Industry

OpenAI and Fanatics Deploy AI-Powered Shopping Recommendations Across Major Sports Leagues

Fanatics launches Shop the Look, an OpenAI-powered recommendation engine that suggests coordinated merchandise to fans browsing team apparel across NFL, NBA, MLB, and NHL official stores.

Last verified:

Fanatics, a global digital sports commerce platform, deployed an OpenAI-powered recommendation engine to match fans with coordinated game-day merchandise. According to OpenAI Blog, the new Shop the Look feature uses OpenAI’s models to suggest in-stock apparel and accessories that pair with items fans are browsing across official team stores. The system was built and tested at scale within days, rolling out to thousands of products and delivering measurable engagement gains over the incumbent recommendation approach.

Shop the Look Deployment Across Major Sports Retailers

According to OpenAI Blog, Shop the Look is now live on Fanatics.com and the official online stores for the National Football League (NFL), National Basketball Association (NBA), Major League Baseball (MLB), and National Hockey League (NHL). When a fan browses for a team jersey, the system automatically surfaces coordinated outfits on the product page—suggesting compatible hats, jackets, and other gear in matching team colors and styles.

The recommendations rely on enriched product data, team information, and real-time inventory status to ensure suggestions are both relevant and purchasable. Fanatics applies its own merchandising rules to validate that paired items align with brand guidelines and style conventions. The engine refreshes daily as inventory and catalogs change, keeping recommendations current across thousands of SKUs.

Rapid Development Using Codex and ChatGPT Work

Fanatics accelerated development by deploying OpenAI’s Codex (code generation) and ChatGPT Work across corporate, commerce, collectibles, and sports betting operations. Using Codex, the Fanatics engineering team and OpenAI built custom tools to enrich product metadata, identify compatible merchandise, evaluate recommendation quality, and generate looks for each product. The initial batch of 50 test looks was completed and evaluated in less than 24 hours, with feedback loops enabling iteration. Within a week, Fanatics approved a minimum viable product ready for production.

After validating the approach across roughly 30,000 MLB products, the OpenAI-powered Shop the Look experience achieved 10% higher engagement than the prior recommendation system, according to OpenAI Blog. This early success demonstrates the business case for scaling AI-assisted merchandising across the Fanatics network.

Why This Matters

This deployment exemplifies a high-velocity enterprise use case for large language models: automating data enrichment and recommendation logic at scale. Fanatics compressed what typically takes weeks of feature development into days by leveraging OpenAI’s pre-trained models and code generation capabilities. For sports retailers competing on fan experience, the 10% engagement lift translates to higher average order value and merchandise attachment rates—metrics that directly affect margins. As LLM-powered recommendation systems mature and integrate with real-time inventory systems, retailers will likely prioritize vendors who can combine generative AI with merchandising domain knowledge, positioning OpenAI and similar platforms as infrastructure rather than pure software-as-a-service offerings. The success on MLB inventory suggests Fanatics will expand this pattern across other sports properties and product categories.

Frequently Asked Questions

What does Shop the Look recommend?

The system suggests in-stock merchandise that coordinates with items fans are browsing, such as complementary apparel and accessories that match their team and style preferences.

Where is Shop the Look live?

According to OpenAI Blog, the feature is deployed across Fanatics.com and the official online stores for the NFL, NBA, MLB, and NHL.

What OpenAI tools did Fanatics use?

Fanatics deployed ChatGPT Work for enterprise use and Codex (OpenAI's code generation model) across corporate, commerce, collectibles, and sports betting functions.

How much did engagement improve?

Testing across roughly 30,000 MLB products showed Shop the Look achieved 10% higher engagement than the previous recommendation system.

#generative-ai #e-commerce #sports #openai #enterprise-ai