How to Track ChatGPT and AI Referrals That GA4 Marks as Direct Traffic

When potential customers ask ChatGPT, Claude, Perplexity, or Gemini for product recommendations, standard web analytics tools like Google Analytics 4 (GA4) routinely misclassify those visits as (direct) / (none).

Why AI Referrals Are Lost as Dark Traffic

Traditional analytics engines rely heavily on standard HTTP Referer headers. However, modern LLM interfaces and mobile chat apps often strip or obfuscate referrer headers when users click embedded link recommendations.

As a result, your marketing reports show an unexplained surge in “Direct” conversions, masking the true return on your AI search optimization and content investments.

The SourceTrack Approach to AI Attribution

SourceTrack uses first-party browser inspection heuristics to detect AI search sessions without relying on third-party tracking cookies or invasive device fingerprinting.

1. Header & Query Pattern Analysis: Detects incoming traffic signatures from AI search engines and conversational assistants.
2. First-Touch Stitched Timeline: Preserves the originating AI discovery channel across subsequent visits before purchase.
3. Revenue Attribution: Connects Stripe purchase webhooks and Shopify orders directly to the initial AI referral touchpoint.

By filling the dark traffic gap, growth teams get clear visibility into which specific AI search platforms drive real enterprise pipeline and recurring revenue.