Smarter results with vector search

Keyword search looks for matching terms — it works best when shoppers know what to type, like a product name or SKU. Vector search, on the other hand, understands natural, conversational queries like “cheap gift for mom,” matching intent with relevant products even when keywords don’t align. It helps avoid zero results, reduce shopper frustration and increase conversions. Explore the full side-by-side comparison in our blog post.

Vector search uses AI to translate complex data, like products, texts and images, into numbers (vectors). These are stored in a vector database so that when someone searches with vague or complex input, the system can quickly surface similar content based on semantic similarity, not just matching words. Learn more about how this works in our deep-dive blog post.

Keyword search remains the best tool for most queries — especially when shoppers use exact terms like product names, SKUs, or brand filters. But when queries are vague, conversational, or incomplete, vector search steps in to understand intent and return relevant results that keyword matching would miss. Read more about how vector search complements keyword search in our announcement blog.

Yes, vector search is valuable in B2B eCommerce, where buyers sometimes describe what they need for specific use cases without knowing exact product names or specs. It understands functional, use-case-based queries like “sleeve for high-voltage cables” or “chemical-resistant gloves for cleaning tanks,” helping users self-serve with confidence. See more ways we support B2B eCommerce here.