Iris · AI SEO, AEO & GEO auditor

← Back to Iris

Free tool · no sign-up

Schema checker: is your structured data readable, and does it say who you are?

Paste a page. Iris reads the JSON-LD it sends, tells you which blocks do not parse and why, and whether the schema says who is behind the page and what the page is.

Free, no sign-up. Reads one page, as IrisBot. This is not Google’s Rich Results Test: it does not validate every property.

What does the schema checker look at?

The structured data in the HTML your server sends, checked the way the Iris audit scores it.

  • Does every JSON-LD block parse? A block that does not is named, with the likely cause.
  • Which types does the page declare?
  • Does the schema say who is behind the page: an Organization, a Person, a LocalBusiness or a WebSite?
  • Does the page have the type that fits what it is, for example Article on an article or Product on a product page?
  • On an article, does the schema name an author and a date?

Why does one broken JSON-LD block matter?

Because it fails silently. The page looks marked up, the code is there, and every reader ignores the whole block. The usual causes are small:

  • a comma before a closing bracket,
  • curly quotation marks pasted from a document,
  • single quotes, or a name with no quotes at all,
  • a comment left inside the block,
  • a bracket opened and never closed.

The checker names which of these it found and shows the start of the block so you can find it in your template.

How is this different from Google’s Rich Results Test?

Google’s test is the better tool for one job: it knows which properties each rich result needs and tells you whether a page is eligible for one. Use it for that. This checker answers a different question, the one Iris scores: does the markup parse at all, and does it tell a machine who you are and what the page is. A page can pass one and fail the other, so it is worth running both.

Does schema make an AI assistant cite a page?

Not by itself. Structured data helps a machine understand a page and who published it, and it is worth 38 of the 100 points in the Iris AI-search score: 15 for JSON-LD that parses, up to 15 for types such as FAQPage, Article, Product or BreadcrumbList, and 8 for saying who is behind the page. The other 62 points are the content, the headings and whether crawlers can reach the page. No markup guarantees a citation.

One limit to know: this reads the HTML your server sends, which is what most AI crawlers work from. Schema added later by a script may not be seen by them, and is not seen here.

Questions people ask about this

Is the schema checker free?
Yes. There is no sign-up and nothing to pay. One connection can run 40 checks a day.
Which formats does it read?
JSON-LD, which is the format Google recommends and the one the Iris score counts. If a page has microdata and no JSON-LD, the checker says so and names the types it found.
Does it replace Google's Rich Results Test?
No. Google's test checks the properties each rich result needs. This checker tells you whether the markup parses, which types it declares and whether it says who is behind the page. They answer different questions.
Why does it say schema is missing when my site has it?
Usually because the schema is added by a script after the page loads. The checker reads the HTML the server sends, as most AI crawlers do, so markup added later is not seen. Putting the JSON-LD in the HTML itself fixes that.
Should I add FAQPage schema to every page?
No. Add a type only where the page really has that content. FAQPage belongs on a page whose questions and answers are visible on it.

The other free tools

The checks and the points are those of the Iris audit, described on How Iris works. For rich result eligibility use Google’s Rich Results Test.