In search, an entity is a distinct thing that exists independently of the words used to name it: a person, a place, an organisation, a product or an abstract concept. Google's Knowledge Graph is built from entities and the relationships between them.
An entity is singular, well defined and distinguishable from everything else. It is not a keyword. The strings Apple, apple and the fruit itself may share letters, but they map to different entities, and a search engine that understands the difference can disambiguate a query without relying on exact word matching. Entities are connected by typed relationships — founded by, located in, is a kind of — that form a machine-readable graph.
Once a search engine resolves a query to an entity rather than a string, exact keyword matching stops being the mechanism that decides relevance. This is what allows results to be returned for synonyms, translations and paraphrases, and it is the foundation of knowledge panels, rich results and AI-generated answers.
Help search engines resolve your entities by marking them up with schema.org types, using consistent naming everywhere you appear, and linking to authoritative references such as Wikipedia or Wikidata through sameAs properties. Consistency of name, address and phone number across the web is entity work even when it is not called that.
Want to learn more? Explore our SEO Glossary or check out our expert guides on SEO strategies.