Fanatics launches OpenAI Codex-powered Shop the Look outfit suggestions

The feature, announced by Fanatics Commerce, matches items on a product page with other in-stock products based on team and intended style, according to CNBC. Fanatics Commerce manufactures more than 45 million items a year and sells to customers in more than 180 countries.

Justin Tsai, chief product officer for Fanatics Commerce, said the company kept seeing customers struggle to work through its large catalog across hundreds of sites. Shopping sessions routinely averaged more than 30 minutes, he said, and that time kept compounding as Fanatics released upwards of 1,000 new products each week.

Traditional curation methods such as photo shoots and styled outfits have lead times that do not always fit Fanatics’ growing product pool or fast-released items like championship shirts and hats, Tsai said. Random suggestions also do not work for sports fans, he said. “Nobody who is a Los Angeles Dodgers fan wants to be matched with a San Francisco Giants item … team and player alignment is extremely important in our world,” Tsai said.

Fanatics founder and CEO Michael Rubin told CNBC’s “Squawk Box” earlier this month that the company’s commerce business will exceed $7 billion in revenue this year, compared with roughly $5.5 billion for its collectibles business, which includes Topps trading cards. Team merchandise remains Fanatics’ biggest revenue source even as it has grown into a multibillion-dollar business.

OpenAI has been experimenting with shopping tools for ChatGPT since the chatbot launched, including a feature last fall that let users buy merchants’ products directly within ChatGPT. But it is ending the Instant Checkout effort and now is working directly with retailers to create dedicated apps within ChatGPT, according to CNBC.

Rosemary DeArago, head of industry GTM at OpenAI, said personalization in retail “is an age-old problem.” E-commerce gave retailers new ways to “figure out the fingerprint of the customer” through past purchase history, location, or a logged-in account, she said, but those efforts still sometimes miss the mark. “In the past we were trying to stitch together something that would be good to show a customer,” DeArago said. “With AI, you can do more deterministic matching to do personalization in this next-level way.”

“Shop the Look” uses OpenAI’s Codex to create a tool that parses product and catalog data to find compatible items and reevaluate recommendations, producing looks that might suggest a hat, socks and other accessories to go with a jersey a fan is considering. Tsai said that after rolling out the feature, views of recommended items rose 10%. He also said the suggestions help shoppers have “more conviction” about items they are thinking of buying by showing how they may fit with other team gear.

DeArago said there will always be room in shopping for more curated looks and trying items on in person, but that AI integration in efforts that narrow massive product selections and show what might work with a valuable purchase like an authentic team jersey shows there is a “really big opportunity to enhance the existing experience.”