FindAstra adds AI semantic search to WooCommerce so shoppers find products by meaning, not exact keywords. Free local engine or one-time paid plans from $89.
FindAstra is a WordPress plugin that upgrades WooCommerce's built-in search from simple keyword matching to AI-driven semantic search. Instead of requiring exact word matches in product titles, FindAstra converts each product's title, categories, description, and custom fields into a numerical embedding (a "fingerprint"), then compares customer search queries against those embeddings using cosine similarity — so a query like "comfortable hiking shoes" can surface relevant footwear even if those exact words never appear in the listing.
Store owners choose from three interchangeable search engines: a free local engine that runs entirely in the shopper's browser using a quantized Hugging Face model (no API keys, no recurring cost), a Hugging Face-hosted engine with multilingual support for up to roughly 30,000 monthly queries, and an OpenAI-powered engine (text-embedding-3-small) for the highest accuracy on high-volume stores, at an estimated cost of about $2/year in API usage. On larger catalogs (5,000+ products) FindAstra uses a hybrid approach, narrowing candidates with MySQL full-text search before applying semantic re-ranking.
The plugin installs in about three minutes with no template editing, works with popular WooCommerce themes (Storefront, Astra, Kadence, Blocksy, GeneratePress, and others), and includes live autocomplete with product thumbnails, a fallback chain to pinned/featured/best-selling products when no semantic match is found, encrypted credential storage for API keys, and search logging to help store owners identify catalog gaps from zero-result queries.