Outfit assembly is categorically different from product recommendation. A recommendation widget surfaces products that match a query. Outfit assembly builds a complete look — selecting pieces that work together aesthetically, for a specific occasion, in a specific shopper's size and taste profile. The output is a decision, not a list. The assembly process requires the AI to reason across multiple dimensions simultaneously: occasion appropriateness (what does a "Diwali party" imply about formality?), aesthetic coherence (which silhouettes and colors work together?), catalog availability (which pieces exist in the shopper's size?), and taste fit (does this match what the shopper has historically liked?). Each of these dimensions filters the candidate set until a complete look emerges. Outfit assembly is why Elara produces higher AOV than recommendation tools. A shopper who receives a complete outfit recommendation and buys multiple pieces from it generates more revenue per session than a shopper who evaluates individual product suggestions. The shopper's decision is also more informed — they saw how the pieces work together — which reduces returns.