In short. RDFa expresses structured data through HTML attributes that connect visible content to vocabulary terms and resource identifiers.

Definition: RDFa (Resource Description Framework in Attributes) is a syntax for embedding machine-readable structured data in HTML attributes. It expresses RDF statements about things and their relationships while reusing the page's visible content.

Key facts

W3C publishes RDFa Core 1.1, Third Edition as a Recommendation dated 17 March 2015, specifying the syntax and processing rules.
The RDFa Lite specification defines a subset with five attributes: vocab, typeof, property, resource, and prefix.
The W3C Primer explains that RDFa-specific attributes leave the visual display of HTML content unchanged.
As of 17 October 2026, Google's structured data documentation lists RDFa among its supported formats, alongside JSON-LD and Microdata.

What does RDFa look like in HTML?

The RDFa Lite specification illustrates a person using the Schema.org vocabulary, a collection of terms for describing things. The following shortened adaptation retains the name and identifier from its example:

<p vocab="http://schema.org/" resource="#manu" typeof="Person">
  My name is <span property="name">Manu Sporny</span>.
</p>

Here, vocab sets the default vocabulary address. typeof identifies the thing as a Person, and property labels the visible name. The resource value identifies the person using a fragment appended to the page address. The example describes a person within a page, with an identifier that other pages can reference.

RDFa Lite also supports values taken from links and images through the HTML attributes href and src. Its specification demonstrates property="url" on a link to a person's homepage. The link's destination supplies the structured value, while the link text remains readable.

How does RDFa express linked data?

RDF, the Resource Description Framework, represents information as statements called triples. Each triple has a subject, a property, and an object: the thing being described, the relationship, and its value. The RDFa Core specification explains how web identifiers distinguish subjects and properties that might otherwise have ambiguous names.

The W3C Primer shows how a prefix declaration abbreviates vocabulary addresses. For example, prefix="dc: http://purl.org/dc/terms/" allows property="dc:title" to identify the Dublin Core title property. This lets a document combine terms from different vocabularies while keeping their identities explicit. RDFa supplies the attribute syntax; the selected vocabularies supply the terms.

How is RDFa checked for Google Search?

Google's structured data introduction recommends the Rich Results Test during development and rich result status reports after deployment. Rich results are enhanced search appearances supported by particular structured data types. Their required properties and implementation rules are documented in Google's feature guides.

For a website using RDFa, checking those requirements accompanies checking the attributes themselves. Google's documentation also requires structured data to describe the content of the page where it appears and advises against marking up information hidden from users. The applicable feature guide determines which properties Google requires, recommends, or treats as optional.

FAQ

Is RDFa the same as Schema.org?

RDFa is the attribute syntax. Schema.org is one vocabulary you can use with it; the W3C Primer also demonstrates combining Schema.org and Dublin Core terms.

Does Google prefer RDFa or JSON-LD?

Google recommends JSON-LD when your site's setup allows it, citing ease of implementation and maintenance. It accepts RDFa when the markup meets the relevant feature documentation.

Can RDFa Lite be extended?

The RDFa Lite specification permits a transition to full RDFa by adding further RDFa attributes to existing markup. A document using those additional attributes is described as RDFa rather than RDFa Lite.

Sources

Check how a page reads to answer engines. Open the GEO score analyzer