AI Traffic Is a Sliver of Your Sessions. Here Is Why It Still Matters

"AI traffic is one percent of our sessions. Why should I care?" It is a fair question, and the honest answer is that the share is small and the shoppers inside it are not. Adobe Analytics, which tracks more than a trillion visits to US retail sites, found that visitors arriving from AI tools converted 42% better than other visitors in March 2026. A year earlier, in March 2025, the same cohort converted 38% worse.

This post lays out what Adobe measured and when, why AI-referred shoppers behave differently, how to separate them from Google Shopping in your own analytics, and what a before-and-after view looks like. The dates matter, so each figure carries one.

What Adobe measured, and when

Adobe's report, covered by TechCrunch on April 16, 2026, found that AI traffic to US retail sites rose 393% year over year in the first quarter of 2026. The volume story is the headline. The quality story is the part that changes how you plan.

In the same report, Adobe found that shoppers who landed on a retail site from an AI source engaged 12% more, stayed 48% longer, viewed 13% more pages per visit, and produced revenue per visit 37% higher than non-AI traffic as of March 2026. The conversion comparison moved from 38% worse in March 2025 to 42% better in March 2026.

The earlier snapshot points the same way. For the 2025 holiday season, Adobe reported that AI referrals converted 31% better than other traffic sources, with revenue per visit up 254% and traffic from generative AI tools up 693% year over year. So the March 2026 figure was not a one-off reading. It was the continuation of a trend Adobe had already seen over the holidays.

On the consumer side, Adobe surveyed more than 5,000 US consumers alongside the traffic data. 39% said they had used AI for online shopping, 85% of that group said it improved their experience, and 66% said they believe AI tools provide accurate results, according to the Yahoo Finance write-up of the same report.

The forward view is consistent too. In its holiday forecast published in late September 2026, Adobe expects AI traffic to US retail sites to rise 130% year over year this holiday season, with the largest jump on Thanksgiving. We covered what to check before that traffic arrives in our holiday product information checklist.

Why AI-referred shoppers convert better

Three things are happening at once, and none of them is mysterious.

First, the assistant does the research before the click. By the time a shopper asks ChatGPT or Google AI Mode for "a crib mattress that fits a standard US crib under $200" and clicks a recommendation, the comparison and most of the objections are already handled. The click lands closer to the purchase decision than a typical search or social click.

Second, in most cases the purchase still happens on your site. The assistant sends the shopper to the product page, so the whole AI funnel flows into the checkout you already measure and optimize. The visitor arrives pre-qualified and buys where you can see it.

Third, the products that get recommended are the ones whose information the assistant could read and match. A product with a complete title, plainly stated attributes, a current price and in-stock availability can be matched to a specific question. One with a thin page and a stale feed is easy to leave out. The recommended products are, by construction, the ones that answer the shopper's question well, which is also why they convert.

Who is capturing this traffic

The distribution is uneven and some of it is public. Similarweb data reported by Digiday in September 2025 showed that one in five of Walmart's referral clicks in August 2025 came from ChatGPT, with more than 20% at Etsy and nearly 15% at Target. Digiday noted in the same piece that ChatGPT referrals remained a sliver of overall traffic for those retailers.

On the platform side, Shopify said in November 2025 that AI traffic to its merchants was up seven times since January 2025 and AI-driven orders were up eleven times. Growth from a small base, again, but orders growing faster than traffic is the same quality signal Adobe saw.

How to split AI traffic from Google Shopping

You cannot plan around a number you cannot see. GA4 added a native AI Assistant channel in May 2026 that catches referrals from ChatGPT, Gemini and Claude, but visits that arrive without a referrer still land in Direct, and Google AI Mode traffic still counts as Organic Search. We walked through the setup, including a custom channel group for the assistants GA4 misses and how to reconcile GA4 with Shopify's own analytics, in how to measure AI referral traffic in GA4 and Shopify.

Once the channel exists, record a baseline: AI-referred sessions, the landing pages they hit, and conversion rate on those pages for the last 30 days. Then compare that conversion rate with Google Shopping and organic search on the same pages. If your own data looks like Adobe's, the case for the work makes itself. If it does not, you have learned something specific about which products the assistants are sending shoppers to.

What before and after looks like

Attribution alone does not tell you why the assistant recommended a competitor. For that you need a fixed set of shopper questions and a habit of running them.

• Referral sessions by assistant. ChatGPT, Google AI Mode, Perplexity, Copilot and Gemini reported separately, so a change in one is not hidden by another.
• Conversion rate by source. AI referrers next to Google Shopping, organic and paid on the same landing pages.
• Product card versus text mention. For a set of twenty or so category questions written the way a real shopper would ask them, note whether you appear as a product card, a text mention or not at all, and who appears instead. Run them before the product information fixes and again after.

Appearances in answers, AI-referred sessions and orders are three different numbers. Report them separately and do not let a growth percentage on one stand in for the others. If you already work with an SEO agency and use Semrush, this sits alongside that work rather than replacing it. Classic SEO keeps the pages ranking. This checks whether the assistant can read the product facts on them and whether it names you in the answer.

The honest caveat

The 393% growth figure is inflated by a small base and will not extrapolate forever. Adobe said as much in its holiday 2025 recap, noting that the base of users remains modest. The more durable signal is the conversion reversal, from 38% worse to 42% better in twelve months, backed by the holiday 2025 reading of 31% better.

Even if the growth curve flattens, the traffic that does arrive from AI is measurably higher value per visit than the rest of the mix, and that is what to plan around.

Frequently asked questions

Why is AI-referred traffic converting better than other traffic?

The assistant handles research, comparison and objections before the shopper clicks, so the visitor lands closer to a purchase decision. Adobe's March 2026 data shows the same cohort engages 12% more, stays 48% longer and views 13% more pages per visit, which is consistent with a shopper who has already chosen and is confirming.

Is the 393% growth stat meaningful given the small base?

The growth rate is a base effect, yes. The conversion rate comparison is not. Going from 38% worse to 42% better is a change in the quality of the traffic itself, measured on the same sites by the same method a year apart, and it is the figure retailers should plan around.

AI is one percent of our sessions. Where should we start?

Make the traffic visible first, using the GA4 and Shopify method above, so you have a baseline. Then pick the products that carry your revenue, write the questions shoppers ask about them, and check whether the assistants recommend you. A free AI readiness report shows how one product page reads to ChatGPT, Google AI and Perplexity, and the AI visibility service runs the full cycle.

This is the work Paz does for lean ecommerce teams. A Paz expert reviews where your products appear in ChatGPT and Google AI Mode and what is missing from the product information behind them, then proposes the fixes with the evidence for each. You approve, and the platform publishes the changes and tracks referrals and conversion before and after. Talk to us.