Build / Discovery
Search & Discovery
Faceting, refinement and relevance that match how people actually search a large catalogue.
Search is where a large catalogue either earns its keep or buries itself. The technical part — indexing, faceting, refinement — is well understood. The hard part is that relevance is a merchandising decision wearing an engineering costume, and the two teams rarely have the same vocabulary.
Work usually starts by making the current behaviour visible: what people search for, what returns nothing, and where they abandon. Only then is it worth touching the index.
Built on Oracle Endeca and Elasticsearch, fed from PIM through loaders I have written more than once.
Where this has been done
Sr. Developer
Endeca-backed search and refinement across the storefront, fed by PIM data through a custom loader.
Sr. ATG Developer
Search refinement on a B2C storefront, built on ATG Repository Search.