Definition
An entity is a uniquely identifiable thing or concept: a person, brand, product, place, organisation or idea that a search engine or language model can recognise independently of the exact words used. 'Google', 'Amsterdam', 'Shopify' and 'core web vitals' are entities. The word 'jaguar' is not an entity but a string, because it could be an animal, a car brand or an operating system.
Search engines have moved from strings to things over the past decade. Google's Knowledge Graph links billions of entities to each other through relationships ('is part of', 'located in', 'founded by'). Language models do something similar in their internal representation: they learn which names go with which properties and which other names. A page that consistently names the right entities is therefore easier to connect to a topic than a page that writes around it.
For SEO and GEO, this means you do more than place keywords: you make clear which things your page is about and how they relate. You do that by naming entities in full, avoiding a different variant every time, and making relationships explicit. Structured data helps: with schema.org types such as Organization, Product and Person, and with sameAs references to Wikidata, LinkedIn or an industry body, you tell machines exactly which entity you mean.
Your own brand is an entity too, and possibly the most important one. Whether an AI engine mentions you for a question like 'which tools automate SEO' depends on how well the model knows your brand and which properties it associates with you. That brand picture is built from your own site, but even more from consistent mentions elsewhere: reviews, directories, articles, podcasts and partner pages that repeat the same name, the same offer and the same facts.
Traze builds an entity profile per customer — products, services, locations, people and related concepts — and uses it as a fixed reference while writing, so every page uses the same names and relationships.