AI Chatbots Drive Small but Growing Web Traffic

According to Ahrefs, AI chatbots sent 3.5 million visitors to the 74,752 websites the company tracks in March 2026, equal to 0.28% of total web traffic. Ahrefs reports that ChatGPT accounted for approximately 2.7 million of those referrals, while Perplexity and Gemini each sent about 230,000 visitors. Ahrefs also reports that Google sent 345.2 million visitors in the same period and that Google's share declined from 35.11% in June 2025 to 30.53% in March 2026. Ahrefs distinguishes human click-throughs from chatbot answers (what it calls AI chatbot traffic) from automated crawlers such as GPTBot and ClaudeBot, which fetch pages for indexing but do not directly generate human visitors.

Editorial analysis - technical context: AI chatbots produce referral URLs when they cite sources inside generated answers; those clicks register in standard analytics platforms as referrals from the chatbot domain. Tracking depends on how the assistant formats links (direct URL, redirect, or no link) and whether analytics capture the chatbot domain in the referrer header. The Ahrefs dataset is based on aggregated referral logs across a large sample of websites, which provides a practical, empirical view but does not capture private or instrumented in-app clicks that lack referer headers.

Editorial analysis: For marketers and analytics teams, the raw volumes reported by Ahrefs show that AI-driven referrals are measurable but still a small fraction of overall traffic. The reported drop in Google share alongside emerging chatbot referrals highlights a distributional shift in how users find answers online, per Ahrefs' measurements. Observers should treat these numbers as an early indicator rather than a replacement for existing search traffic metrics.

Editorial analysis: Practitioners should monitor three indicators: changes in referrer share by domain in analytics, the click-through behavior tied to different assistant interfaces, and evolution in how assistants display and attribute sources. Ahrefs' dataset can be a baseline; teams with access to server logs or first-party analytics should compare those sources to validate chatbot referral attribution and conversion performance.