Why it matters
For buyers, it's mostly a term to decode in proposals: "we'll index your content in a vector database" translates to "we'll make your knowledge semantically searchable so the AI can ground its answers in it". Legitimate architecture, not pixie dust, and worth knowing the plumbing has commodity options.
How it works
Content is chunked, embedded and stored; queries are embedded and matched by similarity in milliseconds, even across millions of chunks. Products range from dedicated services to extensions of ordinary databases: for modest content volumes, simple options do fine.
What to do about it
If you're commissioning a knowledge tool, the questions that matter: what content gets indexed, how often it refreshes, and what happens when the source documents change. The database brand matters far less.
