Shoppers from ChatGPT and Gemini Spend 53% More: What Adobe's May 2026 Data Means for E-commerce

Shoppers arriving from ChatGPT and Gemini are becoming your best customers

Ask a large language model (LLM) about a product, then click through to a retailer from its answer. This entry point is quickly becoming the most profitable traffic source an e-commerce operator can have. New May 2026 data from Adobe Analytics puts hard numbers behind the trend.

Shoppers coming from ChatGPT and Gemini are spending more

According to new May data from Adobe Analytics, U.S. consumers who click through to retail websites from these large language models (LLMs) are lingering longer and spending more.

news.az

According to Adobe Analytics figures, U.S. consumers who reach retail sites via AI generated 53% more revenue per visit (RPV) than visitors from traditional, non-AI sources. Their conversion rate ran 54% higher, and they spent 53% longer on site. They also viewed more pages than non-AI visitors. In other words, AI-referred shoppers stay longer, browse deeper, and buy more.

Vivek Pandya, director of digital insights at Adobe, noted that when products surface in AI suggestions, retailers can deliver a much higher level of personalization to the shoppers who click through to complete a purchase. Because the AI recommends products with context already in mind, intent is largely settled by the time the visitor lands on the site.

By the numbers: AI traffic flipped from laggard to leader in just one year

What stands out is how recently this advantage took hold. AI traffic was long viewed as growing in volume but weak in quality. Between 2025 and 2026, that reputation reversed in a remarkably short window.

Adobe data reported by TechCrunch shows that in March 2025, AI traffic converted 38% worse than ordinary traffic. By March 2026 it had flipped to convert 42% better. Revenue per visit followed the same arc: ordinary traffic was once worth 128% more, yet by March 2026 AI traffic stood 37% higher. By May, that gap had widened to 53%. In barely a year, AI traffic went from dead weight to prized customer.

Volume growth is just as notable. AI-referred traffic in May 2026 rose 138% year over year, reaching the highest share since Adobe began tracking the metric in October 2024. In the first quarter, the same figure hit 393% growth year over year.

Why AI-referred shoppers spend more

The source of this quality lies in how shoppers arrive. Unlike clicking around a list of search results, an LLM-referred shopper has already worked through their requirements in conversation with the AI.

They tell the model their budget, use case, and preferences, read why a product fits, then follow the link. Most of the comparison work is done by that point, so they are less likely to bounce and more likely to buy. During the 2025 holiday season, Adobe reported that AI-referred shoppers were 33% less likely to leave a site immediately. A low bounce rate is one of the clearest signals of traffic quality.

A second factor is that the AI narrows products to a specific context. Ask for "waterproof hiking boots under $200" and the LLM returns candidates that match. The shopper faces a low-noise set of choices and moves to purchase with confidence. In Adobe's survey, 85% of people who shopped with AI said it improved their experience, and that satisfaction likely feeds repeat visits and higher order values.

What e-commerce operators should do: AI search optimization and product data

The numbers above point to a clear agenda. The problem is that many retail sites are still not in a state that AI can read.

Adobe warns that AI traffic is surging while retail sites' AI search visibility lags behind. Roughly 25% of homepage and category page content is not optimized for LLMs, and 34% of product pages cannot be properly accessed by AI systems. LLMs read product names, prices, availability, specifications, and reviews as structured data. Without that, a product cannot even enter the recommendation set.

The work centers on a new discipline known as AEO (AI Engine Optimization) or GEO (Generative Engine Optimization). Where traditional SEO competed for rank in Google's results, AEO aims to be cited and recommended in an LLM's answer. The starting point is to structure product data in machine-readable form, apply structured markup, and provide descriptions trusted as primary sources. At Stellagent we have covered agentic commerce and the practice of AEO on an ongoing basis.

Here is the AI versus non-AI comparison at a glance.

Even after an AI-referred shopper lands, the page has to live up to the recommendation. The destination needs to back up the reasoning the AI used to decide the product fit. Accurate product information, real-time inventory, and a short path to checkout are what keep this high-quality traffic from slipping away.

Conclusion

Adobe's May 2026 data makes it clear that AI traffic now beats other channels not only in volume but in quality. A 53% lift in revenue per visit and a 54% lift in conversion mark a sharp reversal from the laggard position of just a year earlier. LLMs have settled in as an entry point for shopping, funneling buyers with formed intent to retail sites.

Whether you can ride this wave depends on whether your site is readable by AI. Product-data readiness and AI search optimization are no longer a speculative bet but a response to a revenue opportunity already in front of you. With the quality of AI traffic this high, now is the time to prepare to receive it.