Study: ChatGPT ad visits earn about a third of Google search ad visits

A new working paper says a visit from a ChatGPT ad earned far less than a visit from a paid Google search ad in the United States. Search Engine Journal reports that ChatGPT ad visits produced $0.89 in revenue per visit, against $2.45 for paid search. Conversion rates were 0.37% and 1.17%.

The paper is from October 2026 and is still under review. It was written by Maximilian Kaiser of the Vienna University of Economics and Business and Christian Schulze of the Frankfurt School of Finance and Management. The data came from Google Analytics e-commerce tracking supplied by Grips Intelligence. Kaiser is a scientific advisor to Grips, and the paper says Grips had no role in the research.

What the data shows

Among 87 advertisers that bought both channels, 84 earned less per ChatGPT ad visit. Seventy-three of them had no sales at all from ChatGPT ads, across 18,726 visits. Fourteen advertisers generated all 570 ChatGPT purchases, and one advertiser accounted for 66% of purchases across the 102 sites studied.

The authors also estimate a "parity price", the click cost at which ChatGPT ads would match paid search on revenue per visit. They put it at about $0.26 per click, with a resampled range of $0.05 to $0.57. One publicly reported account paid $1.72 per click over 15 days, more than six times that estimate.

On organic traffic, the authors estimate ChatGPT referrals fell about 6% over 13 weeks after ads launched, with a wide range around that figure. Organic conversion rates showed no detectable change.

Limits you should know about

The study tracks last-click data only. It cannot show impressions, prompts or sales that happened later. Advertisers also chose to buy ChatGPT ads, so the paper cannot prove the ads caused these results. Its own pre-launch trend tests were mixed.

OpenAI's FAQ says ChatGPT Ads does not yet have performance benchmarks across advertisers, industries or campaign types. OpenAI has reported an ecommerce advertiser with a 3x return on ad spend over 28 days, and DoubleVerify reported a Weight Watchers cost per acquisition 15.3% below benchmark. Those use different metrics from the paper.

Why it matters

ChatGPT ads are new, and budgets are moving into them before benchmarks exist. This is one of the first outside datasets, and it points to lower value per click, at least on a last-click view. It is a small, early sample dominated by one advertiser, so treat it as a warning, not a verdict.

It also shows why the visit itself matters. If paid AI traffic converts at a third of the rate of search traffic, what happens after the click decides whether the spend pays off.

What to do this week

• Calculate your own parity price. Divide your ChatGPT revenue per visit by your paid search revenue per visit, then multiply by your search cost per click. Bid at or below that number.
• Filter ad-review crawler visits with tracking parameters out of your analytics. The paper's coverage notes they can be mislabeled as paid traffic.
• Cap any test budget and judge it on your own revenue data, not on platform reports.
• Check your landing pages for visitors who arrive from an AI answer with a specific question. Answer it fast, on the page or in chat.
• Keep tracking organic ChatGPT referrals separately from paid ones, so a change in one does not hide in the other.