Michaels' AI Shopping Assistant Doubles Conversion vs. Search

Michaels' AI assistant outperforms traditional search

Michaels has released performance metrics for its Google Gemini-powered shopping assistant that show conversational AI delivering measurably higher conversion rates than keyword search. The craft retailer's "Ask Mike" tool converts shoppers at more than double the rate of customers using traditional search functions, according to Heather Bennett, president and chief customer officer at Michaels.

The assistant, which formally launched on July 21 after a quiet May rollout, has handled 75,000 conversations across the retailer's website and mobile apps. Twenty-seven percent of those interactions result in a product click or add-to-cart action.

Ask Mike responds to open-ended prompts like "Help me plan my kid's birthday party" or "I need fabric for DIY curtains," then asks follow-up questions about project type, colors, materials and budget. This conversational approach differs fundamentally from keyword-based search, where customers must break projects into isolated terms and manually filter results.

Why it matters

The conversion rate advantage suggests conversational AI can solve a persistent e-commerce problem: helping customers who know what they want to accomplish but not which specific products they need. For project-based retailers like Michaels, this translates to larger basket sizes as shoppers bundle complete supply sets rather than purchasing single items. The quick deployment timeline—six weeks from concept to production—also demonstrates that retailers with modernized infrastructure can implement enterprise AI without multi-year transformation projects.

Rapid deployment on existing infrastructure

Michaels built Ask Mike using Google Cloud's Gemini Enterprise for Customer Experience, the same software package powering AI shopping tools at Ulta Beauty, Macy's and The Home Depot. Paul Tepfenhart, global director of retail industry strategy and solutions for Google Cloud, noted the retailer moved from concept to production-grade AI in six weeks.

Kapil Dabi, Americas lead of retail and CPG solutions for Google Cloud, attributed the speed to Michaels' existing cloud foundation and AI-ready product catalog structure. He emphasized that technical infrastructure alone isn't sufficient—retailers also need clear understanding of customer behavior patterns in their specific category.

Behavioral differences between search methods

Bennett reported distinct shopping patterns between the two interfaces. Ask Mike users bundle core supplies and finishing touches in single sessions for complete projects like party planning or framing. Traditional search users more often look for predetermined single items.

"Unlike traditional search, Ask Mike acts as a creative assistant that helps shoppers move seamlessly from inspiration to product discovery and cart action, driving much higher overall engagement," Bennett said.

Dabi positioned the results as evidence that retail is entering an era where conversational AI becomes the primary interface between shoppers and brands, enabling faster discovery and more personalized search experiences.

These details were first reported by Modern Retail.