What is Vector Search?

Vector search is a retrieval technique that finds items by mathematical similarity in a high-dimensional embedding space, rather than by exact keyword match. Each item, such as a product or a document, is represented as a numerical vector. Queries are converted into the same vector format, and results are returned in order of distance to the query vector.

Definition

Vectors are produced by embedding models that capture semantic meaning, so "running shoes" and "trainers" land near each other in the space even if the strings do not match. Specialized vector databases store these embeddings and use approximate nearest-neighbor algorithms to keep response times low even with millions of items. The same approach works for text, images, audio, and mixed media.

Why it matters

Vector search makes storefront search more forgiving and more intuitive. Shoppers find relevant products even with vague queries, typos, or descriptive language that does not appear in product titles. It also underpins recommendation, deduplication, and content matching tasks across the commerce stack.

Use cases

Common applications include semantic product search, image-based search where shoppers upload a photo, content discovery across editorial and product pages, and the retrieval step inside retrieval-augmented generation pipelines that power AI assistants. In composable storefronts, vector search is exposed as a service that the frontend queries alongside or instead of classic keyword search.

Related

Explore Agentic Frontend Management Platform · Personalization.