ChatGPT LLM Referrals Convert Worse Than Google Search: Study

iTech Manthra covers the study by Maximilian Kaiser and Christian Schulze demonstrating that traffic from ChatGPT and other large language models generates weaker conversion rates compared to Google Search traffic. The article examines the research findings on why LLM-generated referrals underperform, highlighting factors such as user skepticism toward AI tools leading to verification through traditional search before purchasing, the negligible volume of AI referral traffic, attribution challenges that typically credit final clicks to Google rather than initial AI assistance, and how AI responses often satisfy user queries without requiring website visits. The piece discusses the study’s analysis of 12 months of data from 973 e-commerce websites, revealing that organic LLM traffic underperforms all traditional channels except paid social media across key financial metrics. The article recommends marketers maintain a balanced approach combining traditional SEO with Answer Engine Optimization (AEO) strategies while monitoring emerging AI traffic patterns.