Macy's launched its "Ask Macy's" AI shopping assistant this week across all digital platforms, and early results are turning heads. In A/B testing with roughly half of Macy's website visitors, shoppers who engaged with the chatbot spent approximately 4.75 times more than those who browsed without it.

The chatbot is powered by Google's Gemini and was built with feedback from thousands of Macy's employees before its wider rollout.

What It Does

The two most popular features are "complete the look" โ€” where the bot suggests accessories to pair with an outfit โ€” and a virtual try-on that lets shoppers preview items on themselves. The try-on feature is also available in physical stores for time-pressed customers.

Chief Customer and Digital Officer Max Magni attributed the spending jump partly to intent: users of the bot tend to be searching for something specific, like an outfit for an event, rather than casually browsing. He also believes the tool is drawing a younger customer base.

From Cold to Conversational

Early versions of the bot had some rough edges. When asked for T-shirt suggestions for a 10-year-old, it replied flatly: "Here's a T-shirt for a 10-year-old." After iteration, the same prompt now generates: "Ten-year-olds can have so much fun with color โ€” do you want a brighter or more muted color selection?"

Context

The launch comes as Macy's works through a multi-year turnaround. Net sales fell 2.4% last year, though comparable sales ticked up 1.5%. The company is projecting $21.4โ€“$21.65 billion in net sales for 2026.

AI-powered shopping assistants are becoming a competitive front across retail โ€” from startups like Wizard (Marc Lore) to browser tools like Phia. Macy's is among the first major department stores to publish real spending data backing the bet.