So What? How to Measure AI Visibility

Well, hey everyone, happy Thursday. Welcome to So What, the marketing analytics and insides live show. I'm joined by Chris and John and my dog barking from the other room because it's hot outside and she is bored. How's it going, fellas? Yeah, trying to speak.

It's hot. Yeah. Uh we're all located in New England. Um, I know it's hot all over pretty much everywhere right now. So that aside, we want to do something productive.

So today we are talking about how to measure AI visibility. This is a hot topic. There's a lot of unhelpful resources, and we wanted to provide you with a general framework of the different data that you can look at in order to understand your AI visibility. So on this episode, we're gonna walk through what you could be doing. But as always, if you have questions or you want to learn more, um, you can reach out and contact us at Trust Insights.ai slash contact if we get the correct banner.

I'm gonna blame the heat. Let's start with what you can't measure. You can't measure what people are typing into chat bots, whether it's AI mode, AI overviews, chat GPT. And anyone who says they can measure that is lying. They are lying that you red flag.

Here we are. We're only what two minutes in. You're welcome. And you can't even necessarily measure what's in the model itself. I know folks have been sharing this one tool that supposedly somebody put together called in the weights, which is when you read about the methodology is completely wrong.

And not to go off on too much of a rant here, but today's AI models are all what are called mixture of experts' models. They are made to have huge amounts of data and then internal miniature versions of themselves that are called experts that and we don't know what they are, neither do the model makers. But every time you ask, what's the best AI consulting company in the Metro Boston area? If you've got other things in the chat that the model interprets, it will route it through different experts. So let's say, as we've talked about with things like AI mode and AI overviews that Garrett Sussman at uh iPool Rank talked about recently.

Let's say in your Gmail inbox, you just got a promotional uh set of emails from Patagonia. That context gets loaded with your search term, and that will invoke different parts of the model when it answers. So you can't even say with definitiveness that you know, trust insights has this much relative brand strength in in this AI model because different portions of it activate at different times. The best analogy I can come up with is imagine you're like at a at a college dining hall. Like if you've set foot in one recently, they're all really nice now, and they all have like you know 82 different food stations.

Um and you start off by saying, I'm hungry, I want tacos, and the the person routes you to the taco station, right? That's one of the experts. But suppose you said, I like Korean food, but I want tacos. They'll route you to the Korean food station first before getting you to the taco station. And they and it might be Korean inspired tacos at the Korean station, in which case you'll never get to the taco station.

And so if you have an AI tool or an AI company saying, Oh, we know how you rank in the taco station, you're like, Yeah, but I'm at the Korean station. That doesn't help me. So all this to say you cannot know what's inside the model itself with any level of confidence, and you cannot trust or believe anybody who tells you here's what your how your brand ranks in the model, because there's so much distortion with the way these models all function. I mean, now I'm hungry for Korean-inspired tacos, which sounds delicious. There's a and there's a chain called Koreanos here in Metro West that does exactly that.

They have like bulgogee tacos and kimchi tacos, it's amazing. Yes, please. Um, but no, but I think it's important. I think that that's a good sort of level set for people of like what you can't do, because there's a lot of misinformation, and I think there's a lot of you know, analysts who are sort of struggling because their leadership is like, you know, it's sort of the old, why aren't we ranking for number one in the search results? Like, because that's not that's not what we have control over.

Like, that's not what we're looking at. And so I think one of the biggest pain points for analysts is trying to explain a large language model is never going to give you the same kind of data that you get from like a Google search console in terms of this is what somebody typed and this is how they found us because it's just it does not work that way anymore. Um but okay, so what can we do? So we can know three things. One, we can know um how AI bots are hitting our sites and resources, so we can know essentially what there's they're looking at from us.

That is a knowable thing. Two, depending on the system, we can know what citation queries or grounding queries, because almost every AI tool does its own searches, and for some tools, we can see what they essentially went searching for. And number three, which is the gold standard, we can know what pe how people found us when they tell us when we ask them, how did you find out about us? And that is the the gold standard. And if you are if you calibrate, if you take away nothing else for this entire episode, please ask people how'd you hear about us.

And I feel like that harkens to why people like John exist, because if you don't have a John on your team, you're missing a lot of great opportunities. So for those of you who don't know, John is our head of business development, he's one of our partners, and he's the one who's talking to prospects all day long. And one of the things, John that you do is you're like, Hey, how did you hear about us? Where did you find us? Oh, did you see Chris?

Oh, did you see Katie? And one of your core responsibilities in those input calls is to find out how people found out about us, and like that to me seems like table stakes. Like, that's just a given. That's what you should be doing. Like, do you have any thoughts on that about like why people maybe don't do that?

It's really, you know, it's it's a trust thing, right? I mean, because you will have their last touch attribution. You'll have something in your system, like, hey, they showed up in HubSpot because they downloaded this or that. But then it's do you have the permission to dig deeper and find out? Because it's always, you know, nobody just landed on the reg page for the white paper, right?

Somebody told them about it or they were looking for something. And then for us too, it's insane. I mean, we'll keep pulling and it'll be like, well, they worked with a guy 25 years ago at this company and met us at some weird podcast event, you know, like it can be anything. And so you really just kind of have to go with the best that you can do. You know, you again, you over time get enough permission to get to the truth.

And prior to that, it's just kind of start with the data that comes in automatically. And if you have enough um trust earn to make a better guess, that's great. But did you really kind of have to just play it by ear and see where it goes? Yeah. I mean, it's all right.

So, Chris, one of the basics, ask people, how'd you hear about us? And if you don't have a John on your team, you should at the very least have it on your contact page as they're filling it out and they're trying to download all of your IP so they can rip you off. You should at least be able to ask them, hey, how'd you hear about us? Yes, exactly. All right, so let's work our way up the chain, because that's the most important.

Ask people. Yeah, and and use AI, right? If you get a lot of contact form submissions, just literally use AI talk and say, you know, what in this response in this field. And Katie, you've shown this and you show this in workshops and things. In fact, you're doing a workshop at the marketing uh AI conference on, among other things, clawed in Word, clawed in Excel, clawed in PowerPoint, et cetera.

And you can have Claude and Excel just write a formula saying, hey, uh, was AI mentioned in this form field response? Um that's level one, the most important level, the foundation level, ask people. Number two, we all have or should have good experience with what the major search companies provide us for data. And there's two tools in this space Google Search Console and Bing Webmaster Tools. Let's start with Search Console.

Um, it is slowly rolling out, but what Search Console currently gives you is gives you your general search results. In some accounts, what you are now seeing is in addition to that, there's now a generative AI sub tab in there that gives you information about uh the the inquiries from generative AI. So this is your regular one that you that everybody has, and then you have just this one, which is generative AI. This goes back to about mid-March of 2026. It does not go back earlier, and you only know Google will only tell you what pages were recommended, uh, what countries, what devices, and what days, and then what impressions you get.

You got get no click data at all, and you get no query data, but you can at least see okay, for our site, it showed up in 4.13 million impressions, and these are the pages that got the impressions. That's what Google is currently giving us as of July of 2026. They might give us more at some point, but this is what you get right now. And this is from a client's uh page, so I can't show any of the the grounding data. What's useful to know here though is that that 4.1 million is a subset of the total 152 million impressions.

So you can also get a sense of how much of our search data is actually AI versus regular Google. And in this case, it's still pretty small. It's still like one 25th or something along those lines. It's it's not a lot for this particular site. But that's a helpful, even if it's small, that's helpful for when the conversation comes up of why is our search traffic declining?

Oh, it must be AI. It's like, nope, that's actually not what's happening. Like, you know, I'm not saying for this particular client, that's what's happening, but like that's a conversation that's happening across a lot of organizations is what am I do about? How do I fix my organic search? And you know, and we know, and we've talked about in multiple instances that you know, you still have to have the basics of SEO, which we cover in our GEO 101 course at TrustInsights.ai slash GEO101.

And the core tenets of SEO still exist. AI visibility is now just another part of it. It doesn't replace it. And so if you're seeing your search traffic go down and you're blaming AI, you're looking in the wrong place. I will point out for this particular website, uh, and I've seen this in a couple of our different clients whose search console accounts we manage, uh, that purple line for generative AI impressions is going up and to the right.

Like it is growing and is growing fast. It's growing, but it is not, it is not the scapegoat that people are trying to make it out to be, I guess, is my point. Is yes, it's growing, but it, you know, generative AI has now been around for what two, two and a half years in terms of like people looking at a large language model to find out information. What's been happening to your organic search in that time? Like you can't just say, oh, it's AI.

Like you can't blame it. So that's my point is like you need to have been doing all this stuff all along. Exactly. The other tool that you should be using is Bing Webmaster Tools. Now, a lot of us like to poke fun at Bing, right?

Even though it powers stuff like Yahoo. But what's really important about Bing, the Webmaster Tools is this simple thing. This is Microsoft Copilot grounding data. And Copilot is the largest installed AI base in enterprise companies. So if your audience works in enterprise companies, so B2B marketers, you should be a hundred percent all in on Bing Webmaster Tools, because that's your that's your if enterprise is your jam, that's your jam.

Even mid-market companies, you know, copilot is the approved AI. But also if you serve people who work at those companies, whether you're B2B or B2C, you should be in here looking at to see what you get. Now go ahead, Katie. No, I you know, I feel like Bing is the unsung hero. Like I remember when, you know, we were working with clients on getting Google Analytics Universal Analytics installed correctly and making sure that they had Google Search Console.

And, you know, there is, well, do I need Bing Webmaster Tools? We're like, it's not a bad idea to have it, but nobody ever seemed to pay attention to it. And now, haha, joke's on them. Um, you know, we've covered in other uh shows, podcasts, why Microsoft Copilot is the system of choice for enterprise uh companies. But the short version is because it's highly regulated, and a lot of companies have already bought into the Microsoft ecosystem.

And so it's just a natural fit, you know. You know, and so if you're looking to use something other than copilot, that's a conversation with your IT team. Exactly. So what do we get with Bing? Uh, as with Google, we get the list of pages, and you get to see how many citations, how many results was this page cited in?

So, for example, our top page on our blog is the post on why AI can't do math from about almost a year ago, actually. Uh, we have other pages, uh, like our some of our white papers. We have some uh some actually a decent number of PDFs, you could tell because it's the WP content thing. So PDFs get some love in and show up in the citations. However, what you get out of Bing that you don't get out of Google is the ground in queries tab.

And this is pretty darn awesome. This tells you what Bing searched for. I refuse to say use Bing as a verb. Uh what Bing searched for, what the intent was of the search, what topic it is, and then how many times your resource your site as a whole was cited for that topic, and what percentage of the citations were yours. So LinkedIn candidate ranking algorithm, informational intent, recruitment and talent acquisition 141 times.

We have been cited, and we are 66% of the citations for when when Bing searches for this particular topic inside of a tool like Copilot and returns results. We copilot says trust insights, 66% of the results are ours. And what though, there's no visibility into which large language models they're hitting, or do they have a short list of where they're pulling that stuff from? Well, so Copilot has four models under the hood. Copilot either has Copil uh Microsoft fee, uh, which is their their autofast one, they have a uh chain of thought version of fee, which is less dumb but still dumb.

Uh, and then they have OpenAI's GPT models, 5.5 and 5.2 are available as the as the models in there. But what happens in Copilot is you ask Copilot something, Copilot says, I'm going to do a web search and goes to Bing and uses Bing, pulls the data, and the and the data that comes back, copilot then interprets. Oh, I see. So even though it's Bing, it is co these are still copilot queries. Yes.

I just had to do a quick search because I need to validate that I was like, wait, isn't LinkedIn part of the Microsoft ecosystem as well? It is. So it, you know, it's interesting because we talk about the Google ecosystem, including things like YouTube and uh other you know major platforms. And now we're seeing the other side of it of now it's the Microsoft, Microsoft's turn to like show off their whole ecosystem. And so I don't know that there's any sort of correlation between LinkedIn showing up as the number one thing and the fact that LinkedIn is part of the Microsoft ecosystem.

But I also can't say that there isn't any correlation. But I think it's you know something for marketers to pay attention to is understanding which platforms belong to which of the big guys. Oh, for sure. In this case, these two queries are they're specifically asked people, you know, Bing and and Copilot. Someone's having conversations about how does LinkedIn's job candidate ranking work, right?

So the HRIS stuff for those who are using uh LinkedIn talent acquisition services. This, by the way, tells us uh, because what it's linking to is the linkedIn unofficial LinkedIn algorithm guide, which is the Trust Insights publication we have about how does LinkedIn itself work. Uh, we've never done a piece specifically on the talent acquisition part. We probably could, and we might want to if it made sense from a uh, you know, it is going to get us any business uh perspective. But there is a whole bunch of papers that LinkedIn has on how their internal ranking systems work for when they're recommending you for job you know searches and things like that.

But that's what people are asking them for there. That makes sense. I mean, it's a big topic right now. It's a hot it's a huge topic. The second one there on the list, co-work plugins.

Well, as of what, two weeks ago, Microsoft Co-Pilot went general availability for Microsoft Co-Pilot Co-Work, which is their licensed version of Claude Co-work. And so people are looking for stuff around that. And Katie, this will be an opportunity for Trust Insights to say, okay, let's make sure that we're promoting our various plugins that we have in our academy as co-work plugins, in addition to just saying they're clawed plugins, because obviously Microsoft co-pilot co-work, because it's a licensed version of Cluck and it should be cross-functional. So we would want to maybe have some content about that. No, and that makes sense.

And you know, for a long time, Microsoft Copilot was sort of, you know, they had that firewall up of like, unless you're a Microsoft user, like you can't know what's inside Copilot. But now that Claude co-work is part of it, it's like, oh, well, we are well versed in Claude co-work. We've taught classes and built instructional materials on Coworks. So, oh, and by the way, we also have things that you can buy to just install into your co-work. So, yeah, we absolutely, I'm gonna write that down somewhere that I'm gonna forget about it.

But like, that's a really great idea. Yep. And finally, we see our years-long addiction to making frameworks for everything actually paying off because the majority of these things are like AI frameworks for marketing consultant, prison framework for prompt writing, marketing consultancy AI frameworks, and so on and so forth. So uh we are winning that in fact, even right there at near the bottom is the five feet framework by just in size. Well, they didn't put the brand name, but the 5P framework is there.

So Microsoft Copilot is citing us when people ask about this this particular framework. And it's great that people are looking for it by name. Now, what we would want to do is we would want to also take a look at our regular search performance to see how does that compare to the what co-pilot is doing. And we see some different stuff here. Um, we see our brand name, we see some YouTube collaboration stuff, we see um get insights on marketing AI agents that look real.

Uh we see a lot of more YouTube-based stuff, people using regular Bing, which is not the co-pilot stuff, regular Bing for this. Search engines like Yahoo, for example, use Bing as their back end. Uh Safari for the longest time used it as its back end. But even here, something like Trust Insights Academy, like people are searching for us by name, so that's a great thing. So that's level two.

Level one is ask people. Level two is take a look at uh Google Search Console, especially if you get the generative AI tab, and take a look at Microsoft uh Bing Webmaster Tools. You know, when we million years ago were talking about Bing Webmaster Tools, it was something that we had to install and set up for our website. I assume that's still true. That's still a step that people have to take.

You can't just open up Bing Webmaster Tools and expect your data to be there. That's correct. Just want to make that clarification. So if you've never gone through, so if you've done the steps to install or integrate Google Search Console, it's roughly, I think, probably the same steps to install Bing Webmaster Tools, but this is a great time to do it, especially as the questions around AI visibility are becoming more and more prevalent in conversations. People are doing their 2027 planning already.

They're gonna want to know. And this is a great data source to be looking at if you don't have it. Yep. Now, the thing, the the third layer is if you have a CDN, a content delivery network that collects the data about what AI agents are hitting your website, you can potentially get at this data. So we talked about, and this is actually in the GEO uh one-on-one course, uh, Cloudflare is one of those CDNs that gives you seven days worth of data about which uh AI bots are hitting your website.

Well, the the challenge here is it kind of sucks, like uh, you know, getting every seven days and downloading a bunch of CSVs. Clearly, there's got to be a better way to do this. And the answer is yes, there is. You can, and I did this last night, set up a Cloudflare worker, which is a little miniature server on their system that intercepts and makes a copy of all the bot data and can send it to other places. And so what I wired up last night for fun, because this is what I do on Thursday nights, is I wired that worker into Google Analytics.

And what we can now see in Google Analytics is I can now see all the Cloudflare data to understand what kinds of AI bots are coming to the Trust Insights website. Are they crawlers, which are looking to vacuum up training data? Are they assistants, the browse on behalf? Like when you're using Gemini or Chat GPT, and you're like, hey, what's a good marketing AI consultancy? And it says, let me go look for that.

And then AI search, uh, search engines like AI mode and AI overviews uh specifically. So those are the three categories. You can see what pages they hit, and of course, you can see uh the event counts. So to clarify even further, so let's say I'm using Claude, and I'm using the Google Chrome extension to, you know, and I say, hey, go pull the data off of this website because I'm too lazy to do it myself. Is that website's Google Analytics going to categorize it as an AI assistant?

In this instance, if if it's Claude bot, yes, Cloudflare will categorize that and send that data. Here's the catch Google Analytics does not do this out of the box at all. In fact, Google Analytics out of the box blocks all bots because they don't want your analytics being filled with junk. So what I had to do was wire this in this in a way that bypasses Google's bot blocker by using what's called the Google measurement Pro uh protocol, uh which is part of their API, and set up a separate Google Analytics property just for this, because you absolutely do not want bots mixing with regular human traffic. But to your question, Katie, you can tell by the user agent of the bot what bot it is.

So if I start a new tab here just in free form and we put in our user agent as one, and I put in our event count as the other, um, we can you can see based on the name. So this is Amazon bot there. There's meta external agent, there is chat GPT user bot. Um is ClaudeBot is that the Chrome extension, or is that something else? That is that is principally going to be Claude's web interface, claude desktop, and claude code.

Got it. So basically, when you say, hey, go find this stuff, and you're using the Claude desktop essentially as a search engine, and then saying this is what I want to search, you go search it. That's what we're talking about when we say like a Claude bot. So if you go into Chat GPT and say, This is the deep research that I want to do, you go find it, then that's where you get the GPT bot, for example. That and I think that that's data that people are craving.

Mm-hmm. Yep. Yeah, you can get all of that data. One of the things I found interesting was that Amazon's crawler bot was the most aggressive of them, uh, which I was like, okay, that's that's interesting. So I put bot category here and then put user agent.

And now we can see crawler assistant, crawler search, et cetera. So we can start to get at what are those different uh bot types. So clawed bot is that is the assistant version of Claude. So when so with Amazon, like what they don't have like a large language model like a chat GPT or like a Google Gemini or like an anthropic cloud. So like what is Amazon's version that users would be using in order to hit our website?

They do. Um when you use Claude or Gemini or any of the models through AWS, um, because you're an enterprise company using Amazon Foundry and AWS, all of that data comes through Amazon bot. So it's behind Amazon's AWS. So that that Amazon bot there is a big old Podge Podge of all the enterprise companies that use AWS for protected models. So for example, we have one client in the healthcare space, they have their own internal AI hub, and that routes through in in their case, Google Cloud.

Um, so all so you would see a Google stuff coming out, even if they're using Claude, they're using a version of Claude that lives on Google Cloud. In this case, you'd be using a version of Claude that lives on AWS. I mean, this is just getting more and more messy. Mm-hmm. It is.

Um, but what's nice about this is if you once you set this up, and it took me a good hour and a half to figure out all the the weird uh tricks to get it working. Um, you can then start to do things like build look data studio dashboards. So as time goes on, this data studio dashboard will populate for us, and Katie will be able to go in and see like how much AI assistant versus search versus crawler traffic are we getting? What are the top pages on our site that they're hitting? And if Katie wanted to, Katie could say, I only care about uh AI search.

So I'm gonna turn off uh oops. It makes it so easy, doesn't it? Welcome to my personal help. And these are the pages that matter most for AI search, right? When someone is look using uh something like Google, or I wanna see, you know, because we know Claude is classified as an assistant, what are the pages that are here?

And we could even put in a drop-down saying if I wanted to specify Claude versus ChatGPT or whatever, Katie could slice and dice this without having to do it in Explore Hub, which is kind of a pain in the butt, or look at raw server logs, because Katie doesn't like looking at raw server logs. And the reason what no, I really don't. I've done it. I don't like, I don't care for it at all. The reason why I would want to look at this information, the so what of it is so that I could look at those particular pages and say, is there enough machine readable content on it so that you know if someone's searching and the machines are saying, Let me go find the most helpful information there is, they're finding us.

So that's where you know we've been talking about for well over a year. You have to have that content for the human as well as the content for the machine. So your pages need to get longer, they have to have more information. Again, in the GEO 101 course, we cover being everywhere and what that means. And so, you know, if you don't have those transcripts included, you know, with your podcasts, with your videos, if you're not putting the descriptions into your YouTube uh channel, into your YouTube videos, like those are missed opportunities because now we're seeing we're being found by these AI assistants, these AI crawlers, because we've been putting in the work to make sure that there is machine readable content.

Uh, I remember, Chris, when you first started putting that onto our website, and it was like, this content is for specifically for, I was like, what is this? Why is this, you know, jacking up the UI on the website we've worked so hard on? But now I'm like, no, I totally get it, because this is the kind of information you're gonna get back. And then I can look at and say, okay, where is there missed opportunities? Where do we need to do more?

Because yes, there's a human behind the large language model searching, but now you have this intermediate step of a large language model. So you're no longer getting sessions from humans, you're getting AI assistance powered by humans, powered by large language models. It's like I'm trying not to overcomplicate it, but basically now instead of a human to your website, you have this intermediate step of human to large language model to your website, and that's what we're looking at. Yep. And it's uh so if you remember a couple of weeks ago we talked about Web MCP.

One of the things we didn't mention in that episode is that Web MCP imperative API, which is the JavaScript version, one of the things that you should consider doing, um, and we built it into our WordPress plugin is sending Google measurement protocol hits to an AI property on your Google Analytics so that you can see it A, is anyone using Web MCP? And if they are, then um, you know, what is the use? But I want to show this because I think this is fascinating. So if I go to our the regular trust insights, Google Analytics Property, and I go into our real-time overview. Uh, this is right now, it's relatively quiet on our website, right?

This is a regular GA4 property. This is fine. It's you know, it is what it is because it's the middle of the day, it's a holiday week. If I then flip over to the Cloudflare bot page, look how many bots are on our website right this minute. And I think that in and of itself makes the case for why you would want to have a separate Google Analytics property for your Cloudflare for the bot specifically, because this tells a very different story.

That you know, I along with Kelsey, look at the data for our website month over month. And this is what we've been struggling with, like how much of that is bot traffic and how much of it is real. And now we can actually differentiate uh note for Kelsey. You're gonna have to look at a different GA4 property for the monthly metrics now. Yep, exactly.

So this is the this is level three of AI visibility, which is so level one, ask people. Level two, look at the approved tools like Google Search Console and Bing Webmaster Tools to say, you know, what are those things to doing? Level three, start looking at your CDN data and put it in your analytics tool of choice so that you can see, okay, here's how the bots are coming to us, what kind of bots they are. Um if you use Akamai, uh Akamai has a similar thing to Cloudflare. It's called Edge Workers, and you'll have to program and build that for Akamai.

If you're using Cloudflare, you're using Cloudflare workers and worker routes to do this. Uh and if you're interested in getting some help with uh this, because I can tell you right up the top, it is is a massive hate to set it up. Uh, you can go to a trusted science.aslash AI bot analytics, and that will connect you with John Wall, who will help you uh talk through how we can get this uh set up for you. So that's level three. And I want to acknowledge, Chris, because you said, oh, this took me about an hour last night.

I want to sort of put into sort of like relative terms. Because you're so well versed in setting things up in different technology systems and infrastructures, and especially Google Analytics 4, you're very efficient with Google Analytics. When you're saying it takes someone like you, who really has that deep level of understanding an hour, we're talking someone like me, probably a good couple of days to like wrap my head around all of it. And then John, who is like, get me away from setting up Google Analytics for, he's like, I'm just gonna outsource it. Where do I find the John Wall version of this?

Um, so I just I mean, like, I just want to like put that in perspective. Like when you're saying it took you an hour, that's you, who is like a super advanced user. Us regular folks, it's gonna take us a while, probably first and foremost, to say, well, what CDN do we even have? And where's the login? Yes.

And do you have permission to use it? Which is one of the things that uh, you know, a lot of companies have IT departments that have it that's separated. Um, the other thing that I would strongly recommend if you do this with Google Analytics, is to tie it in your Google Analytics to your BigQuery instance to Google Cloud BigQuery. Because remember that depending on your Google Analytics settings, you may either have data retention at the event level for either two or 14 months, and then that data gets deleted, it goes away. If you tie it into a big query instance, then you get that you keep that data in perpetuity, but you pay for it.

Uh for us, it's gonna cost us probably about five to ten dollars a month in usage cost, which is not like outrageous. Um, but also the other thing that it will give you inside BigQuery that you can't do in Google Analytics, is you can use BigQuery's native AI with Gemini to do much more advanced modeling and stuff. Again, it will cost you money because everything you do with BigQuery costs money. But if you had stuff that you specifically wanted to do for advanced data analysis, that would be the place to do it because that that database is tuned for doing very heavy machine learning loads. I recall when Google Analytics made the switch from Universal Analytics to Google Analytics 4.

BigQuery was a recommendation we were making at that time anyway. Uh, so if you're a user of Google Analytics 4, BigQuery shouldn't come as a surprise. I'm assuming, Chris, the reason we're recommending BigQuery is because it is natively connected to Google Analytics versus using some other um database or data collection system. Like, could you use something that isn't BigQuery? Um you could use something that wasn't BigQuery if you had the ability to export the data to that system.

But BigQuery will still be your intermediate stopping point to you know that stops by that, which is where you send some of that data. I bring this up because again, I feel like yes, we're promoting that we can do this for you, but there's a reason why you might want to have somebody who's familiar with all of these systems doing this because it's not just setting up a new type of report in Google Analytics. Gosh, I wish it were that simple. But you have your CDN, you have BigQuery, you have these other systems that you need to be well versed in in order to get this level of data. Yep.

So this is what it'll look like in BigQuery. Inside the BigQuery table itself, you can see like AI bot. Is it a bot? You know what kind of bot is it and stuff like that. The stuff we've been talking about is AI assistant.

This is the data that comes from Cloudflare to BigQuery. Now, here's what the reason why you might want to use BigQuery, and um as opposed to Google Analytics or as opposed to the approach Katie was talking about. In BigQuery, if you're skilled at it, you can query across tables. So I can take my AI bot table and I can cross-reference it with my regular GA4 table to see, for example, are humans and machines looking at the same thing. Right now in Google Analytics, you'd have to flip back and forth between properties, it'd be a pain in the butt.

You could all do inside one query inside BigQuery and do the advanced data analysis there. And then you could even, if you if you want to get fancy, um inside Google Data Studio, you can issue a custom BigQuery query as a view that allows that data to be blended at the BigQuery level first and then make you a consolidated, you know, maybe you have five websites uh and you want bot analytics on all five. Uh you could blend it all together into one big one before it goes into Looker Studio. So you can see a big picture of a view of here is uh AI bots for our portfolio of sites, or if you're an agency, here's the AI view, uh AI bot view of all of our clients, and you can see, oh wow, we're getting you know a gazillion bots a day. It's just yeah, go ahead, John.

Uh unless you've got your DBA merit badge, you you just want someone else to do this for you. I agree. Yeah, um, like I have this like little pain that's like starting above my eyebrow. I'm like, oh my goodness. Like I get how like I crave this data.

I've been looking for something like, you know, we've been trying to answer this question, but I also don't have the time or energy to figure out how to set it up. So I acknowledge and appreciate that you did take the time to do this and that you're sharing with us how difficult it can be if you don't, you know, readily have access to all of these systems and or are well versed in them. Um, you know, we're just watching the live active users and we're just seeing the number of bots just continue to creep up. I'm looking sort of in the what the bottom right hand side, event count by event name, number one AI bot hit, like the other thing I should mention is that in Cloudflare, which is where you deploy this, on the free account, you get 100,000 requests per day for free, and then you have to upgrade to the the paid version of workers. I installed this 16 hours ago, and you can see we have 62,000 requests already, right?

So we're we're clocking four to five thousand requests an hour. Uh now, granted, it's it doesn't strain the the worker very diff in a very difficult way, but this is this thing is you know warm. And we have in you know, if we're honest, we have a modest little website. Um if you have like, you know, uh it's on record, one of our uh customers is triple A, right? They have millions of p humans a day, they probably have billions of requests per day.

And so pulling off something like this, they're gonna get a bill. If they implemented this, they would get a pretty decently sized bill from Cloudflare for the servers that would need to spin up to handle this kind of load. Yeah, I mean, we're we often talk about B2B, but when you think about B2C, like the you know, consumer products, things that people are searching for all the time, like it's gonna get out of control really fast. Yes. Yeah, this has the potential to get very, very expensive very, very quickly.

Um that's good. Well, I was gonna ask, like, so when we talk about pricing for Cloudflare, without getting too deep into Cloudflare, um, you know, is it's is Cloudflare? So we recommend people have it regardless for their websites, but is Cloudflare a affordable system? Are there other CDNs like you mentioned, Akamai? Are there other CDNs that like a small business could do this kind of work with?

Cloudflare is the preferred choice for small businesses. Um we are on the $20 a month plan. Um gotcha. And that gives us a lot of features. Um there's like it's almost kind of like AI.

There's like the the this SMB 20 bucks a month plan, there's the 200 a month plan, then there's the pay as you go enterprise plans, because that's you know, the the bills are going to be large there. And it's at the enterprise level where some you know Cloudflare and Akamai start getting competitive in terms of pricing. Uh and below that, like the SMB level, Cloudflare hands down. Like we can't afford Akamai. Akamai wouldn't even talk to us for less than like $4,000 a month.

Um, and we're like, oh, $4,000 a month, $20 a month. Hmm. Well, and I think that you know, when we talk about analytics, a lot of times, like we're saying that the data is directional, the data is good enough. This helps us understand how many bots are hitting our website, or this helps us understand which pages we need to refocus on in order to make them more attractive to AI assistants. Like that's really what I need.

I don't need to know, is it you know 62,000 or is it 62,395? That's not going to make a big difference to me. Yeah, exactly. And the worker stuff, even on the free plan, you get some. So if you have, like, for example, Katie Robert.com is not yet a top 10 internet site.

Uh so this we could install on your site on the free plan, and it would probably have enough coverage to be able to do that off. And it actually is something probably worth doing because I'd be curious just to see how many bots are hitting your site, even though it's not a huge, hugely traffic site, just to see how much do AI bots impact even a tiny site. I'm gonna go ahead and say 100% of the traffic to my site is bots. I'm gonna I'm just gonna put that up. The site itself is like five pages, it's very small.

But I think that it would be a good test and also for us to do a demo of like this is what it looks like to set it up fresh on a website. Um, because I I mean, who knows? I don't think my site's gonna break the bank, but you never know. And more importantly, I would want to know which AI bots are hitting your site and why. Like, is Claude hitting your site a lot?

Is ChatGPT hitting it a lot? Uh and is it assistant? Is it search? Is it crawler? I would I would want to know that as a site owner, just broadly.

Like, yeah, if it's assistant or search, it means people are asking about you in AI tools and it says this is a valid resource. And that's helpful to know. Yeah. And we also, Katie Robert.com, which does actually exist. Um, I don't believe we have Bing Webmaster Tools set up on it.

So perhaps that could be next week's live stream is how to set up Bing Webmaster Tools. Yeah, we could do we in fact we could do all three. Um there'll be some parts that will be very messy because you have to go into the command line with Cloudflare's Wrangler uh uh node package. But yeah, we could do all three next week just for just for giggles. I mean, people are curious.

That said, I would highly recommend at least chatting with us about what it takes, depending on the size of your site, depending on what systems you have. So you go to trust insights.ai slash AI bot analytics. You will get a real human. That real human is right there. That's John Wall.

Um, I can't guarantee what kind of mood he's gonna be in, but he is a real human. That's right, humans are messy, so you know the dice. Um, but no, I think it would be interesting just to show, even just to show how messy the setup itself is, but then like start to finish, like what do you get? Yep. Get ready for a lot of you and John telling jokes while I type.

Well, that's the magic of television. We have pre-baked things. Yeah, yeah. All right, John, we're gonna have to work on our shtick. I think we come up with something.

Exactly. Yep. Now the last part, uh, and I'll recommend this. The last part of uh AI visibility that people do care about is competitive. And we answer that in our GEO 201 course, uh, which we're not going to get into because we're almost out of time for this week's episode.

Uh and also we have a course for that. But that would be sort of the level four uh is is to say, okay, you've asked people, you've looked in the approved tools from that have grounding in reality, you're looking at real bot analytics to see what content is getting hit and when and why based on bot category, and then the fourth would be looking at the activity data that you can get for competitors compared to the two. And that's that's AI visibility in a nutshell right now. The thing that I find interesting, like the Cloudflare stuff that we talked about today, that came out yesterday. Like that literally came out yesterday.

Uh as I was getting ready for the live stream last night to doing the prep, I was like, oh, I can implement this. Um it totally threw my plan off for today's show, but that's the space is changing very quickly. And I think that that's just another good reason why you should probably be connected with an expert like Chris who understands these things and that is their full-time job is to make sure that they're sort of hitting the latest and greatest. Um and yeah, so red flag to anyone who says that they can tell you what someone is searching directly in a large language model without doing any of these steps. Exactly.

Even then, even with these steps, it won't help. Uh that's gonna do it for this week's show, folks. Thanks for tuning in, and we will talk to you all on the next one. Thanks for watching today. Be sure to subscribe to our show wherever you're watching it for more resources and to learn more.

Check out the Trust Insights podcast at trustinsights.ai slash TI podcast and a weekly email newsletter at trustinsights.ai slash newsletter. Got questions about what you saw in today's episode? Join our free analytics for marketers Slack group at trustinsights.ai slash analytics for marketers. See you next time.