What is schema.org?
The shared vocabulary most structured data is written in: an agreed list of kinds of thing, such as Organization, Article, Product or Event, and the properties each one can have. When a page describes itself in schema.org terms, any machine that knows the vocabulary reads the same meaning from it.
Who runs it
Google, Microsoft, Yahoo and Yandex started it together, so that a publisher could describe a page once rather than once for every search engine. Since 2015 it has been developed in the open through a W3C community group, with a small steering group overseeing it. Anyone can propose a change.
Types, and how they nest
Everything in the vocabulary is a kind of Thing, and types narrow from there. An Article is a kind of CreativeWork. A Restaurant is a kind of LocalBusiness, which is both an Organization and a Place — a type can sit under more than one parent.
Every type inherits the properties of the types above it, so a Restaurant can carry opening hours because it is a business, an address because it is a place, and a name because it is a Thing. The more specific the type, the more a machine knows about the page without having to work it out.
One vocabulary, three ways to write it
The vocabulary is separate from the syntax. The same facts can be written as JSON-LD, a block of data that sits on its own in the page; as Microdata, attributes added to the visible HTML; or as RDFa, a similar attribute-based form. All three describe the same data, so a machine that reads them learns the same thing.
JSON-LD is the most common today, because it keeps the facts in one place, apart from the design, where changing a template cannot quietly break them.
Connecting the pieces
A page usually describes more than one thing: the organisation behind it, the site, the page itself, perhaps an author. JSON-LD can hold them together in one @graph and give each an @id, an address that names it, so one entity can point at another instead of repeating it. That is how a machine learns that the publisher of an article is the same organisation the home page describes.
Being honest about it
schema.org is a vocabulary, not a promise about what any engine does with it. Each search engine and AI company decides which types it reads, and those choices change. A type nobody has promised to use still tells a machine plainly what your page is — and that is the part we check: that your markup is there, parses, and names the right things. Whether what it says is true is something only you can check.
Checks that measure it
Related terms
Not sure where yours stands?
A free scan checks your page the way an answer engine reads it, in about 20 seconds — no signup.
Rather have it handled? We do that too. No pitch, just a plain-English chat.
Book a call