Tracking AI-Referred Traffic in GA4: The Step-by-Step Guide

To measure AI-referred traffic in Google Analytics 4, create a segment or exploration that filters visitors whose source matches AI assistant domains (chatgpt.com, perplexity.ai, gemini.google.com, etc.), then track their conversions. GA4 doesn’t yet offer a native “AI” channel, but a few minutes of setup are enough to isolate this traffic.

Here’s the method, along with the list of domains to watch.

Why this isn’t automatic in GA4

By default, GA4 files AI traffic under “Referral” or sometimes “Direct,” without distinguishing it. AI assistants don’t always pass clear origin parameters, and some send traffic with no identifiable referrer. The result: without configuration, your AI traffic is invisible, buried in other channels. Measuring it means defining it yourself.

AI referral domains to track

These domains show up as source or referrer when a user clicks through from an AI answer:

• chatgpt.com and chat.openai.com (ChatGPT)

• perplexity.ai (Perplexity)

• gemini.google.com (Gemini)

• claude.ai (Claude)

• copilot.microsoft.com and bing.com/chat (Copilot)

• you.com, poe.com, and other assistants

Tip: build a pattern-based filter (contains “perplexity,” “chatgpt,” etc.) rather than a rigid list, to catch subdomain variants.

Method 1: a dedicated segment or exploration

1. Open Explore, create a blank exploration.

2. Add the Session source dimension (or Session source/medium).

3. Create a segment with the condition: “Session source matches one of the AI domains above” (use the “matches regex” operator to capture everything at once).

4. Cross it with your metrics: sessions, users, conversion events.

Useful regex example for the Source field:
chatgpt · openai · perplexity · gemini · claude · copilot · you\.com · poe\.com

Method 2: a custom channel group

To see “AI” as its own channel in your standard reports:

1. Go to Admin > Channel Groups, create a custom group.

2. Add an “AI / Assistants” channel rule with the same regex condition on source or referrer.

3. Place it at the top of the rule order so it catches traffic before the generic channels.

This channel will then appear in your acquisition reports, alongside Organic, Direct, and Referral.

Tracking conversions, not just visits

The real challenge is connecting this traffic to value. Once the segment or channel is in place:

• Compare the conversion rate of AI traffic against your other channels.

• Track the key events (demo, signup, purchase) attributed to AI.

• Watch behavior (pageviews, duration): AI traffic is often more qualified, since the user arrives with clear intent.

Limitations to know

• Traffic with no referrer: some AI answers don’t pass a source, so a portion of traffic stays unattributable. Your numbers are a floor, not a total.

• Shifting domains: assistants change URLs and behavior; update your regex regularly.

• Citation without a click: GA4 only measures visits. But a lot of GEO value happens without a click, when AI cites your brand without the user visiting. For that portion, an AI visibility tracking tool is complementary.

FAQ

Will GA4 ever get a native AI channel?
It’s likely as AI traffic grows, but nothing official yet. In the meantime, manual configuration remains necessary.

Why does my AI traffic show up as “Direct”?
Because the assistant didn’t pass a referrer. It’s common. The source regex catches what’s identifiable; the rest falls into Direct.

Is it worth tracking AI traffic if it’s small?
Yes: it’s growing fast and often converts well. Measuring it early gives you a valuable baseline for the months ahead.

The click is only part of the story. To also measure citations without a click.
Discover vizibAI.