Fanatics has rolled out an OpenAI-powered Shop the Look across its network to help shoppers assemble coordinated game-day outfits. The feature surfaces on Fanatics.com and the official online stores for the NFL, NBA, MLB and NHL, offering product-page suggestions that match a browsed jersey or other team item with in-stock merchandise that shares the same team and style.
The new Shop the Look recommendations are generated by OpenAI models that draw on enriched product metadata, team information and live inventory. Fanatics layers merchandising rules on top of those AI outputs to ensure suggested items remain compatible and relevant—so recommendations reflect not just stylistic fit but also availability. Catalogs and the suggested looks are refreshed daily so selections align with current stock and offerings.
From a technical standpoint, Fanatics combined OpenAI capabilities with its internal engineering to build the system. The company uses ChatGPT Work and Codex across multiple business units, including Fanatics Corporate, Commerce, Collectibles, and Betting and Gaming. Codex played a specific role in accelerating the creation of custom tooling that enriches catalog data, identifies compatible merchandise, evaluates recommendation quality and assembles full looks for each product.
The development process was intentionally rapid. Fanatics reports that generating and testing initial looks for 50 products required less than a day. Teams reviewed those early outputs and provided feedback, and after a few days of iteration the company approved a minimum viable product within the same week. OpenAI assisted in designing the solution so Fanatics could move the experience into production and scale it across a much larger set of items.
Fanatics validated the approach with an early trial across roughly 30,000 MLB products. In those tests, the OpenAI-powered Shop the Look experience produced a 10% higher engagement rate compared with the previous recommendation system. The lift suggests that combining generative models with curated merchandising logic and real-time inventory signals can increase shopper interest in complementary items and help fans more easily complete team outfits.
Operationally, Fanatics emphasizes that recommended looks are constrained by merchandising rules to preserve compatibility and relevance. The company also refreshes catalog signals daily so recommendations reflect what is actually available to buy. These measures aim to prevent recommending out-of-stock or irrelevant items while letting the AI surface combinations that customers may not discover on their own.
The integration illustrates a broader pattern in e-commerce: using generative AI to enhance on-page personalization while retaining human-defined controls for commercial and brand considerations. Fanatics’ approach couples automated recommendations with guardrails drawn from merchandising expertise, enabling the retailer to offer tailored suggestions at scale without ceding full editorial control to the model outputs.
Fanatics’ use of Codex and ChatGPT Work across multiple teams signals how the company is adopting AI as a cross-functional toolset. Codex helped speed development of the backend tooling that standardizes product information and evaluates recommendation quality, while ChatGPT Work supported collaboration across business units and solution design. Those internal workflows contributed to the quick turnaround from prototype to production-ready service.
For shoppers, the immediate benefit is simplified discovery: when browsing a team jersey, the Shop the Look feature proactively suggests coordinated items that match team colors and style, and that are in stock. For Fanatics, the test results point to measurable commercial upside—higher engagement can translate into more cross-sell opportunities and larger baskets when customers add complementary pieces to complete an outfit.
As retailers continue to experiment with AI-driven personalization, Fanatics’ rollout offers a case study in blending model-driven creativity with inventory-aware merchandising. The company’s controlled testing and rapid iteration underscore a pragmatic path to bringing generative AI into customer-facing commerce experiences while keeping recommendations grounded in real-world availability.
In summary, Fanatics’ OpenAI-powered Shop the Look uses enriched metadata, live inventory and merchandising rules to recommend coordinated team outfits across its network. Early MLB tests showed a 10% engagement lift, and the system was built rapidly using Codex and ChatGPT Work to enrich catalogs, evaluate recommendations and scale the experience across official league storefronts. The deployment highlights how AI-assisted tools can be integrated into e-commerce product pages to deliver tailored recommendations at scale.
Source: Read the original source
Leave a Reply