Schema.org JSON-LD present: what the check tests and how to fix it

Counts the share of crawled pages with a parseable JSON-LD block in the served HTML. Only all pages passes. Google says AI features need no special markup.

Structured Data · Weight 2 in the w2agent Score

How common is it?

35% of 1223 measured sites where this check applied pass.

Latest audit of each site under rubric v3.1. Sites where the check did not apply or could not be measured (for example, no llms.txt for the llms.txt checks) are left out. A partial result does not count as a pass.

What this check tests

The audit looks at every page it crawled: the homepage plus a sample of same-site pages found through your sitemap and links (the standard audit profile caps the crawl at 10 pages). For each page it reads the HTML as served, finds every <script type="application/ld+json"> element and tries to parse its content as JSON. A page counts when at least one block parses. A block with a syntax error is dropped without a message, so a page whose only block is broken counts as having none.

The score is the share of crawled pages that count, rounded to a whole percent.

  • Pass: every crawled page has at least one parseable block (100).
  • Partial: some, but not all, pages do.
  • Fail: no page does.

The check does not look at what the block says. Types, fields and @context are judged by Appropriate Schema.org type, Schema.org completeness and JSON-LD valid. Microdata and RDFa markup are not counted, and neither is JSON-LD that a script adds in the browser, because the audit does not run JavaScript.

Why it matters to agents

Structured data is machine-readable statements about a page, such as its type, its title, its author or a product's price. Google describes it as a standardized format for providing information about a page and classifying the page content, and recommends JSON-LD when your setup allows it. Schema.org says the vocabulary can be written as Microdata, RDFa or JSON-LD.

The payoff for AI answers is unproven. Google's documentation on AI features says you do not need new machine-readable files, AI text files or markup to appear in AI Overviews or AI Mode, and that there is no special schema.org structured data to add. It does say to make sure your structured data matches the visible text on the page. So treat this check as general hygiene that helps search engines and any tool that parses JSON-LD, not as a lever for chat assistants. It has weight 2, the highest of the seven Structured Data checks.

How to fix it

Add one JSON-LD block to the HTML your server returns for each page template. Start with a type that fits the page. The example is for an article.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "How we price the Team plan",
  "description": "What the Team plan includes and how seats are billed.",
  "datePublished": "2026-03-02",
  "author": { "@type": "Person", "name": "A. Writer" }
}
</script>
  • Describe only what a reader can see on the page. Google's guidelines say not to mark up content that is not visible to readers, and that missing required properties make an item ineligible for rich results.
  • Generate the block on the server or at build time. If your framework only injects it in the browser, this check will not see it. Look at the page source (not the inspector) to confirm it is there.
  • Put the block on every template, not only the homepage. One uncovered section is enough to turn a pass into a partial.
  • Check the syntax. A trailing comma or an unescaped quote makes the block vanish from this check. Google names its Rich Results Test as a tool for validating structured data.

After the blocks parse, pick sensible types with Appropriate Schema.org type and fill in the fields with Schema.org completeness. For background on what structured data does and does not do for AI, see Schema.org for AI.

Examples from measured sites

Up to 10 of the highest-scoring sites in each group, from recent audits. Failing includes partial results.

Passing

Failing or partial

Sources

Verified against these sources on

Does your site pass this check?

Check your own site →