The abbreviation unpacks as JavaScript Object Notation for Linked Data, which explains the format and says nothing about the use. In practice it is a block of structured data placed in a page’s source that describes what the page is about, in a vocabulary search engines already understand. It is the format Google recommends, and it has one property that makes it far easier to live with than the alternatives.
Briefly explained
- The recommended format for structured data.
- It occupies a block of its own, holding nothing a visitor ever sees.
- That separation means the design can change without breaking it.
- It describes the page. It does not change what the page says.
Why is it preferred over the older formats?
Because it does not have to be woven into the HTML.
Microdata and RDFa annotate the visible markup, which means every design change risks damaging the data. JSON-LD sits in its own block, so the two can be maintained independently. For a site where templates change, that is the whole argument.
What should a local business describe with it?
The facts somebody would otherwise have to infer.
- The business: name, address or service area, phone, hours
- The services: what you sell, and where you sell it
- The questions: where they genuinely appear on the page
- The author: for articles written by a named person
See schema markup for which types are worth the effort.
What are the rules that actually matter?
- Describe only what is on the page
- Keep the entities connected: the service belongs to the business, the article to its author
- Validate after every template change, because markup breaks silently
- Never mark up reviews, prices or questions a visitor cannot see
How do you validate it?
There are two, and they answer different questions.
The Rich Results Test tells you what Google can parse and what the page qualifies for. The Schema Markup Validator checks the vocabulary itself, including types Google does not use. Run both after a template change, not after a traffic drop.
What do the terms mean?
- Linked data: Data that connects to other data through shared identifiers.
- @type: The kind of thing being described.
- @id: A stable identifier, used to connect entities across pages.
- Microdata: The older format, written into the visible HTML.
- Rich Results Test: Google’s tool for checking what a page qualifies for.
Owner questions
