Structured Data Analyzer

Schema.org markup is how you tell AI engines and search crawlers exactly what your brand, content and products are. This tool scans any URL, shows the JSON-LD it already has, detects your page type, and hands you copy-paste JSON-LD for the schemas you're missing.

Try:

We fetch the page HTML and parse its structured data. Nothing is stored.

Structured data is one of four signals we score. The full audit also checks AI model recall, crawler access and crawl budget, then returns a single graded report.

Run the full audit

Structured data reference

Schema.org vocabulary expressed as JSON-LD is the machine-readable layer of a page. Search crawlers and AI retrieval pipelines parse it to resolve entities without guessing from prose. These are the specifics this tool measures.

Supported serialisation formats

FormatDetectionRecommended
JSON-LD<script type="application/ld+json">Yes — Google's stated preference
Microdataitemscope / itemprop attributesLegacy; detected but not scored
RDFavocab / typeof attributesRare outside publishing

How the score is calculated

The score is valid blocks / (valid blocks + recommended blocks), expressed as a percentage with a floor of 10 once any valid JSON-LD exists. A page with no parseable JSON-LD scores 0. Recommendations are derived from the detected page type, so a product page is judged against Product and Offer while an article is judged against Article and Author.

Common failure modes

Why AI engines care

Retrieval-augmented systems need to know what an entity is before they can cite it. Organization with sameAs links to authoritative profiles, and FAQPage with question-answer pairs matching real user prompts, both raise the odds a model surfaces your brand instead of a competitor's.

Structured data is necessary but not sufficient. Perfect markup on a page that blocks GPTBot still gets you nothing.

Check all four signals at once
Copied to clipboard