What Is AI Traffic Attribution?

AI traffic attribution is the practice of identifying which website sessions, signups, and conversions originated from AI platforms, ChatGPT, Perplexity, Gemini, Copilot, Claude, and crediting them correctly in analytics. It matters because AI referral traffic, while still a small share of most sites' totals, converts unusually well: visitors arrive pre-qualified by an assistant's recommendation.

The GA4 recipe

The foundation is a referrer-domain segment. In GA4, create a comparison or exploration filtering Session source / medium with a regex such as:

chatgpt\.com|perplexity\.ai|gemini\.google\.com|copilot\.microsoft\.com|claude\.ai

Then register it as a custom channel group ("AI Assistants") so AI sessions appear alongside Organic and Paid in standard reports. OpenAI helpfully appends utm_source=chatgpt.com to links clicked inside ChatGPT, giving you campaign-level cleanliness. Route the channel into conversion reports to compare AI-referred conversion rates against organic, the delta is usually the headline finding.

Known measurement gaps

• Zero-click influence: the majority of AI answer exposures produce no click, so referral data measures the tip of the funnel, not the funnel.
• Stripped referrers: some AI surfaces and in-app browsers send no referrer, and those sessions collapse into dark traffic labeled Direct.
• Delayed brand visits: a user who reads an AI recommendation and Googles your brand two days later is credited to Organic.
• Assistant-side browsing: when an agent fetches your page server-side (e.g. ChatGPT browsing on the user's behalf), no analytics session fires at all.

Example

A dev-tools company found AI-referred sessions were 0.9% of traffic but 6% of trial signups, a 6.7x conversion index versus the site average. That ratio, not raw session counts, justified the GEO budget. Closing the loop requires connecting attribution to answer-level data: citation tracking shows which prompts and pages produce the citations that referral analytics then converts into revenue.