Right now it feels like everyone’s losing organic search traffic. You open up your analytics, compare it to last year, and the numbers are down.
But here’s the thing:
You’re probably not losing as much traffic as it seems. It’s just that most peoples’ analytics are not set up properly to track the AI traffic that you’re getting.
I recently set this up for a client, and not only did we find tens of thousands of visits from tools like ChatGPT, Perplexity, and others, that wasn’t being reported.
But we also saw revenue from those clicks.

And the best part is, it takes around 5 minutes to track this properly.
In this article I’ll show you:
- How to set this all up so you can see just how much traffic the AI engines are sending to you
- The small tweak you NEED to add to your analytics to make sure it reads these correctly and isn’t lost
- As well as how to measure both current and past AI search traffic
The issue with AI search
As you know, more and more people have been using tools like ChatGPT and other AI tools as pseudo search engines. They ask them questions, and the tools provide answers as well as links to sources such as websites etc.
So the big problem is this.
You are almost definitely getting traffic from AI tools. However, it’s not being recorded as organic traffic in GA4. In fact, it’s probably hidden in your referral traffic.
(Which is kind of correct because it is traffic being referred from elsewhere, which is the definition of referral traffic).
However, it’s only being referred because people are organically searching for that topic inside of those tools. So treating it as a referral means we’re not accurately measuring our ‘organic search traffic’ from all channels.
Not only that but referral traffic is notoriously poorly presented in GA and doesn’t always share the exact sources – unless of course you add filters to it which is exactly what we’re going to do now.
Step #1. Make sure we can measure the AI search traffic
Start off by opening up your Google Analytics, and then click on the Admin cogwheel in the bottom left corner. This will then open up the general admin tab.
From here you want to click on Data Display, and then channel groups.
Why?
Well, we need to create a new channel group so that we can add our filter and track these AI tools properly.
So go ahead and click on the blue ‘Create new channel group’ button.

This will then open up a new window. We can ignore most of the stuff on there for now.
We just need to click on the ‘Add new channel’ button.

This will then let us create and customize an organic AI traffic channel. So start off by giving it a name.
I recommend something like ‘AI tools list’.

This is because the filter we’re going to give it will cover more than just ChatGPT and Perplexity.
It also covers:
- AI search engines
- AI copilots & assistants
- AI writing/productivity tools
- AI model providers/research labs
- Other niche/creative AI tools
- Aggregators/portals
Pretty handy right?
Basically, if its an AI tool that scrapes the internet and can list sources and send us traffic, then we want to make sure we can measure it all.
So now it’s time to add the filter.
Go ahead and click on the ‘+ Add condition group’ text. (It’s below where you gave the channel a name).

This will load up some options.
Click on the add new condition and then select ‘Source’ from the drop down menu:

Once you’ve done that, you then need to click on the ‘add a condition’ that appears to the side of this surrounded by a dotted line.

This is basically saying that you can apply a condition to a specific traffic source.
Click on this and then choose the ‘Matches Regex’ option.

This will then allow us to add some custom regex code in the blank box underneath it.

This is where we add our filter code, and it looks like this:

Now obviously this is a fairly large script and I update it when new tools and things come on the market.
Don’t worry though! If you want to grab the latest copy of this code along with a 5 minute video walk through showing how to set it up, then click here and I’ll email it to you.
Once you have it, all you need to do is simply copy and paste it across into that box, and then hit ‘apply’.
You’ll notice that the code text bleeds outside the box but don’t worry about it. All you need to do is click the save channel button in the top right.

The channel group page will then load back up, so go ahead and give it a name and description.
Then once you’ve done that don’t go anywhere because there’s a super important step we need to take to make sure this filter works properly.
Step #2. Make sure this new tracking isn’t lost in referrals
If you scroll down the page in GA4, you’ll notice that the new channel you created is the very bottom option.

We want this to be above referral traffic so that it shows up correctly in reports.
Scroll back up and on the right hand side you’ll see a ‘reorder’ button.

Click that and you’ll notice that all the channels have these 6 grey dots next to them.
You can then click on your new channel and drag it above referrals.
Once you’ve done that scroll back up and click on the apply button.

Then go ahead and click on the ‘save channel’ button. If you miss this you’ll have to start over from scratch!

Give it a few seconds and don’t click anywhere otherwise it might not save.
You’ll know its saved when the page reloads and you see it as an option in your channel groups:

Easy right?
So now that this filter is in place, let’s set up a quick report so we can see how much traffic this is driving to our site.
Step #3. Check out our AI Search traffic!
Go ahead and click on ‘Reports’ from the sidebar menu, and then ‘Engagement’ and ‘Landing Page’.

This will then load up the basic report for all landing page traffic.

We want to create our own custom option, but the easiest way to do this is to simply customize this current report and save as a new option. So go ahead and click the ‘Customize report’ option in the top right corner that looks like a pen.
This will open up a new sidebar menu on the right hand side.
Go ahead and click on ‘Add Filter’.

This will load up another window.
Click on the box that says ‘dimension’ and then type in the name of your channel. This will then load up some different filter options available.

I like to use the ‘Session’ option as this will tell me every time someone comes across from an AI tool and not just first time visitors.
Once you’ve selected that, you then need to set the match type. Select ‘Exactly matches’ from the drop down menu.

Finally we need to select the ‘value’. So from the drop down menu, look for the AI tools channel that you made earlier:

Select that and then hit apply.
Almost done!
Click on dimensions.

Then go ahead and click on ‘session source/medium’ from the drop down menu and then hit apply.

By doing this, we’re helping the report to filter which particular AI tool drove traffic.

It also means that every time we load this up, it will show this setting.
The last thing to do now is hit the save button and select ‘save as a new report’ so that we don’t lose these changes or overwrite the current landing page report settings:

A pop up window will appear, so give it name and then hit save.

And just like that, you now have a report showing all traffic to your site from AI tools.

You can then adjust parameters to get even more insights.
For example
If you change it from ‘source/medium’ to landing page, add an page”url filter, and then search for /blog/ it will then show you specific blog posts that are getting AI search traffic.

This way you can check them out and try to figure out why these are doing well over the others.
Handy right?
But what if after all this, you’re still not seeing that many articles being recommended?
Usually this means your content is not optimized for how AI search finds and ranks content.
So go ahead and read this article next, where I break down exactly how AI search works, as well as what you need to include in your content, so that it gets recommended by all of these AI tools.
