How to Track Claude Traffic in GA4: A Step-by-Step Setup

Visitors clicking links inside Claude to reach your website are hidden inside generic referral reports in Google Analytics 4. You can **track Claude traffic in GA4** by building a custom channel group and using UTM parameters on the links you control. This setup isolates AI-driven visits, showing how automated recommendations convert into leads.

Where Claude Traffic Shows Up in GA4

Google Analytics 4 does not automatically group Claude traffic into a dedicated AI channel. These visits are mixed into standard reporting. Claude-referred traffic typically appears as referral traffic, with the domain "claude.ai" listed as the traffic source.

Session source/medium is a GA4 dimension that identifies the specific website where a user clicked a link (the source) and the broad channel that delivered them (the medium).

GA4’s Default Channel Group categorizes these visits as "Referral," obscuring Claude's impact. To find these visitors, you must look at the specific source and medium dimensions. While GA4 tracks visits from platforms like ChatGPT and Claude, you need custom configurations to isolate them. For a broader view, you can explore the full setup for tracking AI referral traffic in GA4.

Check the Traffic Acquisition Report for claude.ai Referrals

To verify if Claude is driving traffic, you can locate this data directly within your GA4 reports. You must filter your traffic source data manually.

1. Log in to your Google Analytics 4 dashboard.
2. Navigate to Reports, click on Acquisition, and select Traffic acquisition.
3. Change the primary dimension of the table to Session source / medium by clicking the dropdown arrow next to the default dimension.
4. Type "claude.ai" into the search bar above the table to filter the results.

This search isolates sessions where users clicked a link inside Claude. If the table is empty, Claude has not sent you any traffic within the selected date range. This quick search confirms whether your site is winning organic recommendations in user conversations.

Build a Custom Channel Group for Claude Traffic

Grouping Claude traffic into its own channel prevents it from being lost in the generic referral bucket. A Custom Channel Group in Google Analytics 4 lets you view AI-driven visits as a distinct line item in your main dashboards.

1. Navigate to the Admin panel in your GA4 property.
2. Select Data display and click on Channel groups.
3. Click Create new channel group to build a custom classification.
4. Add a new channel named "AI Referrals" and set the condition to match "claude.ai" as the session source.
5. Submit and save the new group configuration.

This method adapts for tracking other platforms. You can use the same channel-group approach applied to ChatGPT traffic by adding extra detection rules. Duplicate the source condition and swap the target domain to capture multiple AI sources in a single report.

Use UTM Parameters for the Links You Control

Custom channel groups capture traffic Claude sends organically. Adding UTM parameters to your URLs tracks specific links you control. You know exactly which resource Claude is citing.

Place UTM-tagged URLs where AI crawlers and users are likely to find them. This includes your public documentation, knowledge base articles, guest posts, and press releases.

1. Define your tracking parameters by setting "claude" as the source and "referral" as the medium.
2. Append the UTM string to your target landing page URL using a link builder tool.
3. Publish the tagged link on external resources, partner sites, or public-facing assets.
4. Monitor incoming sessions in GA4 under the Traffic Acquisition report to identify which specific link prompted the visit.

UTM-tagged links bypass GA4's default categorization limits. The campaign parameters are hardcoded into the URL, so GA4 displays these sessions cleanly in your Traffic Acquisition report.

What GA4 Can't Tell You About Claude Traffic

Google Analytics 4 provides directional data, but it cannot pinpoint the exact prompt, user query, or citation that led a visitor to click your link. You only see that the session originated from claude.ai. Browser security settings and in-app redirects frequently strip away referrer headers. This means a portion of your Claude-referred visitors will land in your reports categorized as "Direct," understating your actual traffic volume. To understand these tracking gaps, analyze how referral traffic from LLMs behaves across GA4 reports.

Another limitation is automated traffic. GA4 automatically filters out known bots and spiders, but it does not catch every automated script or low-quality visit disguised as a real session. You must cross-check your analytics against an independent logging system. Visits delivered through PerkFuel are tracked in both the PerkFuel dashboard and, with analytics tracking enabled, in Google Analytics. This dual logging lets you compare numbers and confirm that the traffic showing up in your reports represents genuine activity.

Criteria for verifying Claude traffic accuracy

• Referrer data preservation — Evaluate whether your tracking setup captures the source domain even when a user's browser or security extensions block standard referral headers.
• Bot detection capability — Determine if your analytics tool can actively identify and filter out sophisticated scrapers that mimic human behavior from Claude-related domains.
• Independent logging — Ensure you have a secondary dashboard outside of Google Analytics to cross-reference and validate total session counts.

Tracking Claude Traffic vs. Connecting GA4 to Claude

Claude traffic tracking measures and analyzes the behavior of visitors who click links within the Claude interface and land on your website.

Many search results conflate this with an entirely different objective: connecting GA4 to Claude to query your data. Connecting the platforms means using Claude as an interface to ask natural language questions about your existing Google Analytics reports. These two configurations are unrelated and share no technical steps.

If you want to analyze your website metrics using Claude as an assistant, you must use third-party analytics connector tools or API integrations. Modifying your GA4 channel groups or UTM parameters will not help Claude read your reports. To see how many prospects the AI engine actually sends to your business, focus on the referral tracking configurations in this guide.

Testing Your Setup With Real Claude Traffic

Your custom channel group and tracking rules remain theoretical until real data flows through them. You need active traffic to test your GA4 setup. You can use PerkFuel to send targeted Claude visits to test your GA4 setup without waiting for organic mentions.

When setting up a project, you choose your specific AI sources, target countries, landing pages, and the delivery pace of the visits. PerkFuel supports 12 AI sources: ChatGPT, Gemini, Claude, Perplexity, Copilot, Grok, Meta AI, DeepSeek, Mistral, Manus, Poe, and Kimi. New accounts receive 10 free visits with no credit card required, providing enough data to immediately verify your UTM and channel configurations.

You can buy AI referral traffic across every supported assistant to scale up your testing. With analytics tracking enabled, these visits are tracked in both your PerkFuel dashboard and Google Analytics, allowing you to compare and validate incoming traffic side by side.

Implement a custom channel group today to cleanly separate your AI-driven sessions from standard web referrals. Privacy-focused browsers and stripped headers will still mask some Claude traffic as direct visits, making secondary tracking logs necessary. You can run an immediate end-to-end test of your new GA4 configurations by using PerkFuel's free tier to send real, targeted Claude visits to your landing pages.