph": 1, "dublincore": 1, "microformat": 0
    },
    "data": {
      "json-ld":    [ { "@context": "https://schema.org", "@type": "Product", ... } ],
      "rdfa":       [ ... ],
      "opengraph":  [ { "og:title": "...", "og:image": "...", ... } ],
      "dublincore": [ { "elements": [ ... ] } ]
    }
  }
}

Backed by extruct

extruct (scrapinghub) is the same library Zyte uses in production and the same library most academic structured-data crawlers rely on. We run it as a Python sidecar accessible only to the crawler API — you get production-grade parsing without managing the dependency.

What's surfaced per syntax

• JSON-LD — every <script type="application/ld+json"> block, parsed and validated. Multiple blocks per page are returned as an array.
• Microdata — itemscope / itemprop trees, resolved to typed objects.
• RDFa — vocab + property triples flattened per item.
• OpenGraph — every og:* meta tag plus type-specific extensions (article:*, product:*).
• Dublin Core — DC.* meta tags + DCMI terms.
• Microformats — h-card, h-entry, h-feed (legacy IndieWeb).

When to use it

Rich-results readiness. Audit every product page for valid Product/Offer JSON-LD before Google's structured-data report catches issues at scale.

AI-search visibility. AI search engines (Perplexity, ChatGPT search, Claude) lean heavily on JSON-LD for entity extraction. Pages without it are invisible to the new search surface. Pair with render-diff to see how much of your schema is post-JS rendered (and thus skipped by non-rendering crawlers).

Migration verification. After a CMS or theme change, run structured-data on every template type to confirm schema didn't regress.

Competitor research. Pull the JSON-LD off a competitor's product page to see exactly what typed properties they emit.

Pricing

One page-credit per call. The $42 Studio tier includes 50,000 page-credits a month. See full pricing →

Where it fits