Structured Data Change Monitor & Rich Result QA

Audit JSON-LD and schema.org markup on public web pages, create a persistent baseline, and detect regressions on later scheduled runs.

This Actor is built for technical SEO monitoring, storefront QA, CMS release checks, and agent/CI pipelines that need stable issue codes instead of a prose-only audit.

Why use it

• Audit one page, a batch of URLs, or public XML sitemaps.
• Parse JSON-LD, including nested nodes and @graph structures.
• Detect Microdata and RDFa schema types.
• Run deterministic checks for Product, Article, BreadcrumbList, FAQPage, and LocalBusiness markup.
• Track schema types, issue codes, fingerprints, QA scores, and HTTP state between runs.
• Emit only changed pages after the baseline if desired.
• Use baseline_only to avoid flooding the first scheduled run.
• Export stable dataset records through Apify API, webhooks, schedules, Zapier, Make, or MCP.

Input

Supply urls, sitemapUrls, or both. The Actor deduplicates targets and applies maxPages after sitemap expansion.

{
  "urls": ["https://example.com/product/widget"],
  "sitemapUrls": [],
  "maxPages": 100,
  "monitorChanges": true,
  "onlyChanges": false,
  "firstRunMode": "emit_current",
  "stateStoreName": "structured-data-monitor-state",
  "stateNamespace": "storefront-production"
}

For scheduled regression alerts, use:

{
  "sitemapUrls": ["https://example.com/sitemap.xml"],
  "maxPages": 500,
  "monitorChanges": true,
  "onlyChanges": true,
  "firstRunMode": "baseline_only",
  "stateNamespace": "storefront-production"
}

Output

Each emitted dataset item contains:

• requested and final URL, HTTP status, title, canonical, and robots metadata;
• JSON-LD, Microdata, and RDFa schema types;
• JSON-LD block count;
• deterministic QA score and rule-based rich-result readiness flag;
• issues with code, severity, message, and associated schema type;
• a content fingerprint;
• change details: added/removed types, new/resolved issue codes, and score delta;
• check timestamp.

The OUTPUT key-value-store record summarizes target count, emitted items, baselines, suppressed unchanged pages, failures, and page-limit capping.

Monitoring behavior

State is stored by namespace and URL in the named key-value store.

• emit_current: the first run emits current audits.
• baseline_only: the first run records state without dataset output.
• onlyChanges: true: later unchanged pages are suppressed.
• Failed fetches are reported in the run summary and never silently turned into successful audits.

Safety and limits

• Only public HTTP(S) targets are accepted.
• Local/private IP ranges and hostnames are blocked before requests.
• Every redirect is revalidated.
• DNS results are checked before fetching.
• Per-request timeouts, retries, response-byte limits, and a page ceiling are enforced.

The readiness result is a deterministic rule set, not a claim that Google will show a rich result. Search engines apply additional eligibility, quality, policy, indexing, and site-level criteria.

Pricing

Recommended Store pricing: $2.00 per 1,000 audited page records, with a minimal Actor-start event. Platform compute is included.

Source basis

The checks are designed around public Schema.org vocabulary and Google Search structured-data guidance. They deliberately produce explicit issue codes and avoid claiming to reproduce Google's private ranking or indexing systems.