If you’ve been looking at your Google Analytics 4 reports lately and wondering why your referral traffic looks a little odd or why “direct” keeps creeping up, there’s a good chance AI traffic is hiding in plain sight. Platforms like ChatGPT, Perplexity, Claude, Google Gemini, and Microsoft Copilot are sending real people to real websites every single day. And right now, GA4 has no idea how to label them properly.
Here’s the thing: how to track AI traffic in GA4 is one of the most important skills a marketer or SEO professional can pick up in 2026. This traffic isn’t just growing, it’s growing fast. AI referral traffic grew 527% year-over-year between January and May 2025. And it converts at 4.4 times the rate of traditional organic search. That’s not a channel you want buried inside “Referral” and forgotten about.
This guide walks you through exactly how to set up GA4 to catch, label, and analyze this traffic step by step, no guesswork needed.
Why GA4 Doesn’t Track AI Traffic Automatically
GA4 ships with 18 default channel groupings: Organic Search, Paid Search, Direct, Referral, Social, Email, and so on. But there’s no “AI” channel anywhere on that list.
When someone clicks a link in a ChatGPT response and lands on your site, GA4 sees chatgpt.com it as the referrer. It drops that session into the generic “Referral” bucket, technically correct, but practically useless when you’re trying to understand a completely different traffic source.
Things get murkier with mobile apps. When someone uses the ChatGPT iOS or Android app and clicks through to your site, the browser that opens often strips the referrer header entirely. GA4 logs that session as Direct traffic the analytics equivalent of “no idea where this came from.” Google acknowledged this gap in late 2024 and added a regex example to their custom channel groups documentation, but it’s still buried in the help docs. There’s no built-in fix.
The good news? You can fix it yourself in about 15 minutes.
Step 1: Do a Quick Check First (No Setup Required)
Before you build anything, check whether you’re already getting AI traffic right now. This takes about two minutes.
- Go to Reports > Acquisition > Traffic Acquisition in GA4.
- Change the primary dimension to Session source/medium using the dropdown above the table.
- Use the search box above the table and type keywords like ChatGPT, Perplexity, Gemini, Copilot, Claude, or OpenAI.
If any of those show up with sessions attached to them, congratulations, you’re already getting AI traffic. You just didn’t know it. This method is quick, but it has a hard limit: it only catches exact source matches and can’t group all AI traffic into a single view. That’s what the next steps are for.
Step 2: Create a Custom Channel Group for AI Traffic
This is the most important step. A custom channel group tells GA4 to label traffic from AI platforms as its own dedicated channel, not “Referral,” not “Direct,” just AI Traffic. The best part? It applies to your historical data too, so you don’t have to wait weeks for results.
How to Set It Up
- In GA4, click the Admin gear icon (bottom-left corner).
- Under Data Display, click Channel Groups.
- Click Create new channel group (or click the three dots next to the default group and select “Copy to create new”).
- Name your group something like Custom Channel Group, with AI works well.
- Click Add new channel.
- Name the channel Artificial Intelligence (using the full name makes filtering easier later).
- Set the condition to: Source > Matches regex – then paste in your regex pattern (see below).
- Click Save Channel.
- Click Reorder and drag your new AI channel above the Referral channel in the priority list.
- Click Apply, then Save Group.
Moving the AI channel above Referral is critical. GA4 assigns traffic to channels from the top of the list down. If Referral sits higher, AI sessions get swallowed by it before your new rule has a chance to fire.
The Regex Pattern to Use
Copy and paste this into the regex field. It covers 30+ major AI platforms as of mid-2026:
chatgpt\.com|chat\.openai\.com|openai\.com|perplexity\.ai|claude\.ai|anthropic\.com|gemini\.google\.com|copilot\.microsoft\.com|copilot\.com|edgeservices\.bing\.com|grok\.com|grok\.x\.com|deepseek\.com|you\.com|phind\.com|meta\.ai|poe\.com|mistral\.ai|character\.ai|writesonic\.com|copy\.ai|cohere\.com|bard\.google\.com|link\.edgepilot\.com|x\.ai|arcsearch\.aiImportant: GA4 regex is case-sensitive. Update this list quarterly as new AI platforms launch and existing ones change their referral domains. A pattern from six months ago may already be missing Grok, DeepSeek, or Meta AI.
Viewing Your New Channel
Once set up, go back to Reports > Acquisition > Traffic Acquisition. Use the dimension dropdown in the table and switch to your new Custom Channel Group with AI. You’ll now see your AI traffic broken out as its own line item alongside Organic Search, Direct, and the rest.
Step 3: Build an Exploration Report for Deeper Analysis
The Traffic Acquisition report gives you the big picture. But if you want to know which pages your AI visitors land on, how long they stick around, and what they do, you need GA4 Explorations.
Setting Up the Exploration
- In GA4, click Explore in the left navigation.
- Start a blank exploration and name it something like “AI Referrals.”
- In the Variables column, click the + next to Dimensions and add:
- Page referrer – the URL of the referring AI platform
- Landing page + query string – which page on your site they hit first
- Session source/medium
- In the Metrics section, add Sessions, Engaged sessions, and Key events.
- Drag dimensions and metrics into the Rows and Values sections.
- Under Filters, click + and select Page referrer > Matches regex, then paste in the same regex pattern from Step 2.
The resulting table shows you exactly how many sessions from each AI source hit which landing pages, all in one clean view. You can also create a Session segment called “AI Traffic” using the same regex on the Session source, which lets you apply that segment across multiple exploration types.
Step 4: Track AI Traffic Engagement and Conversions
Getting people to your site from AI is only half the story. The real question is: what do they do when they get there? This is where track AI traffic engagement and conversions becomes essential.
AI-referred visitors are already pre-qualified. They’ve had a conversation with an AI assistant, your site came up as a recommended source, and they clicked through. That’s a very different mindset than someone scrolling past a Google result.
Metrics to Watch
| Metric | Why It Matters for AI Traffic |
|---|---|
| Engaged Sessions | Shows whether visitors actually read and interact – not just bounce |
| Average Engagement Time | AI visitors spend up to 68% longer on sites than other traffic sources |
| Pages per Session | AI visitors view over 3x more pages per session on average |
| Key Events / Conversions | 73% of AI referral visitors convert in their first session, vs. 23% from organic Google |
| Revenue per Visit | AI traffic revenue per visit improved from 3% to 70% of non-AI traffic between July 2024 and May 2025 |
To see these numbers for your AI channel specifically, open your custom Traffic Acquisition report, filter down to your “Artificial Intelligence” channel, and add secondary dimensions like Landing page or Session source/medium. You can also build a custom Exploration with all of these metrics side by side.
Set up a custom alert in GA4 so you get notified when AI traffic shifts significantly week over week. Something like “Sessions from AI Traffic channel increased or decreased by more than 20% week-over-week” will catch major changes before they catch you off guard.
Step 5: Use Google Search Console for Google AI Mode Traffic
Here’s a blind spot that even experienced SEOs miss: GA4 cannot separate Google AI Mode clicks from regular organic Google clicks. Both show up as google / organic in your reports. You can’t tell them apart inside GA4 at all.
Google added the AI Mode search appearance filter to Google Search Console in June 2025. That makes GSC your only reliable window into this specific traffic type. Check GSC’s Performance report, filter by “Search type: Web,” and look for the AI Mode appearance filter. Pairing GSC data with your GA4 custom channel gives you the most complete picture of your total AI traffic.
Free Tools to Track AI Traffic (Beyond GA4)
GA4 is the foundation, but it has real blind spots – especially around traffic that enters as “direct” from mobile AI apps. Here are the best free tools to track AI traffic that fill those gaps.
1. Google Search Console
Free and already connected to your GA4. The AI Mode filter in GSC is currently your only way to isolate clicks from Google’s AI Mode search results. Use it alongside GA4 for a complete view.
2. Microsoft Clarity
Free behavior analytics tool with heatmaps and session recordings. In 2026, Clarity added dedicated AIPlatform and PaidAIPlatform channel groups specifically for AI-driven traffic, something GA4 still doesn’t offer by default. It’s a solid complement to GA4 for understanding what AI visitors do once they arrive.
3. Bing Webmaster Tools
Completely free. Since Bing’s index powers ChatGPT and Microsoft Copilot, Bing Webmaster Tools has become a primary window into AI-driven search traffic. Teams that previously ignored Bing are paying close attention now because AI search engines route queries through Bing’s index.
4. Similarweb (Free Tier)
Similarweb’s free AI Traffic Checker surfaces AI visits, top prompts, and the AI engines sending traffic to your site. Their GenAI Intelligence module covers ChatGPT, Google AI Overviews, Perplexity, Gemini, Claude, and Grok. Useful for competitive benchmarking alongside your own GA4 data.
5. Looker Studio (Free with GA4)
Connect your GA4 property to a Looker Studio dashboard for visual AI traffic trend reports. You can build a dedicated AI traffic page that shows volume, engagement, and conversions over time, all pulling live from your GA4 data. There are also pre-built GA4 AI Traffic Report templates available that you can connect in minutes.
Comparison: Free AI Traffic Tracking Tools
| Tool | Best For | Cost | Tracks AI Mobile App Traffic? |
|---|---|---|---|
| GA4 (custom setup) | Full referral tracking, conversions | Free | Partial (mobile often shows as Direct) |
| Google Search Console | Google AI Mode clicks | Free | No |
| Microsoft Clarity | On-site behavior from AI visitors | Free | Yes (AIPlatform channel) |
| Bing Webmaster Tools | ChatGPT + Copilot search signals | Free | Partial |
| Looker Studio | Visual dashboards, trend reports | Free | Depends on GA4 data |
Comparing GA4 AI Traffic Methods: Which One Should You Use?
| Method | Setup Time | Shows Historical Data? | Best Use Case |
|---|---|---|---|
| Traffic Acquisition Quick Filter | 2 minutes | Yes | Quick sanity check |
| Custom Channel Group | 15 minutes | Yes (retroactive) | Ongoing reporting and dashboards |
| GA4 Explorations (regex segment) | 10 minutes | Yes | Deep-dive landing page analysis |
| Page Referrer Exploration | 10 minutes | Yes | See exact AI domains sending traffic |
For most websites, the best approach is to do all of these. Start with the quick filter to confirm you have AI traffic, build the custom channel group for ongoing reporting, and use an exploration when you want to dig into specific landing pages or conversion paths.
Common Mistakes When Tracking AI Traffic in GA4
- Not reordering the channel: If your AI channel sits below “Referral” in the priority list, GA4 will assign AI sessions to Referral first and your custom channel never fires.
- Using an outdated regex: New AI platforms appear regularly. A regex from early 2025 is missing Grok, DeepSeek, Meta AI, and others. Review it quarterly.
- Assuming all AI traffic is tracked: Mobile app traffic from ChatGPT and Gemini frequently drops the referrer header and appears as Direct. Your GA4 numbers will always be a partial view, but a much better one than nothing.
- Confusing bot traffic with real AI referrals: Bot traffic from AI crawlers shows 0 engaged sessions, 0% engagement rate, and 0 seconds average engagement time. Legitimate ChatGPT referral traffic has “referral” as the medium and real engagement metrics. Always vet your data before drawing conclusions.
- Ignoring Google AI Mode: This traffic blends into organic in GA4. Without checking Google Search Console’s AI Mode filter, you’re missing a meaningful slice of the picture.
Conclusion
AI traffic is already one of the most valuable sources hitting your website, it just might not be visible yet. Learning how to track AI traffic in GA4 is no longer optional for serious marketers and SEOs. With a custom channel group, the right regex pattern, and a complementary stack of free tools to track AI traffic like Google Search Console, Microsoft Clarity, and Looker Studio, you can go from completely blind to fully informed in an afternoon.
The steps are clear: do a quick sanity check in the Traffic Acquisition report, build a custom channel group with a solid regex pattern, create an exploration for deeper landing page data, and set up weekly alerts so you never miss a significant shift. Pair all of that with GSC for Google AI Mode visibility, and you’ll have the most complete view of your AI traffic website analytics that’s actually possible today.
Start this week. AI traffic patterns move fast, and the sites that are measuring it now are the ones that will be optimizing for it first.
FAQs
1. Does GA4 track AI traffic automatically?
No. By default, GA4 dumps traffic from ChatGPT, Perplexity, Claude, and other AI platforms into the generic “Referral” channel. Some AI mobile app traffic even shows up as “Direct.” You have to manually create a custom channel group with a regex pattern to isolate and label AI traffic properly.
2. What is the best free tool to track AI traffic?
GA4 with a custom channel group is your most powerful free option for tracking AI referral traffic on your own site. Pair it with Google Search Console for Google AI Mode data, Microsoft Clarity for on-site behavior from AI visitors, and Bing Webmaster Tools for signals related to ChatGPT and Copilot search traffic.
3. Why does some AI traffic show up as “Direct” in GA4?
Mobile apps like the ChatGPT iOS and Android apps often strip the referrer header when a user clicks a link. When the browser opens your site, it has no record of where the visit originated, so GA4 defaults it to “Direct.” This is a known limitation, and no analytics tool can fully solve it without cooperation from the AI platforms themselves.
4. How do I track AI traffic engagement and conversions in GA4?
Once your custom AI channel group is set up, go to the Traffic Acquisition report, switch to your custom channel, and add metrics like Engaged Sessions, Average Engagement Time, and Key Events. You can also build a GA4 Exploration with Session segments to see conversion paths specific to AI-referred visitors.
5. How often should I update my AI traffic regex in GA4?
Review and update your regex at least once a quarter. New AI platforms launch regularly, and existing platforms sometimes change their referral domains. A pattern that covered all major sources in early 2025 may already be missing platforms like Grok, DeepSeek, or Meta AI.
6. Can I track Google AI Overview traffic in GA4?
Not directly. GA4 cannot separate clicks from Google AI Overviews or AI Mode from regular organic Google clicks; both appear as google / organic. Your best option is Google Search Console, which added an AI Mode search appearance filter in June 2025. Use it alongside your GA4 data for the most complete picture.
