Codex
Fanatics used Codex to build "Shop the Look" across a 1.5M-item catalog
Fanatics, whose team-merchandise business is its biggest revenue line, built a "Shop the Look" outfit recommender with OpenAI Codex. The company reports a 10% rise in views of recommended items.
What changed
Fanatics Commerce runs hundreds of league and team stores with more than 1.5 million SKUs and roughly a thousand new items a week. Photo shoots and hand-styled outfits could not keep up, and shoppers were spending 30+ minutes digging through the catalog. Working with OpenAI, the team used Codex to build a tool that reads product and catalog data, pairs a jersey with in-stock hats, socks, and accessories from the same team, and re-checks those pairings as stock changes. The feature now appears on product pages.
Results
- Fanatics' chief product officer Justin Tsai told CNBC that views of suggested items rose 10% after launch.
- Posts citing OpenAI's own write-up say it was tested on about 30,000 MLB products and that Codex helped test the first 50 looks in under 24 hours.
- The metric is engagement, not revenue. Fanatics has not published a sales lift tied to the feature.
Who is affected
E-commerce and marketplace teams with large or fast-changing catalogs, and developers using Codex to build recommendation or merchandising tools.
What to do now
The hard part was not the model, it was the matching rules: team and player alignment ("no Giants hat for a Dodgers fan") and in-stock filtering. If you are building catalog features with Codex, encode those hard constraints in the data and the checks first, then let Codex generate and validate candidates in bulk.
What is not confirmed
Reported by CNBC (Oct 26, 2026) with on-record quotes from Fanatics and OpenAI. The 10% figure is the company's own number and is not independently verified. The 30,000-product and 24-hour details come from posts summarizing OpenAI's page and were not checked against that page directly.
Sources
- CNBC: https://www.cnbc.com/2026/10/26/fanatics-openai-codex-sports-merchandise-shopping.html
- X post sharing the CNBC report (@BlueCrewViking): https://x.com/BlueCrewViking/status/2114705472527737040
- X post summarizing OpenAI's page (@Chahatusharma): https://x.com/Chahatusharma/status/2114729260329861257
Sources
- ReportCNBC: Fanatics, OpenAI Codex, sports merchandise shopping · Oct 26, 2026
Get updates like this every morning
- ① Email
- ② Card on Stripe
- ③ 7 days free
Then $2/month · cancel anytime in one click