Free tool · JSON-LD
Does your markup actually parse?
A JSON-LD block with a syntax error is discarded silently. Nothing warns you, the page looks fine, and every engine reading it sees nothing at all.
01Why it matters
Invalid markup fails quietly, which is the worst way to fail.
A broken JSON-LD block does not render an error or degrade gracefully. Parsers drop it whole, so a page can carry beautifully considered structured data and be read as having none.
The second failure is subtler. Schema that describes a page is table stakes; schema that describes the organisation, with a stable @id every other node points at, is what lets an engine treat you as a business it can hold facts about rather than a URL it happened to find.
This reads your homepage the way a parser does and reports what survives: which blocks parse, which types resolve, whether there is an Organization node with sameAs links, and whether anything references an @id that nothing on the page defines.
02Questions
- What is JSON-LD?
- JSON-LD is a format for stating machine-readable facts about a page and the business behind it, embedded in a script tag. Search engines and AI assistants parse it to understand what a page is, who published it, and how it relates to other things they know.
- Does schema still matter now Google shows fewer rich results?
- For rich results, less than it did. For assistants, more. Models parsing pages to build answers still read the graph, and it is the cheapest way to state facts in a form that does not depend on prose being interpreted correctly.
- What is a dangling @id?
- A node that references another node by @id when nothing defines that @id. It is valid JSON and broken meaning: the engine is told to look something up and finds nothing, so the relationship you intended silently does not exist.
- Why does sameAs matter?
- It resolves ambiguity. Without sameAs links an engine has no way to know that your company and a similarly named one are different entities, and merging them is materially worse than knowing nothing about you.
Next
Markup is half of being machine-readable.
The other half is whether the content is in the server response at all, whether the crawlers can reach it, and whether anything on the page is worth quoting. The full check scores all four.