Embedding
A numerical representation of a passage's meaning, used to match a question against your content. It is why a page can be retrieved for a question that uses none of its exact words.
Updated August 21, 2026
Embeddings mean keyword matching is no longer the whole game: a page about “winter tyre pressure” can surface for “how hard should my tyres be when it is cold”. The practical lesson is the opposite of keyword stuffing. Write the way your customers actually ask, cover the meaning fully, and stop contorting sentences around a phrase — the match happens on sense, not on spelling.
Related terms
- Retrieval. The step where a system finds and fetches your page. If a page is blocked, unreachable, or clearly not about the topic asked, it is never retrieved, and nothing else matters.
- RAG (retrieval-augmented generation). The standard way answer engines work: fetch relevant documents first, then write an answer from them. It is the reason your page content, not the model's training, decides whether you are quoted today.
- Prompt coverage. How much of what people actually ask about your kind of business your pages answer. A gap in coverage is a question where an AI system has to use somebody else's site.