Why malformed JSON-LD can look like missing JSON-LD

A safer mental model for structured-data inspection

A page can contain a JSON-LD script and still be reported as having “no JSON-LD.” The usual cause is not detection—it is a parser that silently discards the block after JSON.parse fails.

The parse-and-discard trap

A common implementation finds every <script type="application/ld+json">, parses each text value, and retains only successful results. This looks compact:

const schemas = scripts
  .map(script => {
    try {
      return JSON.parse(script.textContent || "");
    } catch {
      return null;
    }
  })
  .filter(Boolean);

The final array is empty when every block is malformed. If the interface uses that array to decide whether structured data exists, a syntax problem becomes a false “not found” result.

Detection, parsing, and validation are separate states

A reliable inspector treats these as three different questions:

1. Detection: Is a JSON-LD script present in the document?
2. Parsing: Is its text valid JSON, and where does parsing fail?
3. Validation: Does the parsed value follow the expected Schema.org or platform structure?

Skipping directly from detection to validated entities removes the evidence needed to answer the second question.

A source-preserving data model

Keep one record for every script before parsing it. Attach either a parsed value or a parse error, never replace the source:

{
  blockIndex: 1,
  raw: "{ \"@type\": \"Product\", ... }",
  parsed: undefined,
  parseError: {
    line: 7,
    column: 5,
    excerpt: "    \"priceCurrency\": \"USD\"",
    pointer: "    ^"
  }
}

Once parsing succeeds, normalize top-level arrays and @graph entries into inspection items while retaining their block index and JSONPath. The same mapping should also apply to nested typed entities such as an Offer inside a Product.

A five-minute debugging workflow

1. Confirm the script exists in the rendered DOM, not only in the original HTML response.
2. Copy the exact raw text and inspect the reported line and column.
3. Fix JSON syntax first: commas, quotes, escaping, and truncated output.
4. After parsing succeeds, inspect Schema.org structure and platform-specific requirements.
5. Run the live URL through an official validator before shipping.