AI assistants are sending more shoppers to retail websites as Google, OpenAI, and retailers develop shopping tools that connect product discovery with merchant product data and checkout systems.

Adobe Digital Insights found that AI-driven traffic to US retail websites increased 138% year over year in May 2026. AI-referred shoppers generated 53% more revenue per visit and converted at a 54% higher rate than traffic from non-AI sources during the month.

Reuters reported recently that retailers including Walmart, Ulta Beauty, and Wayfair are adjusting their online operations as shoppers use services such as ChatGPT and Google’s Gemini to research products. Many want purchases to remain within their own commerce systems, where they can retain information on browsing habits, basket sizes, and previous purchases.

That information supports functions including product recommendations and loyalty programmes. Retailers interviewed by Reuters said maintaining the transaction on their own platforms also preserves the direct relationship with the customer.

Adobe’s data also shows a change in the quality of AI-referred traffic. In March 2026, visitors arriving from AI services converted 42% better than non-AI traffic, compared with March 2025, when AI-referred traffic converted 38% worse.

A March survey of more than 5,000 US consumers conducted by Adobe found that 39% had used AI assistants for online shopping. Of those respondents, 85% said AI assistants had improved their online shopping experience.

AI shopping moves closer to checkout

Google introduced Universal Cart in May, allowing shoppers to add products while using services including Search and Gemini. Consumers can complete supported purchases using Google Pay or transfer products to a retailer’s website before checking out.

Google said retailers remain the merchant of record regardless of the checkout route. Universal Cart can also use payment-method benefits, loyalty information, and merchant offers stored in Google Wallet.

Nike, Sephora, Target, Ulta Beauty, Walmart, and Wayfair are among the retailers Google said would support selected checkout features through its Universal Commerce Protocol, or UCP. The company said Universal Cart was rolling out across Search and Gemini in the US during the summer, with YouTube and Gmail to follow.

Wayfair was one of the retailers that helped develop UCP. The retailer said its integration enables eligible customers to check out from Wayfair without leaving Google’s interface, while Wayfair remains responsible for pricing, fulfilment, and customer support.

Ulta Beauty is also connecting its commerce systems with Google’s AI services. The retailer announced in April that its assortment would become shoppable through AI Mode in Search and the Gemini app, while Ulta AI launched on Ulta.com.

Ulta said its assistant draws on insights from more than 46 million members to provide personalised shopping guidance. Reuters also reported that the retailer is working with Google to incorporate shopping carts and its Ulta Beauty Rewards programme into shopping experiences within Gemini.

“We always want to be wherever our guest is,” Josh Friedman, Ulta Beauty’s head of digital and e-commerce, said. Ulta is seeing twice the conversion and purchase intent from shoppers finding its products through Gemini and ChatGPT, according to Reuters.

Walmart is also participating in external AI shopping services while developing experiences on its own digital properties. Walmart and Google said their Gemini integration will allow customers to link their accounts, receive recommendations based on previous online and in-store purchases, and combine new items with existing Walmart or Sam’s Club carts.

Product data for AI discovery

Retailers are also addressing how their product information is accessed by AI systems. Adobe analysed US retail sites using its AI Content Visibility Checker, which measures the proportion of webpage content the tool determines is readable by AI systems.

Retail homepages recorded an average visibility score of 75%, while category pages averaged 74% and individual product pages 66%. Customer-service pages scored higher at 79%.

Adobe’s later analysis found that cosmetics led the retail categories it assessed for AI citation readability, which Adobe attributed partly to editorial, educational, and customer-service content.

Structured product feeds provide another route for retailers to supply information directly to AI shopping services. OpenAI’s Agentic Commerce Protocol allows merchants to provide product data for use in ChatGPT’s search and shopping experiences.

OpenAI says merchants can supply feeds containing information such as descriptions, prices, inventory, images, and fulfilment options. Its documentation calls for regularly refreshed feeds so ChatGPT can display current product information, including price and availability.

OpenAI also developed specifications for checkout integrations in which ChatGPT collects buyer, delivery, and payment information while the merchant’s systems handle the transaction. The merchant validates the order, calculates tax, processes the payment through its existing provider, and decides whether to accept the order.

The company changed its broader shopping approach in March after concluding that the initial version of Instant Checkout did not provide enough flexibility. OpenAI said merchants would be able to use their own checkout experiences while it focused on product discovery, although deeper native integrations remain available through ChatGPT apps.

OpenAI says it is not the merchant of record under the Agentic Commerce Protocol.

Measuring AI visibility

The expansion of AI shopping is also producing new measurement tools for merchants. Google’s AI performance insights report in Merchant Center measures how brands appear across AI Mode and AI Overviews.

The report, currently in pilot with a limited number of US Merchant Center accounts, includes share-of-voice data that compares a merchant’s visibility with competitors. Google separates conversational shopping queries into discovery, evaluation, and purchase stages.

Merchant Center can also surface common query types and identify product attributes, such as size, colour, or material, that appear in AI shopping searches but are missing from a retailer’s structured product data. The current report covers organic AI traffic and does not include paid advertising traffic.

(Photo by Eduardo Soares)

See also: Michaels says AI shopping assistant more than doubles conversions

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.

Marketing Tech News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.