Schema.org

What it means

Schema.org is a shared vocabulary of structured data supported by the major search engines. It lets a page mark up in its code that it describes a business, product, article, review, FAQ, event or term definition, and add properties such as address, price or author.

The vocabulary arose from an agreement among the big search engines so that websites would not need a different format for each of them. It is most often written as JSON-LD in the page head, separate from the visible content, which makes maintenance easier.

Structured data using Schema.org help machines understand content and enable rich results in search. They are also increasingly important for AI search and assistants, which draw precise facts about businesses and products from them.

How it works in practice

For an accounting or advisory firm, useful types include Organization or LocalBusiness with address, contacts and opening hours, Service for services, Article for expert articles and DefinedTerm for a glossary. An online shop mainly uses Product, Offer and AggregateRating.

The data should be generated automatically from the system running the website so they are always current. Hand-maintained markup quickly goes stale, for instance when a price or address changes, and mismatches with the visible content can do harm.

After deployment, markup is validated in testing tools and monitored in the webmaster console. On a multilingual site, the data are given in each version's language along with a language tag.

What to watch out for

Markup must match the visible content of the page. Invented data, such as ratings not shown on the page, or marking up content visitors cannot see, breaches search engine guidelines and can lead to penalties.

Not every type in the vocabulary produces a rich result. Search engines support only selected types and change their rules. Structured data are valuable even without rich results, but they should not be expected to work instant miracles for traffic.