Skip to main content

Vector search

Searching by semantic meaning using embeddings: numerical representations of text that capture meaning.

In depth

Products, questions, and policies are converted to high-dimensional vectors by an embedding model. A query is converted the same way. The search finds the closest vectors in space. This is how 'gift for dad' can return tools, grilling gear, or headphones without any keyword overlap.

Related glossary entries

Hybrid search
A product search strategy that combines semantic (vector) search with keyword (BM25) search and fuses the results.
BM25
A classical keyword-based ranking algorithm. Scores documents by how often query terms appear, weighted by rarity.
Product recommendation AI
Software that suggests relevant products to a shopper based on their intent, context, or history. In a chatbot context, the AI picks one or …
Semantic search
Search that matches the meaning of a shopper's query against products in your catalog, not just keyword overlap. A search for 'something warm for …
← Back to full glossary

Try it on your store

Connect your products and see the chat answer shopper questions from your own catalog.

Your dashboard shows what each conversation led to.

  • Free to start
  • Pay per use, no seat fees
  • No card required
Create free account Try the Live Demo
Contact Us