AI Traffic Is Already Here, But You’re Probably Not Tracking It

There’s a new kind of traffic hitting your site. It’s not coming from Google Ads. Not Meta. Not affiliates. It’s coming from AI-powered tools like:

• ChatGPT (with browsing or plugins) 
• Perplexity 
• Claude 
• You.com 
• Other generative search assistants

These LLMs (Large Language Models) are becoming discovery engines in their own right. Nowadays, users don’t start with Google Search or social browsing; they ask AI agents for financial advice, product recommendations, side-by-side comparisons, health tips, wellness guidance, and more. AI tools funnel users to websites via means that are not traditional search, paid ads, or affiliate links.

But here’s the issue:

• Most analytics setups aren’t catching this. 
• GA4 is quietly shoving this traffic into “Direct” or “Unassigned.” 

That means you could be winning new customers… and never even know it.

ChatGPT is sending you traffic. But GA4 isn’t telling you. Here’s how to fix it. 

Why Most Analytics Miss This Traffic

Most marketing teams, especially in high-intent or regulated sectors like healthcare, insurance, supplements, and wellness, rely on Google Analytics 4 to understand where traffic comes from. 

The problem is that GA4 isn’t designed to recognise AI referrals as a separate category. Instead:

Referrer data is often stripped or lost.

Some AI tools, or their front-ends (browsers, mobile apps, plugins), don’t send proper referrer headers when links are clicked. This buries visits under “Direct” or “Unassigned”.

Domain or source names are ambiguous.

Even when data is passed, the sources are often not able to clearly identify the AI tool (for example, chat.openai.com or perplexity.ai). These names don’t map to GA4’s default categories.

Default channel groups ignore LLMs.

GA4’s standard channels (Organic Search, Paid Search, Direct, Social, Affiliates, etc.) assume traffic from traditional sources. AI referrals don’t fit into these categories, so they’re grouped in “Referral,” “Direct,” or other catch-all buckets. 

You may already be getting customers (or at least visits) via AI referrals, but your dashboards don’t show it. Which means you can’t understand, optimise, or compare them with other channels.

Why This Matters More Than You Think

AI-driven referrals grew more than tenfold in the U.S. from July 2024 to February 2025. The conversion gap between AI traffic and non-AI traffic is shrinking. 

Failing to track this traffic impacts:

• Attribution & Revenue: You can’t assign conversions or revenue back to AI-driven sessions. Budget, content, and SEO strategy will be skewed.‍
• Content Strategy, SEO & AEO: AI assistants surface useful, clear, and structured content. Optimise only for traditional search engine signals, and you lose out on AI discovery.‍
• Competitive Edge: Brands that measure and optimise early for AI referrals will likely get more traffic, better engagement, and more conversions.‍
• User Behaviour Differences: AI users may come in with a different intent (they’ve already made comparisons). Their session behaviour, bounce rate, pages per session, etc., might differ.‍
• Privacy, Compliance & Trust: As AI tools evolve, how they pass data (referrer, UTM tags, etc.) may be constrained by privacy policies, or app vs browser behaviour. 

How to Track AI Traffic in GA4 (It Takes 5 Minutes)

The good news: you can set this up quickly. Here’s how:

• Go to Admin → Channel Groups in GA4.
• Duplicate the default channel group so you don’t lose the original setup.
• Create a new channel called something like “LLMs” or “AI Assistants.”
• Add traffic source conditions, for example:
  
  • Source contains chat.openai
  • Source contains perplexity
  • Source contains claude
  • (Or if you’re using UTMs, add a rule: Medium contains LLM)
• Save your new group and set it as the Primary Channel Group.

From now on, when AI traffic reaches your site, it will show up separately, right alongside Organic, Paid, Social, Direct, and Referral.

What You Can Track (And What You Can’t)

What’s Possible

• Traffic from AI tools that preserve referrer headers (e.g., chat.openai.com, perplexity.ai, claude.ai, gemini.google.com, copilot.microsoft.com).
• Identifying traffic using regex or filters in GA4/Looker Studio to match known AI domains.
• Creating segments or custom dimensions to mark sessions with AI sources.
• Building a dedicated “AI/LLM” channel to keep them separate from default channels.

What’s Tough

• Tools that strip referrer headers: those visits still show as “Direct” or “Unassigned.”
• Even if you know the source, you don’t know which prompt or query drove the click.
• Embedded/app-based environments may limit cookie or script tracking.
• New AI tools keep emerging, and existing ones change behavior. Tracking rules must be flexible and updated regularly.

Beyond Tracking: Best Practices For AI Discovery

Tracking is only step one. The next step is to prepare your site to perform well:

Answer Engine Optimisation (AEO)

• Structure content with clear answers, FAQs, subheadings, and schema markup.
• Keep your content concise but add rich snippets. 

Strong Content Quality & Trust Signals

• Back up claims with data and sources. Add trust markers. In regulated verticals, credentials matter. 

Speed, Clarity, Readability‍

• Users coming via AI referrals may have less patience. The content should load quickly, get to the point and avoid heavy clutter.

UX For Deeper Journeys‍

• AI users may skip traditional browsing steps. Optimise landing pages for clarity, conversions, and next steps.

Evergreen Capability To Spot New AI Tools‍

• Continuously update your list of sources, new AI tools, domains, and referrer behaviours to capture.

What Advertisers Can Do Now: Checklist

Here’s a short checklist to get started. You can get started in under 30 minutes:

• List known AI/LLM referral domains you suspect are sending traffic.
• In GA4, run filters in Traffic Acquisition to spot them.
• Create a new Channel Group for AI/LLMs.
• Build an Exploration report to isolate AI sessions and events.
• Compare AI traffic behavior with Organic, Direct, and Referral.
• Identify which content is surfaced most often via AI tools.
• Optimise content to be more “AI-friendly” (clear, structured, credible).

What’s The Future Of AI Traffic?

AI traffic isn’t coming. It’s already here. Generative search and LLMs are fast becoming discovery engines and linking content in ways that bypass traditional referral paths. 

If you don’t measure it, you can’t optimise for it and you’ll lose opportunities to competitors who do. Taking a few steps in GA4 and optimising content for AI discovery will pay off, often more than you expect.

Need a fresh perspective? Let’s talk.

At 360 OM, we specialise in helping businesses take their marketing efforts to the next level. Our team stays on top of industry trends, uses data-informed decisions to maximise your ROI, and provides full transparency through comprehensive reports.