Quick answer
The simplest way to find ChatGPT traffic in GA4
Open Google Analytics 4, go to Reports → Acquisition → Traffic acquisition, change the primary dimension to Session source, and look for chatgpt.com.
OpenAI states that referral URLs from ChatGPT search automatically include utm_source=chatgpt.com. That gives GA4 a source value you can use to identify sessions that began from a ChatGPT link.
This measures visits that actually click through. It does not measure every time your brand appears in a ChatGPT answer. Referral analytics and AI visibility measurement answer different questions.
What ChatGPT sends to your analytics
When someone clicks a website link from a ChatGPT search result, OpenAI says the destination URL includes the source parameter utm_source=chatgpt.com.
Attribution signal
utm_source=chatgpt.comGA4 uses traffic-source information to determine where sessions come from. For this analysis, Session source is the safest starting dimension because the question we are answering is straightforward: which sessions began with ChatGPT as the source?
Do not depend only on a default channel such as Referral. Channel classifications are broader categories. The source value gives you a much more precise way to isolate ChatGPT.
How to find ChatGPT referral traffic in GA4
Open the Traffic acquisition report
In GA4, navigate to Reports → Acquisition → Traffic acquisition.
Use the Traffic acquisition report rather than User acquisition for this analysis. Traffic acquisition uses session-scoped dimensions, so a returning visitor can still be attributed to ChatGPT when a new session begins from a ChatGPT link.
Switch the dimension to Session source
Change the primary dimension in the report to Session source.
You can also inspect Session source / medium later, but Session source gives you the cleanest first check for the source value OpenAI documents.
Filter for chatgpt.com
Search the report for:
chatgpt.comIf sessions exist, you now have your basic ChatGPT referral traffic view.
Add landing-page context
Add Landing page + query string as a secondary dimension, or use the Landing page report and add Session source / medium as the secondary dimension.
This tells you which page first received the visitor rather than just telling you that ChatGPT sent traffic somewhere on the site.
Evaluate what those sessions actually did
Traffic volume alone is not enough. Compare sessions from ChatGPT using metrics such as:
- Sessions
- Engaged sessions
- Engagement rate
- Average engagement time
- Key events
- Revenue, when applicable
Build a repeatable ChatGPT traffic report
The standard Traffic acquisition report is enough for a quick check. For ongoing analysis, create a GA4 Exploration so the same view can be reopened without rebuilding the filters every time.
Suggested GA4 Exploration
Dimensions
- Session source
- Session source / medium
- Landing page + query string
- Page title
Metrics
- Sessions
- Active users
- Engaged sessions
- Key events
Apply a filter where Session source exactly matches chatgpt.com.
Google Analytics Explorations let you combine dimensions and metrics into a reusable free-form table. The result becomes a lightweight AI referral dashboard without needing another analytics product.
Find which pages ChatGPT actually sends visitors to
This is often more useful than the total number of AI referrals. Landing-page data shows which pieces of your website are turning AI search visibility into real site visits.
Educational pages
A guide receiving ChatGPT referrals may be answering a question that users ask directly inside the assistant.
Commercial pages
Traffic to product, pricing, service, comparison, or use-case pages can indicate stronger evaluation intent.
Unexpected pages
A page you did not consider strategically important may be the source ChatGPT currently finds useful enough to surface.
No-click visibility
A page can still influence an AI answer without receiving a click, so referral traffic should not be treated as the complete visibility picture.
Measure whether ChatGPT traffic converts
Once ChatGPT sessions are isolated, measure them against the actions that matter to the business.
In GA4, important business outcomes should be represented by events and, where appropriate, marked as key events. Depending on your business, that might include a demo request, trial signup, qualified form submission, purchase, account creation, or another meaningful conversion.
Sessions
How much measurable click-through traffic did ChatGPT send?
Engagement
Did those visitors actually consume the page after landing?
Key events
Did ChatGPT-referred sessions trigger meaningful business actions?
Landing pages
Which pages convert AI-search visibility into site visits?
Revenue
If revenue is available in GA4, did any attributable commercial value follow?
Why GA4 cannot measure all ChatGPT influence
Referral traffic is observable, but it is only one part of AI search performance.
The user may never click
ChatGPT can answer a question, mention your company, or shape a buying decision without sending the user to your website.
The user may return another way
Someone may discover your brand in ChatGPT and later type the domain directly, search the brand on Google, or return through another channel. GA4 will not necessarily attribute that later visit to ChatGPT.
Tracking information can be lost
Redirects, privacy tooling, browser behavior, and other technical conditions can interfere with referral or campaign information.
Referral traffic does not reveal prompt visibility
GA4 tells you that a click happened. It does not tell you the full prompt set where your brand appeared, which competitors appeared, or how often you were recommended without a click.
Combine GA4 referral data with AI visibility tracking
GA4 measures the downstream visit. It does not measure the complete upstream AI-search environment.
A more useful measurement system combines referral analytics with a broader AI search visibility framework.
GA4
Sessions, landing pages, engagement, key events, revenue, and observable referral behavior.
Prompt-level tracking
Brand mentions, recommendation frequency, competitor presence, citation visibility, context, and change across a repeatable prompt set.
If you need a repeatable prompt methodology, use our prompt-level AI visibility tracking playbook.
Common mistakes when measuring ChatGPT traffic
Looking only at default channel groups
Use the actual Session source so you can isolate chatgpt.com instead of depending on a broad channel classification.
Using First user source for every question
First user source answers how the user was originally acquired. Session source is usually more appropriate when you want to know which sessions arrived from ChatGPT.
Counting sessions without examining landing pages
Knowing that ChatGPT sent ten sessions is less useful than knowing which pages attracted those visits and what happened next.
Treating traffic as total AI visibility
A brand can be visible without earning a click. Referral traffic and AI answer visibility should be measured separately.
Ignoring conversion quality
A small number of highly qualified AI-referred visits can matter more than a much larger pool of low-intent traffic.
What if GA4 shows no ChatGPT referral traffic?
Zero measurable referrals does not automatically mean zero AI visibility.
First confirm that your GA4 installation is working and that campaign parameters are not being removed by redirects. Then separate the traffic question from the visibility question.
Use an AI visibility audit to establish whether your site has technical, entity, content, evidence, or point-in-time visibility gaps before assuming analytics is the problem.
ChatGPT referral traffic FAQ
Can GA4 track traffic from ChatGPT?
Yes. OpenAI states that ChatGPT search referral URLs automatically include utm_source=chatgpt.com. GA4 can use that source information to identify sessions arriving from ChatGPT links.
Where do I find ChatGPT traffic in GA4?
Start with Reports → Acquisition → Traffic acquisition. Change the primary dimension to Session source and filter for chatgpt.com.
Should I use Session source or First user source?
Use Session source when your question is which sessions arrived from ChatGPT. First user source represents how the user was originally acquired and answers a different attribution question.
Can GA4 tell me which ChatGPT prompt sent the user?
Not generally from standard referral attribution alone. GA4 can show the source and landing behavior, but it does not provide a complete prompt-level visibility history for the AI answer that preceded the click.
Does no ChatGPT referral traffic mean my brand is invisible in ChatGPT?
No. A user can see a brand in an AI response without clicking its website, or discover the brand in ChatGPT and return later through another channel. Referral traffic is therefore only one part of AI visibility measurement.
What metrics should I use for AI referral traffic?
Start with sessions, landing pages, engaged sessions, engagement rate, key events, and revenue where relevant. Then compare those downstream metrics with prompt-level brand mentions, competitor presence, and citation visibility.
Traffic is one signal
Measure what happens before and after the click.
Growthract combines technical AI-search analysis, prompt-level visibility signals, competitor observations, and measurable website evidence so you can understand both visibility and downstream traffic.
Source notes
This guide was checked against OpenAI's current Publishers and Developers FAQ for ChatGPT referral tagging and Google Analytics documentation covering Traffic acquisition, traffic-source dimensions, landing-page reporting, and Explorations.
Continue exploring
Related insights
Case study
E-Commerce Brand: Securing #1 Recommended Status in Gemini & Claude AI Shopping Summaries
Illustrative strategy: how an e-commerce brand could improve product data, entity clarity, third-party evidence and AI-shopping visibility to compete for recommendation-level visibility across Gemini and Claude.
See the case study →Insight
The Cost of Being Invisible to AI Search (And How to Measure It)
AI-search invisibility is difficult to value with one universal metric. Learn how to measure brand inclusion, competitor visibility, citations, referral traffic and business outcomes without inventing an AI visibility score.
Read the insight →Insight
Measuring AI Search Visibility: A Practical Framework for B2B SaaS
Stop chasing proprietary AI visibility scores. Learn how to build a rigorous, evidence-based framework to track your brand’s presence in AI-assisted search across ChatGPT, Perplexity, and Gemini.
Read the insight →