Google Analytics 4 launched its “AI Assistant” channel in May. I’ve since accumulated data before covering the feature here. This new channel measures and analyzes traffic from generative AI platforms, specifically tracking visitors who arrive via interactions with chatbots like ChatGPT, Gemini, and Claude. By capturing this data, the feature provides a distinct view into how users are discovering websites through conversational interfaces rather than traditional search queries.
How the AI Assistant Channel Works
According to Google’s documentation in the Analytics Help portal, the AI Assistant channel evaluates traffic based on several key performance indicators. The reporting tool displays metrics such as engagement rates, events per session, and time on site to offer a full view of user behavior. It is important to note the distinction in how Google categorizes certain interactions; specifically, clicks originating from AI Overviews and AI Mode are treated as standard Organic Search traffic rather than being classified under the new AI Assistant channel.
To access this specific data, users must handle to the acquisition reports within the interface. The process involves going to Reports > Acquisition > Traffic acquisition. From there, the “AI Assistant” Session Channel Group dimension displays the breakdown of traffic sources. This reporting capability is particularly useful for identifying specific pages that are being cited within AI answers and subsequently driving clicks. Users can locate these high-traffic pages by accessing Reports > Engagement > Pages and screens. By adding a filter for “Session default channel group” set to “Exactly matches” AI Assistant, website owners can isolate which pages perform best when driven by these AI platforms.
Related: AI Marketing Strategies Resemble Human Tactics
Comparing Performance
The feature allows for a direct comparison of AI-assisted traffic against other acquisition sources. Users can click “Add comparison” on the graph and select the same “Exactly matches” filter for the AI Assistant channel. This creates a side-by-side view, allowing analysts to see exactly how users from AI platforms behave compared to the site’s overall traffic. It is also possible to compare AI traffic against specific channels, such as organic search, by creating a separate comparison and removing “All users” from the selection. In practical tests, the top pages driving traffic from AI and organic search sources rarely overlap, highlighting a distinct pattern in how these channels contribute to site visibility.
While the default AI Assistant channel simplifies tracking by grouping sources together, it does not list the specific platforms included in the aggregate. For a detailed breakdown of individual sources, users can apply a regex filter to “Session source/medium.” Pasting a specific string of URLs for platforms like Perplexity, Copilot, and Grok reveals the exact breakdown of AI traffic. This granular approach is essential for understanding which specific conversational tools are most effective for driving visitors. One limitation in this method is that “Organic” and “(not set)” sources do not appear in the AI Assistant channel, pointing to minor data inconsistencies in the reporting.
