Configure Meilisearch for ecommerce product search with facets, custom ranking, and filterable attributes

Steps

1. Create or update the index settings via PATCH /indexes/{uid}/settings, setting filterableAttributes to all fields you want to facet or filter on (e.g., ['brand', 'category', 'color', 'price', 'in_stock']) and sortableAttributes for fields users will sort by (e.g., ['price', 'rating']).
2. Add custom ranking rules to the rankingRules array after the six built-in rules; for ecommerce, common additions are 'desc(in_stock)', 'desc(rating)', or 'desc(sales_count)' — limit total rules to 8 to avoid performance degradation.
3. Index products via POST /indexes/{uid}/documents with an array of product JSON objects; each must include the configured uid field as the document identifier.
4. At search time, send POST /indexes/{uid}/search with a body including q (search term), filter (e.g., 'in_stock = true AND price < 100'), sort (['price:asc']), and facets (['brand', 'category']). The response includes hits, facetDistribution (counts per value), and totalHits.
5. For Meilisearch Cloud, authenticate requests with a bearer token passed in the Authorization header (Authorization: Bearer YOUR_KEY); self-hosted instances can also run without authentication for dev environments.