Alhena tested 15 live AI shopping agents on foundation shade matching. The best scored 2.75/3.0. Five ignored the selfie entirely and returned bestsellers.
Alhena tested 15 live AI agents on a fragrance-free skincare brief. Five forgot the constraint mid-conversation. In skincare, the forgotten constraint is not a preference – it is a reaction.
Alhena tested 15 live AI shopping agents across 11 verticals in 2026. Category difficulty did not predict failure. Architecture did, and the widest gap sat inside one category.
Three technologies wear the same chat box. Alhena's 2026 study found each has a hard ceiling: personalisation stops at Recommend, AI search at Sell, agentic reaches Act and Remember.
Alhena tested 15 live AI shopping agents in 2026. Only 1 remembered the shopper on a return visit. Memory is the rarest capability in agentic CX, and the hardest to fake.
Alhena tested 15 live AI shopping agents in 2026. All 15 explained how to process a return. Only 4 actually did it. Inside the widest performance gap in agentic CX.