A product catalogue, a documentation site and a multi-tenant application ask different things of a search engine. Each page below walks through one job: the data model, the requests, and the parts of the platform it leans on.
Index Markdown, HTML or a sitemap, filter by version and product, and add cited AI answers and an MCP server for coding agents on the same index.
Catalogue search with facets, stock-aware filtering, merchandising rules and revenue analytics — fed by CMS modules, a YML feed or the API.
Isolate customers by collection or by a filter forced on the key, mint short-lived browser keys per session, and keep quotas and analytics per tenant.
Solutions are organised by the kind of business you run — a shop, a marketplace, a publisher, a SaaS product. Use cases are organised by the job the search has to do, whatever the business. Most teams need one of each.
Explore more
The best way to evaluate a search platform isn't reading one more comparison table.
Only then make the decision.