Department stores have been fighting for relevance for two decades. Their answer, so far in 2026, is a chatbot.
Macy's launched Ask Macy's publicly in late March — a Google Gemini-powered AI shopping assistant that helps customers discover products, receive personalized style recommendations, and try on apparel virtually. According to Bloomberg, shoppers who interact with the tool spend roughly 4.75 times more than those who don't. Fortune puts that at "nearly 400% more." Either way, it's a number the retail industry is taking seriously.
The tool wasn't built overnight. Retail Dive reports that Macy's ran a dark launch in December 2025, with employees testing the assistant internally before a wider pilot covering roughly half of website visitors over several weeks. The public rollout on March 23 marked the first time all Macy's digital platform users could access it — including the app, where the virtual try-on feature has seen particularly strong adoption.
What It Does
Ask Macy's is less a search tool and more a digital stylist. CX Dive describes the core interaction loop: a shopper describes what they're looking for — a dress for a wedding, a gift for a dad who hates shopping — and the assistant responds with brand recommendations, "complete the look" outfit pairings, and direct links to purchase. The virtual try-on feature allows users to see how selected items would look on a model with their approximate body type and coloring.
The emphasis on "complete the look" is deliberate. Macy's long-term revenue problem isn't just about getting shoppers to the site — it's about getting them to buy more than one item per visit. The AI assistant's basket-building function directly addresses the chain's average order value challenge, which has been under pressure as more consumers treat department store websites as comparison tools rather than buying destinations.
The 400% Number: What It Actually Means
Before the retail industry collectively rewires its digital strategy around this stat, it's worth asking the obvious question: is this correlation or causation?
The honest answer is probably both. PYMNTS notes that Macy's has not released a controlled experiment showing that the AI assistant caused higher spending — the 4.75x figure compares users of the tool against non-users over the same period. There's meaningful selection bias baked in: a shopper willing to engage in a multi-turn AI conversation to find an outfit is probably already more purchase-motivated than someone bouncing from page to page.
That caveat matters, but it doesn't eliminate the signal. Even if half the differential is selection bias, a 2x spending lift from AI-assisted users is a result most retailers would take immediately. The more interesting question is whether usage of the tool can be scaled — whether Macy's can train enough customers to reach for Ask Macy's first, rather than treating it as a novelty.
Why This Matters Beyond Macy's
Chain Store Age noted that Macy's is the latest in a growing list of major retailers deploying conversational AI at scale. As Endcap has covered, Gap launched a Google Gemini-integrated checkout, Sephora embedded ChatGPT, and Walmart's Sparky now handles both shopping assistance and supplier management.
The Macy's deployment is notable because it's happening at a company whose strategic position is so precarious. The retailer is mid-way through closing dozens of stores, under pressure from off-price competitors, and trying to navigate a consumer who is simultaneously trading down and seeking more personalized experiences. If any company needed AI to pull its weight, it's Macy's.
The test isn't whether Ask Macy's works in a pilot. It's whether it can move the needle on the metric that matters most for department stores in 2026: getting customers to choose Macy's over Amazon when they already know what they want — and over no one when they don't know yet but might be convinced.
That's a hard job. A well-built AI shopping assistant might actually be equipped to do it.