Entity
An entity is a distinct, well-defined concept or thing, such as a person, place, organization, or object, that can be uniquely identified independent of language or phrasing. Search engines map entities and their interrelated attributes into a knowledge graph to understand meaning, context, and user intent beyond literal keywords.
In plain English
In early search engine technology, search systems were purely lexical. They treated language as a flat series of character strings. If you typed the word “jaguar” into a search box, the engine simply scanned its index for pages containing the letters “j-a-g-u-a-r.” The system had no native way to tell whether you were researching the wild predatory cat, the luxury automotive manufacturer, the vintage electric guitar, or an operating system release.
An entity represents a fundamental shift in search technology from “strings to things.” An entity is an identifiable concept, person, place, organization, or object that exists in the real world. In a search engine’s knowledge graph, every entity is assigned a unique permanent identifier along with defined attributes, characteristics, and relationships to other entities. By analyzing entities rather than plain text strings, search algorithms understand the real-world meaning, context, and intent of human language.
An example
Consider how an entity-driven search engine processes the conversational query: “who played iron man in the marvel movies.”
A traditional lexical search engine would search for documents containing that exact sentence structure.
An entity-based search engine resolves the query by traversing connected nodes in its knowledge graph:
- Entity recognition: The system extracts the entity “Iron Man” (the Marvel Comics superhero entity, represented by identifier
/m/03s2_). - Contextual disambiguation: The mention of “marvel movies” connects the comic book entity to the “Marvel Cinematic Universe” film series entity.
- Graph relationship traversal: The algorithm queries the knowledge graph relationship “portrayed_by” specifically for the film character “Tony Stark.”
- Node resolution: The query directly resolves to the person entity “Robert Downey Jr.” (identifier
/m/017m_b).
Because the search engine understands the underlying real-world entities, it does not just return blue links. It generates an interactive Knowledge Panel featuring Robert Downey Jr.’s biography, image gallery, and filmography directly at the top of the results page. If the user subsequently enters a follow-up query like “how tall is he,” the search engine remembers that “he” refers to the entity Robert Downey Jr. and answers instantly without requiring the user to repeat the actor’s name.
Why it matters
Entity-based understanding powers modern semantic search, Knowledge Graph panels, voice search responses, and AI overviews. Modern search optimization requires moving beyond repetitive keyword targeting. Webmasters must clearly define their organizations, authors, products, and concepts using structured data (such as Schema.org markup) so that search engines can accurately map their content to recognized global entities.
Related terms
Read the full guide to search queries, or explore our complete guide to entities in search.