Measuring traffic from AI models

Measurement service

The AI channel your reports can't see

Standard GA4 misfiles traffic from language-model answers as referral or direct. It arrives with no channel grouping and no attribution, so a channel that's quietly growing stays invisible in your reports. On top of that sits a newer problem: agentic traffic - an agent acting on a user's behalf - which GA4 doesn't separate from a bot or a human.

What I set up

• Channel grouping rules and a source list that route model traffic into its own channel
• Custom dimensions and attribution configuration for that channel
• Separation of agent traffic from human traffic in reporting
• A report on this channel's share of revenue

How the method stays correct

The source list and grouping rules are built to your stack and kept current as models and their referrers change - there is no fixed pattern that stays accurate for long, so this is maintained, not set-and-forget.

What this is not

This is measurement only. It tells you how much the AI channel is worth - it does not change how models rank or cite you, and I don't promise any visibility inside a model. If you also want to shape what models can read from your site, that's the data-layer job of my AI Visibility product - a separate thing, which this service can measure but does not replace.