How Many People Found Your Site With AI? AI Tools Have Your Answer

Table of Contents

• Introduction
• AI Traffic vs. AI Crawlers vs. AI Visibility: What Actually Counts
• Why Doesn’t Google Analytics Show AI Traffic in One Report?
• How to Check AI Traffic in GA4 in Under a Minute
• How to Build a Custom “AI Tools” Channel in GA4
• Is GA4’s Native “AI Assistant” Channel Enough on Its Own?
• What Should You Do With Your AI Traffic Data?
• How Does Direct Online Marketing Improve AI Search Traffic?
• Key Takeaways

If you work in marketing, odds are somebody on your team has already asked this question. Maybe it was your boss, maybe it was even you at 11 PM, squinting at a traffic report. The question is: “How much of our website traffic is coming from AI sources?” The honest answer most people give is a shrug, or a mumbled “I’m not sure.”

Marketers can see the signs of AI-driven traffic, but the reportable data can be slippery. The good news is that the number already lives in your analytics. No new subscription, no fancy dashboard, no extra budget, and no begging your dev team for a favor required. A handful of clicks and one simple setup, and you can watch your AI numbers climb week over week. Let’s find that number so you can start to leverage AI data like a professional digital marketing analyst.

AI Traffic vs. AI Crawlers vs. AI Visibility: What Actually Counts

Before we quantify anything, let’s pin down exactly what we’re looking for. AI traffic means a person asked ChatGPT, Gemini, Claude, or Perplexity a question, spotted your page cited in the answer, and clicked through to your site. That click becomes a session in your analytics, the same as any other visitor.

Two red herrings can easily get tangled up with the measurable human traffic you’re after, so let’s name them early. First, there are the crawler bots, GPTBot, ClaudeBot, etc., that scrape your pages but never appear in your reports, because they aren’t browsers loading your site. Second, there’s “AI visibility,” which tracks whether a tool cites your brand as a source in its response. A mention and a visit are two separate data points. Visibility signifies the LLM mentioned your brand, while traffic indicates a human visitor arrived on your website.

Why Doesn’t Google Analytics Show AI Traffic in One Report?

Step one is now complete; we’ve identified the metric we’re trying to isolate. Here’s where things can be frustrating. Google Analytics won’t hand you a tidy “AI traffic” number wrapped up with a neat little bow. Instead, it scatters those visits around the interface for you to hunt down. It files some traffic under “Referral,” mixed in with every other site that links to you. Worse, it dumps a big chunk into “Direct,” the same category as folks who typed your URL from memory.

Somewhere between 35% and 70% of AI sessions arrive with no referrer attached, usually because a mobile app stripped the data before the visitor even reached your site. Whatever number you manage to measure is a floor, not a ceiling. The real figure runs higher. Don’t lose sleep over this, though. Every site on the internet has this same blind spot, so you’re in good company.

How to Check AI Traffic in GA4 in Under a Minute

Want proof that AI is sending your site visitors before you commit time to a setup? This proof is simple to find. Start by opening GA4, then go to Reports > Acquisition > Traffic Acquisition. Find the dimension dropdown at the top of that table and switch it to Session Source/Medium. Now use the filter box above the table and type “ChatGPT.” Hit it with “perplexity” after that, then “gemini,” then “claude,” until you’ve run through the major LLMs. The data that pops up next to each source represents real people who found you through AI.

Keep in mind this number isn’t the full count. The method is a lot like peeking through the blinds rather than stepping into the room for a full headcount. Session Source/Medium only catches exact name matches, so plenty of AI visits can slip past it. This quick check is perfect for when your boss asks, “Are we getting any traffic from AI yet?” and you need an answer before your coffee goes cold.

How to Build a Custom “AI Tools” Channel in GA4

The peek-through-the-blinds move is handy in a pinch, but running it manually every week wastes your time. You need something permanent. In GA4, click the Admin gear, then find Channel Groups under Data Display. Create a new channel group, add a channel, and give it a name like “AI Tools.” Assign it one rule that scoops up all the big players at once. Match the session source against this pattern:

chatgpt.com|chat.openai.com|perplexity.ai|claude.ai|gemini.google.com|copilot.microsoft.com|meta.ai

Now save your channel group, and wait for the magic to happen. Every acquisition report you open carries a permanent “AI Tools” row, right alongside Organic, Paid & Social. You can finally compare them head-to-head and see how AI measures against the traffic channels you’ve reported on for years.

Here is the best part about GA4: custom channels apply retroactively. The moment you save your new ‘AI Tools’ channel, GA4 backfills it against your historical data. You don’t have to wait weeks to build a baseline. You can instantly look back at the past year and see exactly how your AI traffic has been trending. Just make sure you circle back every few months to update your channel rule as new AI tools enter the market.

Is GA4’s Native “AI Assistant” Channel Enough on Its Own?

For marketers who’ve been playing around this space for a while, you might be thinking, “Wait, doesn’t Google already do this for me?” The answer is, sort of. Back in May of 2026, GA4 rolled out a native “AI Assistant” channel that auto-tags recognized traffic without any setup on your end.

The problem is that it only recognizes three tools: ChatGPT, Gemini & Claude. Perplexity and Copilot visits remain buried in “Referral,” right back where we started, so the native channel undercounts them. Additionally, because Google only introduced this native channel in May 2026, it won’t show you any AI traffic history prior to its launch, which is exactly why building your own retroactive custom channel is so important.

We recommend covering all your bases and using both. Google’s channel gives you a fast read the second you log in, and your “AI Tools” channel catches the players Google ignores. Run them side by side, and you get the whole picture instead of half of it.

What Should You Do With Your AI Traffic Data?

An AI traffic number by itself is one part of a much larger story. Making it useful means uncovering the rest of the plot. Add “Landing Page” as a secondary dimension to your AI Tools view to see the exact pages these tools cite when they drive traffic back to your site. This is the most valuable piece of data in the whole exercise, because knowing which of your pages the LLM trusts tells you exactly what to focus on. Leveraging this data to enhance content to show up in more AI queries is called Generative Engine Optimization, or GEO.

The next piece is conversions. You’ve intrigued a browser enough to pick your page as the destination, but where do they go once they land on the site? Are these visitors buying, subscribing, or filling out your contact form? At the end of the day, conversions are the numbers your boss is looking for. Many marketers are surprised that AI traffic tends to run low-funnel, meaning these visitors come further along in the buying decision. So while the raw volume may look low, the quality runs much richer than your other channels. A modest line in the report can deliver outsized value to your bottom line.

How Does Direct Online Marketing Improve AI Search Traffic?

Diving headfirst into GA4 to set up tracking takes an afternoon, but career marketers know the real work starts afterward. Optimization requires you to read what the data says and shape your content so more of those AI citations point back to your site. That’s how you stay a key player in your industry while others fall behind.

This is how Direct Online Marketing helps our clients stand above the rest. We’ve been helping companies get found across AI search since it became a channel worth counting. Want a set of expert eyes on your numbers? Grab a free GEO audit, and we’ll show you where your AI traffic really comes from. Sign up for a free consultation today.

Key Takeaways

• AI traffic means a real human clicked through to your site from ChatGPT, Gemini, Claude, or Perplexity. Crawler bots and brand mentions are separate things that don’t count.
• GA4 won’t show you this number by default. It scatters those visits across “Referral” & “Direct,” so they’re easy to miss.
• Anywhere from 35% to 70% of AI sessions land with no referrer, so whatever you measure is a floor, not a ceiling.
• Want a fast gut check? Go to Reports > Acquisition > Traffic Acquisition, switch to Session Source/Medium, and filter for each AI tool by name.
• For a permanent fix, build a custom “AI Tools” channel group in GA4 with a regex that catches all the major platforms at once.
• Custom channels don’t backfill, so build yours today. Any data you skip now is gone for good.
• Google’s native “AI Assistant” channel only recognizes ChatGPT, Gemini & Claude. Run it alongside your custom channel for full coverage.
• Add “Landing Page” as a secondary dimension to see which of your pages the AI tools actually cite.
• AI traffic often runs low-funnel, so the volume can look small while conversions run much higher than your other channels.