Do This in GA4 Before It's Too LATE

Brian Dean (Backlinko)
Brian Dean (Backlinko)May 19, 2026

Why It Matters

Establishing a historical benchmark now lets you measure AI-search-driven traffic growth and attribute it to specific LLMs and content, informing SEO and content strategy. Without this baseline you’ll lack meaningful historical comparisons when AI referral traffic accelerates.

Summary

Mark a baseline in GA4 now to measure impending AI-driven search traffic. In GA4’s Explore, create a blank exploration, set primary dimension to Default Channel Group, filter for the AI Assistant channel, and compare the last 90 days with the prior 90 days—export and save the result. Then switch the primary dimension to Session Source with medium exactly matching “AI assistant” to see which LLMs (ChatGPT, Perplexity, Claude, etc.) are sending traffic and which content is being cited. Note that the feature is new and rollout may be incomplete, so some data or sources may not appear yet.

Original Description

Set your AI traffic benchmark in GA4 now, before it spikes. Otherwise you'll have nothing to compare it to.
Here's how to break down exactly which LLMs (ChatGPT, Perplexity, Claude) are sending traffic to your site.
If you're into:
AI search tracking
GA4 reports
LLM traffic attribution
SEO analytics
What LLM are you tracking first? Drop it below.
Topics covered: GA4 Explorations, AI Assistant channel filter, LLM source breakdown, 90-day benchmarking.
GA4, AI search, AI traffic, Google Analytics 4, AI Assistant channel, ChatGPT traffic, Perplexity, Claude, LLM analytics, AI SEO, GEO, AI search analytics, GA4 Exploration, session source, AI SEO 2026, SEO benchmarking, AI attribution, Backlinko, SEO tips, AI Visibility, Search Everywhere Optimization
Hashtags:
#GA4 #AISearch #AISEO #ChatGPT #Perplexity #SEO #Backlinko #AIAnalytics #LLM #SEOTips

Comments

Want to join the conversation?

Loading comments...