Marketing Science · 2026
ChatGPT Referrals to E-Commerce Websites
How Do LLMs Compare Against Traditional Channels?
What the paper shows
The first large-scale empirical analysis of organic LLM traffic (oLLM) — visitors who reach online stores through ChatGPT rather than search, ads, email, or social. It draws on 12 months of first-party Google Analytics data (Aug 2024 – Jul 2025) from 973 websites with $20 billion in combined revenue: more than 50,000 ChatGPT-referred transactions measured against 164 million from traditional channels. (ChatGPT accounts for over 90% of all LLM-referred sessions.)
Key findings
A new shopping channel, measured for the first time
When someone asks ChatGPT and clicks through to a store, that visit is “organic LLM” traffic. This is the first large-scale evidence of how it behaves — across 973 stores and 50,000+ ChatGPT-referred purchases, benchmarked against every traditional channel.
Today it is a niche, not yet a mass channel
Adjusting for the data's sparsity, oLLM converts and earns more per session than paid social — but trails every other channel (organic search, email, affiliate, direct, paid search). At under 0.2% of all visits, it is promising but not yet a meaningful source of sales.
It is improving — and we can see why
Over the year, conversion rates rose while average order values fell, and the same pattern appears where shoppers are more LLM-savvy. That points to growing consumer proficiency with LLMs as the engine: more targeted, less exploratory buying.
It already wins for complex products
On stores selling complex products, oLLM's traffic share is about 4.6× higher and it out-converts several traditional channels — a high-intent complement to search, exactly where buyers need more guidance before purchase.
Downloads
Also available at the published version (INFORMS) and the SSRN preprint.
Among the most-downloaded papers on SSRN in 2025, covered in 60+ media reports worldwide — see the coverage →
Authors
Abstract
We investigate organic Large Language Model traffic (oLLM) versus traditional digital channels in e-commerce. Analyzing 12 months of first-party data from 973 websites with $20 billion combined revenue, we examine over 50,000 transactions from ChatGPT referrals alongside 164 million transactions from traditional channels. Using regression models that account for data sparsity, we assess financial metrics (conversion rate, average order value, revenue per session) and engagement metrics (bounce rate, session duration, page views). Results are consistent across extensive robustness checks. One year after launch, oLLM exhibits conversion rates and revenue per session above paid social but below all other traditional channels. Product complexity moderates the effects: oLLM's financial outcomes and traffic shares are stronger in complex product categories. Engagement metrics show favorable bounce rates but lower session duration and page views. Temporal analysis shows increasing conversion rates but declining average order values, yielding only moderate revenue-per-session gains over time. Cross-website analyses support growing consumer LLM proficiency as the underlying mechanism. The descriptive study positions oLLM as a new and developing channel. With low volumes and modest revenue per session, oLLM currently serves niche informational needs of proficient consumers and does not yet function as a broad conversion channel.
Kaiser, M., & Schulze, C. (2026). ChatGPT Referrals to E-Commerce Websites: How Do LLMs Compare Against Traditional Channels? Marketing Science. https://doi.org/10.1287/mksc.2025.0489