MW These three clothing retailers might be ahead of their rivals in the AI race
By Bill Peters
UBS analysts suspect recent productivity gains in some retail chains might be due to AI, and say the technology could have a bigger impact on the industry as IT budgets evolve
Abercrombie & Fitch is among the retail chains that may be further along in adopting AI.
As the retail industry figures out how to put artificial intelligence to use, clothing chains like Abercrombie & Fitch, Gap and TJX could be further along than their rivals in adopting the technology, UBS analysts said Monday.
The analysts said that while calculating the exact impact of AI on those retailers was difficult, they based their assessment on improvements in sales per employee - which they used as a gauge of productivity - and on improved profitability as measured by earnings before interest and taxes.
The number-crunching exercise came as rising gas prices due to the Iran war, and the potential impact on consumer demand, take investors' focus off AI adoption at retail chains. AI adoption has, at times, also shown signs of running up against roadblocks, as OpenAI reportedly rethinks the way it works with retailers following hiccups in bot-assisted checkout features.
The UBS analysts, led by Jay Sole, compiled employee-count data at the "softlines" retailers that they cover - those that sell things like clothing and linens. Their analysis indicated that the average sales per employee across those companies rose to around $272,000 in 2021-25, from around $222,000 in 2017-19.
Those employee productivity gains, the analysts said, likely came from the increase in digital-shopping demand during the pandemic. The analysts' calculations were also a function of the store closures that followed.
The analysts said their research also showed that e-commerce adoption was leveling off. Thus, they said, the increase in sales per employee, at least since 2024, required another explanation.
"We suspect the reason is softlines companies' increasing use of AI," they wrote.
Based on that analysis, they said Abercrombie $(ANF)$, Gap $(GAP)$ and TJX $(TJX)$, which owns TJ Maxx and Marshalls, "are likely ahead of peers with respect to AI adoption."
Some retail chains have been upbeat about the spring shopping season or have downplayed the threat of higher gas prices. But consumers have remained cautious, and retailers over this decade have already been dealing with tariffs and spikes in energy and shipping costs.
Against that backdrop, the UBS analysts said, investors have been skeptical that AI can help boost clothing retailers' financials. But they believe that it's still early days for the technology, and that the impact from AI will increase as the technology improves, IT budgets evolve and retailers figure out what practices work best.
"Investors are interested to know if and to what degree companies are using AI to benefit sales and margins," the analysts said. "However, AI isn't an 'application' which can alone boost companies' financial performance. AI is a tool which companies are integrating into hundreds of business processes across their organizations."
Through last year, companies such as Walmart $(WMT)$, Etsy $(ETSY)$ and Shopify (SHOP) signed on with OpenAI to allow shoppers to make purchases through ChatGPT and Instant Checkout, a shopping tool also launched by the AI startup.
But last month, according to The Information and CNBC, OpenAI was backing away from the Instant Checkout feature, which allowed direct purchases through ChatGPT, following challenges in getting those purchases to work consistently. Instead, the AI startup was working on developing apps in ChatGPT that would steer consumers to retailers' websites to complete purchases, CNBC said.
Since then, OpenAI has tried to find more ways to help shoppers find what they want and to compare items, offering visual browsing options and recommendations based on parameters like shoppers' budgets.
Mizuho analyst David Bellinger, in a note last month, said those changes were an improvement for ChatGPT, but that shopping on the platform still needed refining.
"This is by no means perfect and another iteration with many, many more to follow," he said. "Our early trials of ChatGPT's update are still rather clunky and not ideal. Consumers aren't yet seeing a materially different path to purchase."
-Bill Peters
This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.
(END) Dow Jones Newswires
April 06, 2026 13:21 ET (17:21 GMT)
Copyright (c) 2026 Dow Jones & Company, Inc.
Comments