Structured data validator

The structured data validator checks that markup on a page is recognised correctly. It supports microformats, Schema.org, Open Graph, HTML microdata and RDFa, and accepts either a URL or pasted code.

Why use this rather than a generic validator

Yandex states the difference explicitly: unlike Validator.nu, Structured Data Linter and others, its validator also checks whether the markup meets the requirements of Yandex’s own services, including all the additional types. So it catches things no general tool can — the Yandex-specific resultPhoto field on recipes, the schema types actually supported for organisations, the field combinations required for a product snippet.

Markup that passes a generic validator and produces nothing in Yandex is a routine outcome. This tool is where that gap becomes visible.

The errors

• Page does not load — it does not exist, or is unavailable to the robot. Worth noticing on its own: markup is irrelevant if the page cannot be fetched.
• Micro-markup not detected — no root element was found, or it was set up incorrectly. Yandex’s example is a Schema.org root without the itemscope attribute, which is the most common structural mistake: properties are labelled but nothing declares what they belong to.
• Field missing or empty — a required property is absent or has no value. On generated markup this usually means a template emitting an empty attribute where the data was null.

How to use it in practice

Paste code to test before deploying, or enter a URL to check what is live. Run it again after any template change: markup is the first thing to break silently in a redesign, because nothing about the page looks different when it disappears.

The limitation to remember

Valid markup is necessary and not sufficient. Data must also be indexed and, for partner programmes, pass moderation — typically about two weeks — before anything appears in results.