Definition
Query decomposition is the process by which an AI system breaks a single complex question into smaller, answerable sub-questions before searching. A question that bundles several needs, a place, a constraint, and a comparison, gets split into separate retrievals, each aimed at one part. The model then assembles the pieces into a single answer.Why It Matters for AI Search
A buyer rarely asks one clean question. When someone asks an assistant for a walkable resort that suits a specific diet near good restaurants, the model does not run that as one search. It decomposes the request and retrieves for each part, so a business is found only if its pages answer one of those parts directly. Content that addresses single, specific questions is what survives decomposition. Content that answers everything vaguely is retrieved for nothing.Related Terms
Query understanding
Broader term
Query fan-out
See also
Multi-step retrieval
See also
Query expansion
See also
Semantic search
See also