How Fanatics and OpenAI streamline the fan experience

Fanatics and OpenAI are collaborating to improve the fan experience. As part of that work, Fanatics is launching an OpenAI-powered version of Shop the Look, which uses AI models to recommend coordinated merchandise to help fans complete their game-day outfits.

Fanatics, a global digital sports platform, uses ChatGPT Work and Codex across Fanatics Corporate, Commerce, Collectibles, and Betting and Gaming.

Helping fans complete the look

Shop the Look is now live across many sites across the Fanatics network—including Fanatics.com and the official NFL, NBA, MLB and NHL online stores. With it, fans can more easily find recommended merchandise to match with items they’re looking to wear. When fans browse for their favorite team’s jersey, Shop the Look suggests coordinated outfits on the product page, using in-stock items that match their team and intended style.

OpenAI models draw on enriched product data, team and inventory information to assemble the recommendations. Fanatics’ merchandising rules keep the selections compatible and relevant, and looks are refreshed daily as catalogs change.

Building and testing the first 50 looks in 24 hours

Using Codex, OpenAI and the Fanatics engineering team quickly developed custom recommendation tools to enrich product and catalog data, find compatible merchandise, evaluate recommendations, and generate looks for each product.

Testing the new approach on 50 products took less than a day to complete. Fanatics reviewed the looks and gave feedback directly to the builders. After a couple days of iteration, the company approved the minimum viable product that same week. OpenAI helped design the solution which enabled Fanatics to implement the new Shop the Look experience in their production environment and scale the approach across thousands of products.

After testing the recommendations across roughly 30,000 MLB products, the OpenAI-powered Shop the Look experience achieved 10% higher engagement than the previous experience.