Traffic from artificial intelligence tools to US retail sites now generates 83% higher revenue per visit than non-AI traffic, according to Adobe's analysis of more than 1 trillion visits to US retail websites.
AI-referred visitors also converted 90% better than shoppers arriving through other channels in September, the strongest performance Adobe has recorded for the category so far. Visitors from AI sources spent 73% more time on retail sites, viewed 25% more pages per visit and added products to baskets at a 53% higher rate than non-AI users.
The data points to a change in how online shoppers research products before buying. Consumers are increasingly using large language model tools to narrow choices, compare options and identify deals before clicking through to a retailer's website, Adobe said.
That appears to be changing the value of a site visit. Adobe said AI traffic has gone from being worth less than half of a typical visit a year earlier to nearly twice as much in revenue terms.
AI traffic to US retail sites grew 71% year on year in September and has increased every month since October 2024, when Adobe began tracking the segment.
Holiday demand
Adobe expects another jump over the main US holiday shopping period. It forecasts AI-driven retail traffic to rise 159% year on year on Thanksgiving, 95% on Black Friday and 88% on Cyber Monday.
Retailers are preparing for one of the busiest periods in the consumer calendar as they compete for online spending from price-conscious shoppers. Adobe expects the revenue-per-visit premium from AI-referred traffic to widen further during the season.
The analysis covers online transactions across 100 million stock-keeping units and 18 product categories. Adobe also drew on a survey of more than 5,000 US respondents to add context on how consumers use AI in daily life.
Visibility gap
The research also identified a problem for retailers trying to benefit from the shift. The average US retail homepage scored 63% on Adobe's Citation Readability Score, a measure designed to assess how much webpage content can be understood and cited by AI systems.
That means more than a third of homepage content is not easily visible to the AI tools directing shoppers to retail sites, Adobe said. The issue was more pronounced in some parts of the market, where layouts focused heavily on navigation or promotions gave AI systems less context to interpret.
Electronics recorded the highest overall score at 79%, helped by category pages and customer service pages with setup guides, troubleshooting information and detailed specifications. Grocery followed at 55%, while general merchandise homepages scored 40%.
Explanatory content such as product information, buying guides, frequently asked questions and return policies was more likely to be picked up by AI systems than heavily promotional homepage material, Adobe said.
The findings suggest retailers may need to rethink how information is presented on their sites if they want to improve their chances of appearing in AI-generated answers or recommendations. That matters because, by the time users arrive from an AI tool, they may already be closer to making a purchase than someone browsing through search, email or social media links.
Adobe said the trend marks a reversal from the earlier stages of AI referral traffic. In August 2025, AI traffic converted 9% worse than non-AI traffic, but by September that had shifted decisively in the other direction.
The figures add to growing evidence that AI tools are becoming an important gateway for online commerce, not only as a source of website visits but also as a route to shoppers showing stronger intent to buy. For retailers, the data indicates that visibility within AI systems may increasingly influence digital sales outcomes as much as traditional online marketing channels.