How to Track AI Referral Traffic in GA4 (Step-by-Step)

ChatGPT, Gemini, Perplexity, and other AI platforms have grown quickly, so more businesses are seeing these tools refer relevant traffic and leads to their pages. That traffic tends to be worth more than most. Visitors who arrive from an AI assistant often come with high intent and context, and many teams report that AI-referred visitors convert at a higher rate than visitors from traditional organic search.

That is why it pays to learn how to track AI referral traffic. Once you know where your leads come from, you can optimize your content strategy around it and measure ROI more accurately.

So how do you do that? Below, we cover the GA4 blind spot you have to work around first, then four steps you can follow to get started.

First, understand GA4's AI traffic blind spot

Before you start, it helps to know why AI referral traffic is harder to find than it should be.

Google Analytics 4 (GA4) has no built-in channel for AI assistants. By default, the platforms you most want to measure get scattered:

• ChatGPT, Perplexity, Gemini, and Copilot often land in Direct traffic, because the referrer is stripped when a user clicks through from an app or an in-chat link.
• When a referrer is passed, the visit usually falls into a generic Referral bucket alongside every other website instead of being recognized as AI traffic.

So unless you go looking for it, AI referral traffic hides inside Direct and Referral, and you underestimate how much of your pipeline AI search is already driving. The steps below show you how to surface it.

1. Check your referral sources

To start tracking AI referral traffic, you need to look at what sources are referring your business. ChatGPT, Claude, Perplexity, Gemini, Copilot, and others can all be sending traffic to your website.

To figure out which ones, head into GA4 and follow this path:

Reports → Acquisition → Traffic acquisition

Once you're in the Traffic acquisition report, change the primary dimension to Session source / medium (use the dropdown above the table). Then add a search filter and look for the domains and source names that AI platforms use, such as:

• chatgpt (and chat.openai.com / openai)
• perplexity (perplexity.ai)
• gemini (gemini.google.com)
• copilot (copilot.microsoft.com)
• claude (claude.ai)

These referral sources are usually tagged with the website URL they came from (for example, gemini.google.com or perplexity.ai). Some visits will show up as "direct" because the referral data was incomplete or stripped, but the visits that do carry a referrer get recorded with the proper URL. Searching one platform at a time gives you a quick, reliable read on which AI sources are already sending you traffic.

2. Consolidate AI traffic into a custom channel

Checking platforms one by one is fine for a quick look, but to review all your AI traffic in one view, you need to group it together. The most durable way to do that in GA4 is to build a custom channel group that pulls every AI source into a single "AI / LLM Traffic" channel.

To set it up:

1. Go to Admin → Data settings → Channel groups (under the property column).

2. Click Create new channel group, give it a name, and add a new channel called AI / LLM Traffic.

3. Define that channel with a regex rule that matches AI sources. Apply it to Source using a condition like "Source matches regex":

   chatgpt|openai|perplexity|gemini\.google|copilot|bing\.com/chat|claude\.ai|you\.com
   

4. Drag the new AI channel above the Direct, Organic, and Referral channels so AI traffic is classified first, then save.

One advantage worth knowing: custom channel groups in GA4 apply retroactively, so you can analyze AI traffic for past date ranges, not just going forward. Once the channel exists, you can view all AI referral traffic in one place and compare platforms side by side. You might spot that Perplexity is driving more AI referral traffic than Gemini, for instance, which tells you to lean in on Perplexity optimization. When the data sits in separate buckets, those trends are much harder to see.

Prefer not to touch your channel groups? You can get the same consolidated view with a GA4 Exploration instead. Create a Free-form exploration, add Session source (or Page referrer) and Landing page as dimensions, pull in Sessions and Conversions as metrics, and apply a filter that matches the same list of AI sources. This approach analyzes AI referral traffic without changing how the property classifies channels.

3. Decide your tracking cadence

Think about how often you want to track your AI referral traffic. Some companies check it every week, while others only look once a month.

AI search and user behavior are still shifting quickly, so check in at least once a week to see how much referral traffic you're getting. That way you can act on any trends or patterns you spot.

And if one or more platforms are a major traffic source for you, keep a close eye on whether that performance is rising or falling, so you can make decisions based on current numbers.

4. Monitor for patterns

When you track AI referral traffic, look for patterns and trends. Like traditional search, AI search engines show seasonal swings and longer-term growth. Catching those shifts and acting on them helps you grow your referral traffic from these platforms and understand your visibility better.

So what do you do when you spot a pattern? Look back at the work you did before that spike. Did you optimize content, add schema markup, or change something else? Tying the change back to a specific action tells you where to focus next.

A note on GA4's new native "AI assistants" channel

If you'd rather not maintain a workaround, there's relevant news. On May 14, 2026, Google announced a native "AI assistants" channel in GA4 that groups traffic from ChatGPT, Gemini, Perplexity, Copilot, and similar platforms automatically.

As that channel rolls out across properties, it reduces the need to maintain your own regex-based channel group. Until it's fully available in your account and backfilled to your satisfaction, the custom channel group and Exploration methods above stay the most reliable way to capture AI referral traffic, so keep them in place even after the native channel shows up.

Keep tabs on your AI search performance with OmniSEO®

Tracking your AI search visibility matters if you want to understand how it shapes your overall search performance. Google Analytics 4 gives you part of the picture, but only part, because GA4 only sees traffic after the click.

A tool like OmniSEO® tracks the visibility behind that click. You can see which prompts you show up for, where you're visible, and how you stack up against competitors, the visibility that drives the referral traffic GA4 measures after the fact.