Do visitors sent by ChatGPT buy more than other visitors?
Published on
Yes, in the measurements published in 2026, but these visitors are still few in number. Among the U.S. retailers tracked by Adobe, visits coming from an AI led to a purchase 90% more often than others in September 2026, and brought in 83% more per visit. In August 2025, they bought less often than average. On its own site, Ahrefs measured that AI search brought 0.5% of its visits and 12.1% of its sign-ups. The explanation put forward lies in the timing of the visit: the visitor has already compared the offers with the AI, and arrives ready to choose.
For what type of question? This is a question about the commercial value of the visits sent by AIs, asked by a company wondering whether being cited pays off. The available figures come from large U.S. sites and from software vendors; no public measurement covers French SMEs.
What Adobe measures among U.S. retailers
Adobe draws on more than 1,000 billion visits to U.S. retail sites. In its article of October 29, 2026, it counts the visitors who clicked a link from an AI. In September 2026, these visits rose 71% year on year. They led to a purchase 90% more often than visits coming from elsewhere (email, search engine advertising, social networks…), the widest gap ever observed by Adobe. They also brought in 83% more per visit.
The reversal is recent. In August 2025, these same visits bought 9% less often than the others. Adobe also observes that the visitor coming from an AI stays 73% longer, views 25% more pages and adds a product to their cart 53% more often. Its explanation: when the customer finally clicks through to the site, they have already done their research with the AI, and many are ready to buy.
What a software vendor measures on its own site
Ahrefs, a vendor of SEO tools, published on June 16, 2025 the share of AI search in its own traffic, over thirty days, with its own measurement tools. It brought 0.5% of visits, but 12.1% of sign-ups: a visitor sent by an AI signed up 23 times more often than average. Ahrefs measures sign-ups, not revenue.
Other cases are published, but you need to look at who publishes them. The agency that worked for the note-taking app Granola states that its new users coming from AIs were multiplied by 47 between January and August 2026, after work on its content and on its presence on other sites. The figure is published by the agency itself, with no independent measurement.
Why these visitors buy more
The simplest explanation is the timing of the visit. Someone who goes through an AI has asked it their questions, compared, eliminated. When they click, they are no longer discovering: they are checking a price, availability, a contact. A visit coming from an AI therefore looks more like the end of a journey than a first contact. That is also why the figure does not carry over as is: these visitors buy more because they have already been steered towards you.
This means that the decision is made partly before the visit, in the AI's response. A company that the AI does not cite does not receive these visitors, nor does it see that they went to someone else. That is what G2 observed among B2B software buyers in March 2026: 69% chose a different vendor from the one they had planned on, because it appeared in the AI's recommendation.
What these figures do not say
They do not say that an AI visitor is worth more than a visitor from Google in particular: Adobe compares with all the traffic that does not come from an AI, which mixes emails, advertising and social networks. Nor do they say how many customers an SME would gain. The volume remains low: 0.5% of visits at Ahrefs in 2025. And a high purchase rate on few visits does not, on its own, make revenue.
To find out how things stand for you, you need to count these visits in your own measurement tool: another page in this section explains how to do it in Google Analytics. Then compare, over a few months, what these visitors do and what the others do.
Where FaireDuBruit stands
FaireDuBruit, the publisher of this page, does not sell traffic and does not promise visitors. It is a distribution service: before each distribution, you receive an email, then an article is written and published on your site and on three specialist media, in your language and in English, one to three times a week depending on the plan. The aim is to multiply the sources that AIs can read when a customer looks for what you do, from €299 excl. VAT per month, with no commitment, plus a setup of €690 excl. VAT paid once.
Its limits should be known. It does not measure the visits that AIs send you and has no access to your sales. A monthly report only says which AIs read your articles on our media. The articles are written with AI functions and proofread by a second AI, and those on our media are labeled as sponsored. No citation by an AI, and therefore no visitor, is promised.
What we do not know
Adobe's figures cover retail in the United States, in 2025 and 2026, and compare visits coming from an AI with all the others. Adobe sells analytics and customer experience software. Those from Ahrefs cover its own site only, over thirty days, in 2025, and measure sign-ups. The Granola case is published by the agency that did the work. We found no public measurement on French SMEs.
Sources
- Adobe, Vivek Pandya — “AI traffic now generates 83% higher revenue per visit for U.S. retailers…”, .
- Ahrefs, Patrick Stox — conversions of AI search traffic at Ahrefs, .
- Discovered Labs — Granola case study (agency that carried out the work), .
- G2, Tim Sanders — “The Answer Economy”, .
- FaireDuBruit measurements, from September 4 to 16, 2026 — see “How we measured”.