Data report
AI as a shopping channel: retail traffic up fourfold
Traffic to retail sites from AI sources rose almost fourfold in a year, and most consumers have already tried shopping with AI. Data with sources.

Traffic to US retail sites from AI sources rose 393% year over year, and around 60% of consumers have already tried shopping with AI. An important caveat: the growth is measured from a very low base, so the percentage is more striking than the absolute volume.
Key figures
- Traffic to US retail sites from AI sources: +393% year over year Industry AI commerce research, 2026
- Around 60% of consumers have already used AI for shopping Consumer surveys, 2026
- ≈80% plan to use generative AI for shopping in 2026 Consumer surveys, 2026
- ≈76% would like an AI shopping assistant Consumer surveys, 2026
How fast is the channel growing?
Very fast in percentage terms. Traffic-source data shows traffic to US retail sites from AI sources rising 393% comparing Q1 2026 with Q1 2025 Industry AI commerce research, 2026 .
Read that with the base in mind. Fourfold growth from a small volume produces an impressive percentage and a modest absolute increase — which is exactly the stage this channel is at.
How many people already shop this way?
More than most sellers expect. Surveys show around 60% of consumers have already used AI for shopping Consumer surveys, 2026 , and roughly 80% plan to do so in 2026 Consumer surveys, 2026 .
The second figure is intent and should be discounted: the gap between «plan to» and «do» in surveys about new technology is usually significant. The first is more reliable because it describes experience that already happened.
What do shoppers themselves want?
Roughly 76% say they’d like an AI shopping assistant Consumer surveys, 2026 — with the same caveat about stated preference.
More interesting in practice: people describe the value not as «talking to an AI» but as «finding what fits faster». The demand exists for the outcome, not the interface.
What changes for a store?
The nature of visibility. In classic search a page can rank fifth and still receive traffic. An AI answer usually doesn’t hold ten options — you’re either mentioned or you aren’t.
That makes structured product data and verifiable facts on the page more practical than before. Not as «optimising for AI», but as a condition of appearing in an answer that has few slots.
Where do these figures come from?
This report aggregates 2025–2026 consumer behaviour research and traffic-source data for retail sites. Intent metrics come from surveys and describe stated rather than observed behaviour. Traffic growth is measured from a low base, so percentage change isn’t proportional to absolute volume; the data relates primarily to the US market.
Frequently asked questions
-
01 How much did AI-sourced traffic grow?
By 393% year over year, comparing Q1 2026 with Q1 2025 across US retail sites. The growth is measured from a very low base, so the percentage is more striking than the absolute volume.
-
02 How many people have already shopped with AI?
Around 60% of consumers have done so, and roughly 80% plan to use generative AI for shopping in 2026. The second figure is intent rather than behaviour, and is typically inflated.
-
03 Do shoppers want AI assistants in stores?
By survey, yes: roughly 76% say they'd like an AI shopping assistant. That's stated preference; actual usage is lower, as it always is with new interfaces.
-
04 What does this mean for store visibility?
That a channel has appeared where you're either cited or you aren't. Unlike search, an AI answer usually doesn't present ten options — so visibility here is more binary than in classic SEO.
-
05 Is AI replacing classic search in shopping?
The data doesn't show that. It shows rapid growth of a new channel from a low base. Absolute search traffic volumes remain orders of magnitude larger, so it's more accurate to describe this as an additional channel rather than a replacement.



