AI referral traffic has moved from a curiosity to a commercial channel. New Adobe research on U.S. retail sites found that in September, shoppers arriving through AI assistants generated 83% more revenue per visit than non-AI visitors and converted 90% better.
The reversal is stark. In August 2025, AI traffic converted 9% worse than other traffic. By September 2026, the same segment was outperforming, with visitors from AI sources spending 73% more time on site, viewing 25% more pages and adding items to cart at a 53% higher rate. Adobe also put year-over-year AI traffic growth to U.S. retail sites at 71% in September, and 1,772% since October 2024.
AI referrals arrive further down the funnel
Adobe’s data does not show exactly what a shopper did before clicking through. The reasonable inference is that many AI-referred visitors have already compared options, asked questions and narrowed their choices inside an assistant. They tend to arrive with more context than a typical referral visitor.
That changes the job of the landing page. Instead of always starting the journey, it often needs to finish it, by answering the final question, building confidence and converting an already-informed shopper.
Most retail content is only partly AI-readable
Adobe’s Citation Readability Score put the average U.S. retail homepage at 63%, meaning more than a third of the content was not readily understandable or citable by AI systems. This is a problem for brands hoping to be referenced, recommended or summarised accurately.
The scores varied widely by category. Electronics led at 79%, helped by category pages at 79% and customer service pages at 78%. General merchandise homepages scored just 40%, which Adobe attributed to navigation-heavy, promotional layouts that gave AI models less context.
Explanatory content such as setup guides, troubleshooting articles and detailed product specifications tended to score highest, because it gives AI systems more material to interpret.
Treat AI referrals like an acquisition channel
If AI visitors convert better and deliver more revenue per visit, raw referral volume is no longer enough. Marketing teams should evaluate what happens after the click, not just count it.
- Track conversion rate and revenue per visit for AI-referred shoppers.
- Compare add-to-cart rates and customer value against other channels.
- Segment landing pages for AI traffic rather than viewing AI as one source.
- Audit product, FAQ, policy and support pages for AI readability.
What teams can do now
Adobe’s research does not prove that improving readability causes higher revenue. It does show that AI referral traffic has become significantly more valuable in a year, and that content which explains products and policies is easier for AI systems to understand.
The immediate step is diagnostic: find out which pages AI systems can cite, which pages AI-referred shoppers actually land on and what those visitors do next. From there, teams can decide where to improve structure, add explanatory content and track commercial outcomes rather than visibility alone.
Source: MarTech
