What Agentic Commerce Really Means in a Beauty Aisle
Agentic commerce describes AI systems that do more than answer questions—they can independently orchestrate a shopping task from discovery through purchase. In Ulta’s case, that means embedding its catalog into Google’s AI Mode in Search and the Google Gemini app so an AI shopping assistant can recommend products, compare options and even complete a streamlined checkout inside the conversation. Instead of users hopping between search, product pages and carts, the assistant pulls Ulta’s product data, applies business rules via the Universal Commerce Protocol, and then acts on the shopper’s intent: surfacing beauty product recommendations, refining choices, and guiding payment-eligible items to purchase. Practically, agentic commerce turns a static product listing into a responsive, goal-driven agent that can interpret needs—like “a hydrating foundation for dry skin”—and coordinate everything from shortlists to final confirmation, all within a single conversational shopping flow.
Ulta AI: Turning Loyalty Data into Conversational Shopping
Ulta is extending this agentic model into its own ecosystem with Ulta AI, an AI shopping assistant built on Gemini Enterprise for Customer Experience. Embedded across ulta.com and the Ulta Beauty app, the assistant taps into insights from more than 46 million loyalty members to deliver retail AI personalization at scale. As guests browse Ulta’s expansive digital assortment, the AI shopping assistant can interpret everyday language—“lightweight serum that won’t irritate sensitive skin”—and translate it into targeted beauty product recommendations. Over time, it learns patterns in routine, preferences and responses to past suggestions, making conversational shopping feel increasingly like a one-to-one consultation rather than a generic search. Importantly, Ulta frames Ulta AI as a complement to in-store associates, not a replacement, using it to extend human expertise into digital channels where guests still expect guidance, reassurance and curated choice.
From Browsing to Buying: How Agentic AI Rewrites Product Discovery
Ulta’s move aligns with a broader shift toward integrated AI shopping, where discovery and purchase are embedded directly inside chat-based environments. Similar to how David’s Bridal plugs its catalog into conversational platforms and restructures data around detailed attributes like silhouette and fabric, Ulta’s agentic commerce approach leans on attribute-rich catalogs and Universal Commerce Protocol to enable precise matching and smoother paths to purchase. In this model, conversational shopping becomes a front door to the entire assortment. Instead of wading through filters, shoppers describe their needs—skin type, finish preferences, routine complexity—and the AI assembles shortlists, compares options and builds smarter bundles. Embedded checkout inside Google Gemini compresses the funnel further, capturing intent at the moment it’s articulated. For retailers, this AI-driven discovery ecosystem changes where merchandising happens: inside the assistant’s dialogue, not just on category pages or search grids.
Benefits for Shoppers: Faster Matching, Smarter Bundles, More Confidence
For beauty shoppers, agentic commerce promises less guesswork and more tailored guidance. An AI shopping assistant can quickly narrow thousands of SKUs into a handful of beauty product recommendations matched to skin tone, concerns and budget, then build coordinated routines—cleanser, serum, moisturizer, SPF—without forcing users to understand every ingredient. Attribute-based modeling, like that used by David’s Bridal for silhouettes and fabrics, translates in beauty to texture, coverage, finish and skin type, allowing the assistant to propose combinations that actually work together. Because Ulta AI learns from loyalty behaviors and feedback, it can refine suggestions over time, helping guests avoid redundant purchases and discover new brands aligned with their preferences. The result is a more confident purchase decision: shoppers can ask follow-up questions, compare products in plain language and see curated sets that reflect both their past habits and emerging needs, all within a single conversation.
Retailer Upside and the Trust Question in AI-Driven Commerce
For Ulta and other retailers, agentic commerce unlocks richer first-party data and better conversion. Conversational logs reveal how people describe their needs, which attributes matter most and where they hesitate, feeding back into merchandising, promotions and product data enrichment. Integrated AI shopping within platforms like Google Gemini also reaches customers earlier in their journey, opening new avenues for targeted offers and cross-brand discovery. Yet this power raises critical questions around trust and transparency. If an AI shopping assistant curates most options, retailers must clearly signal why items are recommended—whether based on suitability, popularity or commercial priorities. Over-personalization can also risk narrowing exposure, so systems need guardrails that still allow exploration and serendipity. The next generation of retail AI personalization will hinge not only on technical sophistication, but on how openly retailers communicate data use, recommendation logic and the ongoing role of human expertise.