Google and Macy's Launch AI Shopping Agent in Four-Week Sprint
Google and Macy's deployed an AI agent called Ask Macy's across Macy's website and mobile app in late March 2026, scaling from a small beta group to all users within a week. The companies announced the project publicly on April 22, roughly five weeks after starting work.
The agent, built on Google's Gemini Enterprise for Customer Experience, handles text and image inputs to help customers find products across Macy's 2.5 million stock-keeping units. Early data showed revenue per visit was 4.75 times higher among users who engaged with Ask Macy's compared to those who did not.
How the project came together
Macy's had spent six months developing its own AI agent before learning about Google's Gemini Enterprise offering. The retailer abandoned that work and pivoted to Google's platform, setting a launch target of March 24 to coincide with the Shoptalk retail conference.
With less than six weeks to launch, teams from both companies held daily standups starting February 9. The compressed timeline forced Macy's to skip some standard processes. "The pace of change and innovation in AI left no room to stand still," said Chad Westfall, senior vice president of technology product development and customer experience at Macy's.
What Ask Macy's does
The agent answers product questions and provides recommendations. It can process customer photos and show how clothing would look on the person wearing it, with options to display different settings like an office or restaurant.
Macy's tuned the agent's responses to sound more conversational. Rather than returning "Here's a T-shirt for a 10-year-old," Ask Macy's might say: "10-year-olds can have so much fun with color - do you want a brighter or more muted color selection?"
The companies plan to add size recommendations based on product feedback in the near future, warning customers if items run large or small.
Development timeline and approach
The actual build took four weeks from decision to launch. The speed required both teams to streamline workflows and prioritize core functionality over polish. Macy's framed the project as bringing "the concept of hospitality to online customers at scale," mimicking the curated, personalized experience of in-store shopping.
The rapid deployment reflects how Generative AI and LLM platforms are enabling faster implementation cycles. For development teams, the Ask Macy's case shows how AI Agents & Automation can reduce friction in customer workflows while maintaining quality outputs through careful prompt engineering and feedback loops.
Your membership also unlocks:
