A knowledge graph is a large collection of facts about things in the world and the relationships between them. Instead of words on a page, it stores entities, such as people, businesses, places and products, together with their properties and connections. "Bath", for example, can be a city, a bathroom fixture or a brand name, and the knowledge graph helps a system keep those meanings apart.
The best known is Google's Knowledge Graph, whose results you see in the knowledge panel next to the search results. But other major players also build their own knowledge bases, and open sources such as Wikidata play a role as a shared reference.
In essence, a knowledge graph consists of three elements: an entity, a property and a value. For example: a fictional "Wheatsheaf Bakery" has the type "bakery", is located in "Bristol" and sells "sourdough bread". Each of those values can itself be an entity with its own relationships. That creates a network in which a search engine not only recognises text, but also understands what that text is about.
Google draws those facts from various sources: public databases, trustworthy websites, business profiles and structured data. Exactly which sources, and how much weight they carry, Google doesn't fully disclose. What is clear is that information repeated consistently in several independent places is more easily accepted as a reliable fact than a one-off claim on a single website.
If your business is recognised as a clear entity, a search engine can better link you to the right searches, distinguish your name from similar names and show your details correctly. That's especially important for branded searches: anyone searching for your business name wants to see the right information straight away.
The language models behind ChatGPT, Gemini and other assistants learn from huge amounts of text and often supplement that with search results. An LLM doesn't work the same way as a knowledge graph, but both benefit from consistent, unambiguous information about your business. In AI Overviews, Google combines its language models with its own search systems, in which entity information plays a part. The more sources tell the same story about who you are and what you do, the smaller the chance of confusion in AI answers. That is a core idea of GEO.
A roofer in Manchester with the same name as a company in Dublin, for example, has every interest in search engines keeping the two apart. Consistent details and a clear regional focus help with that. On the platform for visibility in search and AI systems, you can see how your business is recognised today and where the information about you still differs.
Request your free AI visibility plan. An expert analyses your business and contacts you within 1 working day to present your plan in person.
