Step 1: Identify AI search traffic in your GA4 reports
Navigate to Reports → Acquisition → Traffic Acquisition in GA4. Look for sessions where the source/medium contains values like chat.openai.com / referral, perplexity.ai / referral, gemini.google.com / referral, or claude.ai / referral. If you don't see these, your AI search visibility might be the problem — not your analytics setup.
Pro tip: Create a Traffic Source dimension filter specifically for these domains so you can isolate AI-driven visits from other referral sources. GPTBot and ClaudeBot traffic will also show up under Direct in some configurations — tag them using UTM parameters where possible.
Step 2: Create custom dimensions for AI search data
Default GA4 reports treat all referral traffic the same. That's useless for AI search optimization. You need custom dimensions to separate ChatGPT traffic from Perplexity traffic from regular social referrals.
- Go to Admin → Custom Definitions → Create Custom Dimension
- Set scope to
Event - Create dimensions for:
ai_search_source(ChatGPT, Gemini, Perplexity, etc.)ai_prompt_category(the topic of the query)ai_citation_position(where your brand appeared in the response)
These dimensions let you build exploration reports that show exactly which AI platforms drive high-converting traffic versus low-quality curiosity clicks.
Step 3: Build an AI Search Traffic Exploration report
Use GA4's Explore section to build a freeform report:
- Rows:
ai_search_sourcedimension - Columns: Sessions, Engaged sessions, Conversions
- Filters: Include only events where
sourcematches AI platforms
This report tells you at a glance which AI engine sends the most traffic (usually ChatGPT), which AI traffic converts best, and month-over-month trends.
Step 4: Track which landing pages AI users visit
ChatGPT, Perplexity, and Gemini all cite specific pages in their answers. Users click through to those exact URLs. Your Landing Pages report filtered by AI search sources shows you which of your pages have enough authority to get cited by AI engines, whether AI users land on your best-converting pages or random blog posts, and content gaps where your high-value product pages are invisible to AI.
Step 5: Set up conversion tracking for AI search visitors
The ultimate question: do AI search visitors convert? Create a Comparison in GA4: Segment 1 = All AI search traffic (filtered by source). Segment 2 = All converting users. Analyze the overlap. If AI visitors convert at 0.3% while organic search converts at 2.1%, you have a visibility problem — AI engines are citing pages that don't convert. Track this metric monthly to prove ROI to stakeholders.
System Summary
AI search traffic is real, measurable, and growing. If you're not tracking it in GA4, you're flying blind on a channel that already drives 5-15% of referral traffic and will double in the next 12 months. Set up the dimensions this week, check your numbers, and use the data to prioritize which pages need GEO optimization first.
Recommended Action
Measure Your AI Search Traffic
Author & Expert
Arjun Pandit
Founder & SEO Lead
“SEO strategist with 8+ years building organic search structures for Indian startups and SaaS platforms.”
Connect on LinkedIn →