What does AI traffic analytics answer?
It answers the question every board asks sooner or later. Is AI search sending us any visitors, and to which pages? Visibility in AI answers matters, but most businesses also want to see visits and what those visits do.
The term AI traffic analytics refers to pulling AI referral visits out of Google Analytics 4 and Search Console and setting them next to the rest of your search data.
GA4 holds the answer, but it is not easy to get at. AI visits are mixed in with every other referral, spread across several domains. This feature pulls them out and puts them next to the rest of your search data.
How does it work?
Setup takes a few minutes and runs daily after that.
- Connect GA4 and Search Console. You sign in with Google and grant read-only access. The platform cannot change anything in your accounts.
- Daily sync. Sessions from GA4 and clicks from Search Console are synced every day.
- Split out AI referrals. Visits from the main AI engines are grouped as AI referral traffic.
- Join by page. A page level funnel lines up Search Console clicks with GA4 sessions for each page.
- Add AI Mode data. Upload the Search Console Generative AI performance report so its daily impressions build up over time.
Which AI sources does it pick out?
It groups visits by the referring domain. The table shows the main ones.
| AI engine | Referring domain |
|---|---|
| ChatGPT | chatgpt.com |
| Perplexity | perplexity.ai |
| Google Gemini | gemini.google.com |
| Claude | claude.ai |
| Microsoft Copilot | Copilot domains |
Other AI sources are picked up too. The glossary entry on AI referral traffic explains how referrals work in more detail.
What is AI traffic attribution?
AI traffic attribution means working out how many visits, enquiries and sales came from AI tools such as ChatGPT, Perplexity, Gemini, Claude and Copilot. Axiom GEO does it from your own GA4 data, splitting AI referrals out from the rest of your traffic.
What do you see?
You see AI referral traffic over time and the pages that receive it. Each AI source shows its sessions, users, conversions and revenue from GA4, with its share of your organic traffic and a comparison with the period before. That tells you whether the AI visits turn into enquiries or sales, not only how many there are. The page level funnel then shows each page with its Search Console clicks and GA4 sessions side by side. That makes it easier to spot pages that do well in Google but get little from AI, or the other way round.
It helps to look at landing pages first. The pages AI engines send people to are often not the ones you expect. A detailed guide or a pricing page can pull more AI visits than the home page.
The same screen has tabs for source and medium, landing pages, Search Console queries, keyword to page matching and paid traffic. The paid traffic tab reads Google Ads, when it is connected, so AI, organic and paid visits sit side by side.
What are the limits?
We would rather be clear about these up front.
- Some visits from AI apps arrive with no referrer and show up as direct traffic, so the AI figure is a floor.
- Clicks from Google AI Overviews count as ordinary Google search traffic and are not split out.
- Google's API does not expose AI Mode, which is why the report upload is there.
- Many AI answers get no click at all, much like zero-click searches.
That last point matters most. Low AI traffic does not mean low AI visibility. It is worth tracking both, which is why this sits next to AI visibility tracking.
How should you read the numbers?
It helps to compare AI referrals with the same period last month rather than day by day. The numbers are often small at first, so single days jump about. Look at which pages gain AI visits after you improve them, and whether those visitors go on to enquire or buy. That tells you far more than the headline total.
Why does it matter for AI search?
Visits are the part of AI search that shows up in revenue reports. Being able to show them, page by page, turns AI search from a theory into something a finance director can see. According to Pew Research Center, people clicked a normal search result in just 8% of visits when a Google AI summary appeared, against 15% when none did. That is even more reason to watch the visits AI search still sends you.
It also closes the loop on work. When a task in Work is linked to a page, the platform records a baseline and later measures whether clicks rose. The same data feeds content opportunities, such as low click-through pages and pages losing clicks over the last 28 days against the 28 before.
For a step by step look at the GA4 side, our guide to AI traffic in GA4 covers the reports and filters in detail.
Who uses it?
In-house teams use it to report AI traffic to leadership without building custom GA4 reports. Agencies use it in monthly client reports, so clients see visits as well as visibility. B2B and SaaS teams use the page view to see which guides and product pages AI engines send people to.
Where it fits
AI traffic analytics is the visits side of AI search. AI visibility tracking shows how often and how well engines mention you. Google AI Overviews covers the overviews and AI Mode. AI crawler analytics shows which AI bots read your pages before any visitor arrives.