What was announced
Google Search Central announced on August 21, 2026, that it has changed how its crawler extracts and parses JSON-LD structured data from web pages. The parser now applies only a single pass of HTML unescaping when reading JSON-LD script blocks, aligning with the JSON specification defined in RFC 8259.
What changed
Previously, Google's JSON-LD extraction process applied multiple passes of HTML unescaping, which meant that double-escaped HTML entities in structured data were automatically corrected before being processed. For example, a value containing & would be unrolled through multiple passes until it resolved to a plain ampersand character.
Under the new behavior, only a single pass of HTML unescaping is applied. Double-escaped entities such as & or ✔ will no longer be fully resolved. Sites that relied on the previous multi-pass behavior to correct improperly escaped content in their JSON-LD markup may find that their structured data is now parsed differently.
Google said that sites using JSON-LD for structured data should update their code to use standard JSON escapes or Unicode hexadecimal escapes (such as & for an ampersand).
Details and availability
The change applies to all JSON-LD structured data that Googlebot encounters during crawling. It affects any structured data format that uses JSON-LD script blocks (<script type="application/ld+json">), including schema.org markup for products, reviews, recipes, FAQs, articles, and other rich result types.
Google's Gary Illyes referenced RFC 8259, section 7, as the authoritative specification for proper JSON string escaping. The RFC defines exactly which characters must be escaped in JSON strings and how escape sequences should be formatted.
Context
JSON-LD is the structured data format that Google recommends for marking up page content. It powers rich results in Google Search, including product listings with prices and ratings, recipe cards, FAQ expandable sections, and article metadata. Many content management systems and plugins generate JSON-LD markup automatically, and some have historically produced output with double-escaped HTML entities.
The change brings Google's parser into stricter compliance with the JSON standard. Sites that already produce correctly formatted JSON-LD with proper single-pass escaping are not affected. The impact is limited to sites whose structured data contains double-escaped entities that previously relied on Google's parser to correct them.
What has not been confirmed
Google did not specify whether the change is already fully deployed or still rolling out. The announcement does not describe a specific effect on rich result eligibility or search visibility for affected pages. It is not clear whether Google Search Console's structured data reports will flag newly broken markup resulting from this change.
