UX Case Study: Reorienting E‑commerce Search Around Intent Signals
Tech · 5 min read
RheoShop, a direct-to-consumer marketplace, noticed that searchers with the same query had different behaviors: some browsed for ideas, others had purchase intent. The original relevance algorithm treated all queries identically, yielding mixed results. The design and data teams decided to infer intent from session signals — prior page views, time-of-day, and query modifiers — and surface different UX for different intents.
For inferred buyers, the results UI prioritized best sellers, clear pricing, and one-click options, with a streamlined product card. For browsers, the UI highlighted editorial recommendations, lookbooks, and a save-for-later affordance. A small cohort of A/B test users saw a 15% increase in add-to-cart for inferred buyers and a 23% increase in saves for browsing users.
The project required close collaboration between search engineers and designers to create intent-aware templates and lightweight heuristics for switching experiences. The startup rolled out the approach gradually, with analytics dashboards that tracked intent accuracy and downstream conversion, enabling continuous tuning.