A new academic study finds that traffic from ChatGPT and other large language model (LLM) platforms is starting to gain ground in ecommerce but still falls well short of traditional digital marketing channels.

The research — “ChatGPT Referrals to E-Commerce Websites: Do LLMs Outperform Traditional Channels?” — was conducted by Professor Maximilian Kaiser of the University of Hamburg and Professor Christian Schulze of the Frankfurt School of Finance & Management. It represents one of the first large-scale, data-driven evaluations of how conversational AI affects online sales performance.

The study compared how shoppers referred by ChatGPT performed against those arriving through established digital channels such as organic search, paid search, email and social media. Drawing on 12 months of ecommerce data from August 2024 through July 2025, the researchers analyzed 973 websites with a combined $20 billion in annual revenue. The dataset included more than 50,000 transactions from ChatGPT referrals and 164 million transactions from traditional channels, making it one of the largest first-party data sets yet assembled for this kind of analysis.

How ChatGPT compares to other digital marketing channels

According to the report, organic LLM (oLLM) traffic underperformed every traditional channel except paid social media across key financial metrics such as conversion rate, average order value and revenue per session. While ChatGPT referrals achieved lower bounce rates — suggesting that users found relevant content — they generated fewer completed purchases and lower revenue than both paid and organic search traffic from Google.

The authors wrote that their results “contradict widespread expectations of oLLM superiority,” adding that while ChatGPT traffic is “clearly relevant to users,” it “does not yet translate into comparable sales outcomes.”

Even so, the researchers found early signs of progress. Conversion rates for oLLM traffic improved steadily throughout the 12-month period, though declining average order values muted overall gains.

“Time-trend analyses suggest gradual convergence with traditional channels, but projections indicate oLLM will not achieve parity with organic search within the next year,” Kaiser and Schulze wrote.

Those findings held up across multiple robustness checks, including variations in data aggregation, observation thresholds and sample composition. The consistency of results, the authors noted, reinforces the conclusion that oLLM traffic remains a developing acquisition channel rather than a disruptive one.

Kaiser and Schulze concluded that the evidence “challenges narratives of LLMs as immediate Google killers” but also “points to potential for long-term channel evolution.”

Their research offers one of the clearest pictures yet of how conversational AI is reshaping digital commerce. While ChatGPT referrals are beginning to influence product discovery and engagement, the study shows that their commercial impact remains modest — growing in relevance, but not yet ready to rival search-based traffic as a dominant driver of online sales.

Sign up

Sign up for a complimentary subscription to Digital Commerce 360 B2B News. It covers technology and business trends in the growing B2B ecommerce industry. Contact Mark Brohan, senior vice president of B2B and Market Research, at mark@digitalcommerce360.com. Follow him on Twitter @markbrohan. Follow us on LinkedIn, X (formerly Twitter), Facebook and YouTube.