How to Track LLM (AI) Referral Traffic in Google Analytics

AI-powered search is changing how people find and interact with websites.

Millions of users now rely on ChatGPT, Perplexity, Gemini, and other LLMs to search for information, compare products, and make purchasing decisions – bypassing traditional search engines such as Google.

We’re seeing anywhere from 0.5% to 3% of a company’s total website traffic now come from LLMs. And the chances are that it will grow.

Businesses need to track visibility, traffic, and conversions coming from these sources.

Here’s how to create a simple GA4 referral report that shows the Session and Key Event counts from LLMs, such as ChatGPT, Perplexity, Gemini, Copilot, and more:

Step 1: Access GA4 & Create a Custom Exploration Report

1. Log in to Google Analytics (GA4).
2. Click “Explore” in the left-hand menu.

3. Click “Blank” to start a new exploration.

Step 2: Configure Your Report

1. Rename the Report

• Click on “Untitled Exploration”.
• Rename it to something like “LLM Referral Traffic & Key Events”.

2. Add Key Dimensions

Tracking referral traffic requires two key dimensions:

• Session Source/Medium → Identifies where the traffic originated.
• Page Referrer → Shows the exact URL that sent the user.

To add them:

1. Click the “+” sign in the Dimensions section.

2. Search for Session Source/Medium and Page Referrer, (these will help filter referral traffic).

3. Click “Import” to add them.

3. Add Key Metrics

These metrics help track AI-driven traffic performance:

• Sessions → Total visits from LLMs.
• Key Events → Specific conversions or actions taken by visitors.

To add them:

1. Click the “+” sign in the Metrics section.

2. Search for and select:

• Sessions
• Key Events
• (Optional: Other relevant metrics, such as Engagement Time or Conversions)

3. Click “Import” to add them.

Step 3: Filter for LLM Referral Traffic

⚠ Important: If you don’t set a segment, no data will appear.

Set the segment to “All Users” for this report:

Apply the Referral Filter

1. Drag Page Referrer into the Filters section.

2. Add a “Matches Regex” filter with the following regex:

^.*(ai|\.openai|copilot|chatgpt|gemini|gpt|neeva|writesonic|nimble|outrider|perplexity|bard|edgeservices|astastic|copy\.ai|bnngpt).*$

This filter ensures only LLM referral sources are included.

📌 Tip: If you want to exclude certain sources, use the “Source/Medium Contains” filter.

And that’s it. You’re all set!

In the case of the site used to set up this sample report, we can see that across the last 28-day period (Feb 13 – Mar 12, 2025):

• The site received 1,547 visits from AI tools.
• 11 key events were triggered by LLM traffic.

Step 4: Analyze & Visualize Data (Optional)

Once your report is set up, you can adjust how the data is displayed:

• Table View → See a detailed breakdown of each referral source.
• Line Graph → Track traffic trends over time.
• Bar Chart → Compare referral traffic across different LLMs.

Bonus: Add the Date as a breakdown to analyze traffic trends by day, week, or month.

Step 5: Save & Share Your Report

1. Click “Save” to store the report in GA4.
2. (Optional) Click “Share” to generate a link or export the data as a CSV/PDF.

Final Thoughts

Tracking AI-driven traffic in GA4 helps businesses:

• Understand how LLMs impact visibility and conversions.
• Identify which AI tools drive the most engagement.
• Start measuring so you can optimize content for discovery in AI tools (LLMs).