Schema.org structured data, usually implemented as JSON-LD, tells search engines explicitly what a page is about: a product, an article, a local business, a recipe, an FAQ. It is also one of the most commonly mistranslated or entirely ignored parts of a multilingual site, either left in the source language while everything else is translated, or broken by translating fields that should stay structural. Getting multilingual schema right helps translated pages qualify for the same rich results, star ratings, FAQ accordions, product pricing, that the source-language version gets.
What in a Schema Block Should Be Translated
| Schema Field Type | Translate? |
|---|---|
| @type (Product, Article, FAQPage, etc.) | No; this is a structural identifier, not visible content. |
| name, headline, description | Yes; these are the text properties search engines display in rich results. |
| url | No, but should point to the translated version of the page, not be translated as text. |
| FAQPage question/answer text | Yes; these drive FAQ rich results and need to match the visible translated FAQ content exactly. |
| price, priceCurrency, availability | Currency code stays structural; availability values follow schema.org’s fixed vocabulary rather than being translated. |
Why Schema and Visible Content Need to Match
Search engines cross-check structured data against a page’s visible content, and a mismatch, for example an FAQPage schema block still in English on a page whose visible FAQ has been translated to French, can cause the rich result to be ignored or, in more serious mismatch cases, treated as a quality signal problem. Schema translation is not optional polish; it needs to be treated as part of the same translation job as the visible page content, updated together rather than the visible content being translated while the schema block is forgotten.
Common Multilingual Schema Mistakes
- Leaving the entire schema block in the source language while translating the visible page content.
- Translating structural fields like @type or fixed-vocabulary values, which breaks schema validity.
- Pointing the schema’s url field at the source-language page instead of the translated version.
- Letting FAQPage or HowTo schema drift out of sync with edits made to the visible translated content.
How GPTranslate Handles Multilingual Schema
GPTranslate parses schema markup at the field level, translating text properties while leaving structural fields and fixed-vocabulary values untouched, and keeps schema URLs pointed at the correct translated page. This follows the same field-aware approach used for RankMath and Yoast schema settings, and ties into the broader multilingual SEO structure covering hreflang, canonical tags and sitemaps.
Frequently Asked Questions
Does every page need schema markup in every language?
Any page that has schema markup in the source language should have equivalent, translated schema markup in every language version, since the whole point of schema is to help search engines understand the page a visitor is actually looking at.
Can mismatched schema hurt a page’s search ranking?
Significant mismatches between structured data and visible content can affect whether rich results are shown and, in some cases, are treated as a quality signal, which is why schema should be kept in sync with the translated visible content rather than left stale.
Conclusion
Structured data is part of what a translated page communicates to search engines, not a separate technical layer that can be left behind. Treating schema translation as part of the same job as translating visible content, with the same field-level care used for other metadata, is what lets translated pages earn the same rich results as the original.