Most marketers are only using 20% of GA4’s capabilities. Whether you’re an SEO, media buyer, or marketing strategist, this conversation is packed with actionable insights to make your analytics truly accessible and actionable.
https://page2pod.com - Most marketers are only using 20% of Google Analytics 4’s capabilities — and it’s costing them big. In this episode of the Page 2 Podcast, we sit down with Brie Anderson, founder of Beast Analytics and one of the earliest voices bringing clarity to the chaotic transition from Universal Analytics to GA4.
Brie shares why small tracking errors can lead to massive strategic blind spots, how to fix bad data before it misleads you, and why GA4’s Advertising tab is a goldmine — even if you never run ads. She also breaks down her BEAST Framework, revealing exactly how to audit, analyze, and act on your data for real business growth.
Whether you’re an SEO, media buyer, or marketing strategist, this conversation is packed with actionable insights to make your analytics truly accessible and actionable.
📊 In This Episode:
• Why most marketers only tap into a fraction of GA4’s potential
• The high cost of bad data — and how to prevent it
• Brie’s BEAST Framework for reliable, strategic analytics
• Common GA4 migration mistakes (and how to fix them)
• Unlocking the hidden power of GA4’s Advertising tab
• How to spot and leverage multi-touch attribution data
• Tips for integrating Google Search Console with GA4
• Future trends in AI-driven analytics (and their pitfalls)
Brie delivers a masterclass in making your marketing data work harder for you — with a healthy dose of real talk about the industry’s blind spots.
Subscribe to never miss a deep dive into the tools, tactics, and trends shaping digital marketing.
💬 Comment below: What’s the most valuable insight you’ve learned from your analytics data lately?
🔗 Links & Resources Mentioned
• Brie Anderson on LinkedIn
• Brie Anderson on YouTube
• Beast Analytics Official Site
[00:00:00] Google gave us a free analytics tool with enterprise level capabilities, but the problem is most people are using only 20% of what it's capable of and when that data is wrong, it's not just a missed insight, it's a bad business decision dressed up like a good one. Ree Anderson is the founder of Beast Analytics and one of the earliest voices pushing for clarity and accountability in the post universal analytics world.
It's not just a coding GA four. She's guiding marketers through the fallout of an analytic system that changed faster than anyone could adapt. This episode is about the cost of bad data and it isn't assumptions baked into most digital reporting. We dive into why GA four feels so alien compared to universal analytics.
What marketers are still getting wrong about attribution and how small tracking errors became major strategic blind spots. Bree walks us through her Beast framework. Why she always starts with an audit and the unexpected power of Google's advertising tab. Even if you never run ads. Bree's mission is to make analytics accessible.[00:01:00]
She's constantly reminding clients, if you can't explain your numbers, you probably shouldn't be acting on them. If you appreciate a good conversation around data, you're gonna like this one.
Welcome to another episode of the Page two podcast. As as always, I'm joined by Joe devita, my partner in crime at Moving Traffic Media, and today we have Bree Anderson on the show. Welcome to the show, Bri.
Hey, I am excited to be
here. I'm definitely looking forward to talking through these analytics and how you manage that business, but also some of the deliverables that you put together for clients.
But in. Researching ahead of the show. I was excited to come across a MySpace reference. So take us back to when you were 13. So that was your first, as I understand it anyway, correct me if I'm wrong, that was your first sort of introduction to this wonderful [00:02:00] ecosystem and world that we all work in today.
Yeah, my MySpace was my life there for a minute. And you A lot of those, yeah. You're un unknowingly learning like what HTML is and like how to manipulate code a little bit. And then I, so I, I got MySpace. I was on MySpace for about a year and a half, two years, and it just fizzled out, just died. And all of my friends were in punk bands and that's MySpace is how we knew like when they were gonna have shows.
It was like. Google for musicians, right. Obviously. Um, and it, it died out. So I was like, when am I, how am I gonna know when you're playing or if you're gonna have merch or any of that stuff? And then coincidentally, the next, like what, when MySpace died, I started taking a Dream weaver course.
Oh yeah.
Then I learned, oh, this is what, this is what all that stuff was like.
We were just coding, we didn't even know. So [00:03:00] yeah. But also at the same time, like Wix was around and that's just how I was like, oh, maybe I could build websites for the bands and things of that nature. And I'm not a very creative person, and so I can't say that I ever did anything. Good. But the information was on the websites that, that I was building at the time.
And I just, I really just wanted to stay active in like, the music community. I, I moved around a lot as a kid and my way of doing that was staying connected online. 'cause it was really hard to find. I, I was part of a really good scene when I, and when I lived in New York and then I moved and trying to find stuff was hard, but staying connected with everyone, like the best way to do it at that point was online.
And so I, the websites and then Facebook finally came out with pages. And so then we started doing, there were a lot of like promotion pages and then I started working with, with promotion pages. Then another company that was a startup that wanted to become the next like music. [00:04:00] Social media obviously did not work out, but yeah.
So I've been doing this stuff since. Yeah, I was a youngin.
Yeah. Yeah. That's awesome. So bring us or fast forward for us to, I guess when your analytics journey started, again, as I understand it, it came out of a potential client loss and so you're trying to figure out how do I retain this nonprofit's budget.
So take us through that early time through, I guess, discovering like deep analytics in terms of how it can sing's perception and maybe how you fell in love with it from there.
Yeah, so it's, they're really, it's really not that much fast forwarding, I worked with that startup and then they were like, at that time I was, by that time I was like a freshman in college and they were like, you should drop out and move into, move to Nashville and do this full time with us.
And I was like, one, you're not paying me. And two, that's just the music industry in [00:05:00] general. Right. But also it's in two my parents. Kill me. There's just no way that's feasible. But I had, I was playing soccer in college and while doing this stuff and I just had this moment where like I was going to get a communications degree.
I thought maybe I'd go work on like a radio station or something and, but that was the first time, 'cause I was just doing social media for these people and I was like, this might actually be like something. That I can make a living doing. And there was one school in Kentucky, which is where I was at the time, that had a degree in social media marketing, I say in air quotes.
And so I transferred there and after my sophomore year of college, I got an internship. And at this internship I was just given like free range. It was a small agency and they were just, they were one of the only agencies doing digital at the time, like really focusing in on digital. And they had, I had done the Google ad certification, the Google Analytics [00:06:00] certification was just like really wanting to learn as much as I could.
And they had this nonprofit come in and they were like, we think that you can handle this. They have a thousand dollars budget and they're trying to raise money for whatever the initiative was at the time. And I was like. So you're just gonna hand me a thousand dollars. 'cause that seems like a crazy thing to do for a 19-year-old that's four months rent at this point.
And I was terrified. I did not wanna mess it up. So the first thing that I did was like, research, how will I know, like how can I set this up to make sure that I know whether or not we're making money on this? And that was my first time in Google Analytics and then figuring out how do you connect it to Google ads and like how do you see all that traffic and analytics?
And that was that point too. Like I, I like said, was interning at the agency. I learned very quickly that actually maybe I hate social media. It was really fun to do social media when I got to go to [00:07:00] concerts and get free merch and like that kind of content's really easy. Making content about sandwiches or blast resistant buildings, just not my thing.
Uh, I couldn't. Like I said, I'm not very creative. So coming up with interesting things to say about them, I was like, I got nothing. So the paid side of things really work. My, my like foyer. Foyer and I guess, I don't know what the word is, but into analytics because you get so much more data with paid, right?
And so right away I have this, oh, this is fun because I can see if I change this, what's the impact gonna be here and if I do this. And so I like unknowingly you do that with social media as well, right? Oh, these are the posts that go, at least the ones that aren't doing so hot. But then you get even deeper when you have paid.
So then it was like the kind of manipulation of that. And then, then I started learning about SEO. And it's the same thing right now. We have all these new data points and it's okay if I change this, what's [00:08:00] the impact on the click through rate? And if I change this? And all of it really is data driven and its own sense.
All of the pieces are data driven. It's just Google Analytics was the place that ultimately, at least for now, still our websites are where the meaningful actions are happening most of the time. And I had to at least have a base knowledge of Google Analytics, and I didn't really decide to go all in on the analytics thing until like 2020.
But even before then, like I made a lot of my money, like I just remembered this the other day that I made. One semester. Basically all of my money, I made writing content about how to use Google Analytics my junior year of college for this random website. Like I would just, I created back, this is when you could share like dashboard templates in Google Analytics.
So I would make templates and then talk about how you use it and what does this data mean and how would you move forward with that. [00:09:00] Well, I never considered that. Like people can't, people don't do that. In my mind it was like, yeah, you don't wanna just take people's money and not know what you're doing with it.
So, yeah, it's interesting. I think it all goes together and that's really why analytics ended up making sense for me. Right. I was always really interested in the strategy, but really only because it was data driven and that was a,
got it.
One of those like buzzwords for a long time. But truly it should be data driven to an extent.
There's no other way to describe it really.
Yeah.
Yeah. So I think one of the things that your agency is known for is the This Beast framework in terms of how you approach, or I guess really how you standardize your approach, right? It's, it always encompasses these things. Can you take us through what that framework is and how you apply it to, uh, potential client?
Yeah. Yeah. So the Beast Method was something I came up with in 2021, and it's really just, it's basically the scientific [00:10:00] method, but sounds cooler and branded for sure. Sure. But basically you have to grab your benchmarks, right? So B is for benchmarking, which. Generally, the first time we do it as well is like, how do we make sure that we're getting the right data, that we have clean data and reliable data?
'cause we don't wanna make bad decision or good decisions based off bad data because then it's still a bad decision, right? Make sure you have the right data and then it's clean and everything's good there. And then we take our benchmarks, which is basically what are we trying to do? So you define the goal and how are we, how well are we doing it today?
'cause if we don't have that idea, then once we run the test, we have no idea if it was better or worse. Right? Grab your benchmark data and then the ease is for explore. Explore all of your data sources. Look at what Google Analytics is saying. Obviously look at competitive data. Look, it obviously depends on the strategies that you're working with, right?
That could be like meta ads. You could be looking at your metadata, you could be [00:11:00] looking at com. Competitive research on simr, right? It's basically just get as much data as you can. Anything that could be relevant to what you're trying to do. And then once you have all the data, it's really just, I encourage people to play.
Like it's all gonna be messy. Nobody really knows what they're doing, so be messy. Go put two things together, see if they make sense. If they don't make sense, see what data you can find. And then the next part is the analyze the data. And that's where most people are like, all right, cool, I'm out. That's your job.
And sure, I'll take that. It keeps me employed. But analyzing data doesn't really have to be that hard. Generally speaking, you're looking for outliers. Anything that's an outlier is worth investigating. So you look for the outlier and then you find the trend without, within the outlier, right? Oh, we have this huge spike this day.
What happened? We look at our data sources. Oh, we can see that all that data came from being mentioned in this article. On this website, right? Or, oh, [00:12:00] we saw a dip in conversions. What happened? Oh, this key page that we were sending people to just got a lot harder to, to find because we restructured the website or something like that, right?
And it's finding the outliers, investigating the outliers, or finding the trends, the things that you can rely on that, that are happening over and over again. Like the biggest one that most websites see is, oh, our traffic goes down on Saturday and Sunday and then it comes back up on Monday. Right? Some more so than others.
But that's an example of a trend, right? Um, or the infamous, like every Thursday our traffic goes up. It's like, yeah, you send HRES to go crawl your website every Thursday. Like you things to be aware of, right? But they're all worth investigating. So then you get, you gather all these things, right? Here are the trends and the outliers great, but now how do we move forward?
What's. How do we actually use the data, right? Because a lot of people will go into it with, oh, I think we should do Facebook. So then they just look at their Facebook data and then they say, oh, like this is what I found on Facebook, and then that's [00:13:00] what we're gonna move forward with. Right now we have a test when in reality, like you have to be open minded one, which is why I think it's so important to have a third party that works with your analytics because you look at the same data over and over again, and you get tied to it.
And so I came up with this framework called the bucket method, which is the first question is did it work? And if the answer's no, you can't keep doing it. You gotta stop. That doesn't mean like full on stop. It just means you can't keep doing it the way you've been doing it because something's not working right.
It's not a good business choice, so you've gotta find something to change. If you're writing blog content and we're not seeing the value of it, we can't keep writing that kind of blog content. Maybe you're doing like in depth articles on the impact of medication on this specific, so you'd like this medicine's impact on this condition.
It's okay. Maybe that framework isn't working, so we have to come up with a different thing, right? Where it's maybe we don't do deep [00:14:00] dives. Maybe we're like the top medicines for this specific thing or whatever that might be, right? So you gotta change the format. Doesn't mean that the strategy has to go away forever, and we're like, Nope, never doing that again.
Just has to change. Something has to change. If it did make you money, great. Now the next question is, did you mean to do it? Because sometimes. We don't have a strategy for something, we just find, oh, this one website sends us a lot of traffic and it converts really well. And that's a great thing to stumble upon.
My recommendation would be like, let's start having strategy behind this. Right? So you have stop and you have start. And start would be like, we found this website. Let's reach out to them. Let's see if there's a partnership. Let's see if they have a newsletter. Maybe we could sponsor. Let's see if there's a opportunity for collaboration or whatever that might be.
Right? Or you find a trend and you're like, oh, all of these posts do really well. We didn't even really notice that they all have this one thing in commons. Great. Now [00:15:00] let's test more content like that in in, in a strategic way, right? And then. The other side of that is, yes, we did mean to do it. It's cool.
Now let's scale it. If the blog content's doing really well organically when we post about it on social, let's put some money behind it. That's the best way to find what you should be posting, like as ads, by the way, is if it does well organically, it will do well paid. You just put money behind that stuff.
So yeah, so you have the, basically the two questions, which puts everything into a bucket. These are things you could start, stop, or scale. And so now you have all of the data that you analyzed and the trends and the outliers, and everything gets bucketed into one of those three buckets. And then you choose one to test because just like in a scientific method, you have to have controlled testing, which means you can only change one thing at a time or else you don't know.
What the impact came from. And let's be honest, in [00:16:00] this industry, it's not like we have control over much of anything. So to an extent, none of our tests are ever actually gonna be controlled tests, but we'll do our best. But you pick one, and generally it's what's gonna be the least amount of effort that could, has the potential to show the most return.
And then again, that's just backed by the data. Oh, we'd have to get approval from six people to make this change on the site. Or we could use some of this floating budget we have and put a hundred bucks behind this organic post and see if it works well as an ad. Right? One of those things is gonna be a lot easier than the other.
And just for the amount of time that we would save it would probably make more sense to do that. And then. Once you test it, you track it, whatever the test is, that's you have to create a new measurement plan all over again and make sure, okay, are we tracking? Can we isolate this data in the places that we need to be able to isolate it?
How are we gonna compare it to the benchmarks? Is there anything new we need to track? And then you run the test and you compare it back to the [00:17:00] benchmarks. And that's why it's called the cycle, because it'll go on basically forever. So yeah, that's the cycle.
When you take on a new client, you always start with an audit.
We read that about you always. I assume even before that audit, you have a conversation. It's like a little bit of a discovery call. I'm curious, this has happened to us a few times where in that discovery call, the client says, this is who my target, this is who my customer is. She's this, she lives here, she does this, she likes this.
But you jump into their analytics and you do that audit. Do you, have you ever had to. In that next conversation say, you know what? Your customers aren't really who you think they are. There's a different pool of people. Walk us through a story like that, because I always find interesting where it was like a big, or even a subtle difference between who a company thinks their customer is and who it actually is.
Yeah. That, that does happen a lot. So to be clear though, like my audits are specifically, I'm just looking at tracking setup because if [00:18:00] we don't have a good tracking setup, then we don't have good data, and that's generally the first problem that we need to make sure gets solved. But then, yeah, when we're doing like strategic audits and people are like, oh, like we're using, this is what's driving all of our, our money, and these are the people.
I, I can't think I, it's been a while since people have come to me with a specific target audience, I would say, and I've been like, no, that's not who it is. Because people aren't generally coming to me to talk about audiences. It's more so, oh, this channel, right? This is the channel that, that's having the impact.
And yes, I, I have to kindly say, so yeah, you do see a lot of traffic from your display ads. This is generally the one that I get. I worked with a big company a couple years ago, a couple million dollar marketing budget. They had an outside agency. They, it was a decent sized company and they were like, yeah, so we're putting 60% of our budget into display [00:19:00] ads because we know that's what's driving our traffic.
And then we'll have some of this stuff, a traditional, and these sponsor, these newsletters and do this and this. That was their media plan. And I was like, okay, cool. Let's meet in a a month and we'll see how this is going for us. And the agency was on the call, that was that did the media planning on buying everything.
And I had to go on there and go, we did get a lot of traffic from display. Unfortunately that was like a 96% like bounce rate, which meant only three or 4% of people are engaging. The average engagement time is about two seconds. If it were me, I would maybe wanna shift some of that and like right away you're getting the, it's for brand awareness and it's this and it's that.
And it's, yeah. But that goes back to to like in the benchmarking conversation, what was the goal? The goal was to get people to these specific pages and ultimately for them to become a lead. And if they're spending three seconds, two [00:20:00] or three seconds on the pa, the side, they're not achieving those goals.
So either we need to change the goals because the goals are actually different than what we had agreed on, which. Now you're saying it's traffic. That's fine if that's what you want the goal to be, but if the goal is actually we're trying to get people to these pages and then have them convert and become leads, then we have to change something.
I'm not saying you can't run display ads. I'm just saying the way that they're being done right now is not working. Simple as, and it sucks to like especially have to come into an agency where you know that they're like, they're making money, they're happy. They're like, yeah, this, we know our stuff. And you kinda have to come in as like the third party and be like, Hey everyone, how you doing?
Good. So what you're doing not working?
That's the beauty of data, right? I mean it, but yeah, it doesn't lie in terms of what's working and what isn't. Um, yeah. Some of the things I like about data, it's a little bit more concrete. You can definitely, I don't know if the word manipulate [00:21:00] is the right word, but you can definitely skew data to make it look Oh, a thousand percent how you want it to.
But
yeah, and that's honestly that. That's kind how I like really honed in. It was twofold, but one of the reasons that I really honed in on like analytics and specifically educating people around analytics and like what data means and where it lives and how you access it and all of that was because just working in agencies around agencies with agencies, not to say that all agencies are bad, but you do see how the data gets manipulated and clients know no better, right?
So they don't know that they should be going in and looking at this other number, right? Because they're just seeing all our traffic went up. Great. We're seeing all these green numbers. It's okay. But when you break that traffic down to the pages that you wanted it. Them to go to, right? You have to teach people to, to have these conversations and to look for those kinds of things.
And it's not as [00:22:00] common now, but like agencies or businesses in general, holding clients' data hostage or connecting their own things to people's data or getting the data and inflating the numbers to reflect their commissions, all of those things, it's okay, but we have to be transparent about that and that's why I'm such a big proponent of education is because I never want somebody to be like taken advantage of because they didn't know to ask the right questions.
So for a typical client, you always start with the audit, and then depending on the length of the engagement, do you provide the performance reports for them? If they have someone buying media and someone else doing organic, are you the kind of central person to share how everything is doing? When looked.
Together.
Yeah. I mean that it's kind of, I, that's the ideal situation. That's not always how it goes, especially if there's an agency involved that's like a one stop shop kind of agency where they're like, we [00:23:00] handle everything. They wanna take that. And generally what ends up happening is I will, I try and have meetings with the agencies or providers on the line because I don't want there to be any miscommunications.
I wanna be available to have the questions asked either way. But I do have some clients that are just like, we just wanna know what's going on. It's more of a monthly audit, so to speak, of, okay, this is what we were told and this is, this is what we're seeing. What do you think about this? But really just because it just depends on the client.
If we are in some sort of like monthly engagement, then yeah, I'm going through the data and then coming to them and saying, Hey, this is what I'm seeing. And the. That's only like the first half of the meeting is, okay, this is what I saw. These are some things that I kind of wanna flag just to make sure that we're all on the same page as like to what happened here.
But now just to check in what are we doing this coming up? Are there any new initiatives launching? Are we [00:24:00] trying any new channels? Do we have tracking plans for those? What are we expecting to see and what are the goals moving forward? Because if I don't have that information and then I get in and I'm like, a week before the meeting, I'm like, wait, where did all of our meta traffic go?
Something broke or whatever, and they're like, oh no, we pulled it. We decided to go do this in-house or go sponsor a conference. Now I have to go. Cool. Except for how are we tracking if those people came to our website because we didn't come up with any sort of like special link or, oh, we don't have a page for them.
And now all of our data is skewed and we weren't really ready for that. And now my analysis is junk. So we, I like this month's meeting is now, but ideally we're more in cahoots. Staying closer in communication than that, but it has happened and people don't think, think about it, but the goal is to like work with people to integrate the idea of tracking and [00:25:00] analytics into their processes, to where it becomes second nature.
Like we don't have to have a conversation every month about, okay, and now what are you doing on this platform? And what are you doing on that platform? And we might wanna change the way that this is structured. If we're doing that well, we might need to create an event here. If you're adding that, that functionality on the website, those kinds of things where eventually becomes like a, oh yeah, like we've got that.
We already set that up. Ever. Ever. That's the.
Have you, have you learned something from a client or have you had like a client request, something so unique that you thought, oh my God, this is new for me. I need to like, I need to spread the word on this. There's just something really cool that everybody should be thinking about.
I just worked with a client LA the first half of this year that really, they came from like a product background and they were, then they became like the head of growth at a company and they asked for a bunch of things that I never would've even [00:26:00] considered tracking and it really sent me down this tailspin of, which is hilarious.
I had to take a sales class in college and I thought it was the dumbest thing ever and I hated sales and I still hate sales, which is where I could probably be making a lot more money, but. I just hate that part of it. And, but we had to read this book called The Power of Influence and they, he has these seven I or principles of influence.
I have them up on my whiteboard because I literally, it sent me down this rabbit hole of, okay, there are all these different things that you do to influence people. And is an analyst is someone that, that works on the very logical side of things. It was always like, do they start the form? Did they complete the form?
Where are they dropping off? Where did they come from? How long do they spend here? And this person came in and they were like, we wanna know if like, how many testimonials they read. And we wanna know if they're sharing what social platform buttons they're engaging with the most. And just like all of these things that [00:27:00] I was like.
Holy shit. Yeah, those are super
product oriented questions like
that makes sense because if you look at like the principles of persuasion, it's or influence, persuasion, it's reciprocity community. Let's see, authority, scarcity, liking, like all of these things and you're like, yeah, that's what all of this is.
It's just manufactured and we're putting it in front of 'em. And so that's actually like really changed the way that I like look at people's websites and go, okay, what could be the trigger here that made them start the form? Like before they even, because we look at it, especially someone that worked in SEO, I'm like, okay, what content are they reading?
What pages are they interacting with? But to get down into that like very specific, we're trying to do X with this specific modal on the website, like that kind of really shook me recently in a good way.
So if you're, so if you're working with a client, let's assume they're using Google analytics. Are there.
I don't know, like core KPIs that you, [00:28:00] I don't know. You think about applying or tracking for like most, if not all of them.
It's basically at the end of the day, it, they're all business objectives, right? You're in the business of selling t-shirts. How many t-shirts did we sell? That's what we need to track.
You're in the business of bringing people in the door so that you can sell 'em X service. The closest that we can get to tracking on your website is when they fill out this form or they call you or they email you. Right? So like B2B, it's always like form fills, calls, emails. Perfect. Got it. E-commerce, it's always.
Purchases and then all those parts of the funnel as well, right? FU has like their own recommended events section for GA four that kind of like breaks all those down and now they even have that for leads, but then that requires using measurement protocol and a lot of people mess that up. So I'm like, now we'll just do generate lead when they do one of those three actions and we can break it down and then you can decide from there whether or not that person became like an actual lead or whatever.
But yeah, it's pretty much, [00:29:00] that stuff is generally the same. And then the rest of the tracking plan just comes down to what are we doing with our website here? What are we trying to do, what can people do and how do they interact? And can we even change some of these things, right? It's really easy to go in and be like, oh, I have to check all these things.
And so if we're never gonna change it, if you're never gonna get approval to take the sister brand's logo out of your header that people are clicking off to go to. I don't even wanna know. I don't wanna know because we can't change it. They're not gonna let us.
Right. Universal analytics. I wouldn't call myself an expert, but I felt very comfortable in that ecosystem.
And then of course, J four came about. Give us your opinion about the two. Were you, if upset is the right way to characterize it, but were you struggling when they made that transition? And I guess, how do you view it today? Is it better, worse, more complicated job security? What? Like, how do you think about it?
Definitely
job security, I would say is [00:30:00] probably the number one thing I think. But I wanna do this series. I always have these ideas of, oh, I wanna do this series where I created videos within a month and a half of GA four coming out. Where I'm like going through it and being like, what in the world is this?
Like, where's my landing page report? What's an analysis hub? What is, because that's what Explore was called when it first came out. And. Like why, where is this metric and what's this dimension? And so when GA four came out, I was very bitter because I was teaching at the time. And literally the night before I had been like, okay, we can set tag manager up using on Weebly sites.
So we'll use a Weebly site and if we do tag manager, then we can get this and we'll put Google Analytics on site like this. Perfect. Like just making sure everything was gonna work because you can only control so much. And then the next day I went in to teach, I was like, oh, we'll create a GA or a Google Analytics account.
And I was like, oh, this looks a little different, but it's okay. Here we'll just go like this. And then I was like, and you'll grab an ID and it will start with UA Dash. [00:31:00] And then I looked at it and I was like, that's a G guy, what is this? And then I go, oh, welcome to digital marketing. Everyone. Literally overnight they created this.
So then they had to create both a UA and GA four account because at the time you could still do both. I also sold the writing on the wall with GA four very quickly. Like what? Where. We were going and why we had to go there, which was basically a shit ton of litigation around, we're collecting a bunch of data and storing a bunch of data that we should not be.
And you know, Europe doesn't like where it's stored and now Google's getting sued for all these things. So basically that's what happened. Like it was illegal, like a lot of it was deemed illegal. And so Google had to make a change and two cookie list browsing and ad blockers and everything. They had to come up with some sort of solution for that, which is GA four the best solution for that?
I don't know, but at least it's better equipped than UA would've been. And so I knew and figured [00:32:00] that it would stick around and everyone's, it's not sticking around. It's not sticking around. No. Like first of all, they've made these changes before. Granted, we got to keep all of our information and like the data structure wasn't completely different, but like they've made the changes before.
And two, if. Taken like the litigation part of it, like this, it seems legit, like it's gonna happen. I don't know when it's gonna happen, but it's gonna happen. And I just decided like eventually it is going to happen. So I'm going to learn it and I'm gonna figure it out and I wanna know because I don't wanna get left in the dust.
And thankfully it just happened at a really good time. 'cause GA four came out in October of 2020. And that was like right around when they like made our, I was teaching and they made us go back in office and I told them like, I'm not coming back in office. Like my wife is trying to get pregnant. I have terrible health anxiety.
We are in a pandemic in case you forgot. And so I told him like, the fall's gonna be like my last semester teaching. [00:33:00] I will adjunct in the spring. I'll teach my two or three courses, but I'm not gonna have office hours. I'm, I don't wanna be like the director of this program anymore. I'm just gonna do. The class is in dip and then this is gonna be my last semester.
And I went all in on on GA four. And I think that's what kept me alive after that, like making the leap into entrepreneurship, being one of the only people that talk about GA four. That's what got me my speaking gig at MozCon is I was the only one talking about, and I was talking about it a lot and just sharing everything that I was finding and that just snowballed from there.
And so I am thankful in a weird way, but also like I truly do believe, like I really do believe that GA four is such a powerful tool that we are lucky as free. Right? It has, you didn't. So many features that like you would have to pay, maybe not in today's age, but when it came out, thousands of dollars for like, it, it [00:34:00] had anomaly detection out of the gate that had anomaly detection.
It had insights which are just like alerts that you can send to your email. Like it had all these things that you would've not, like, you would've had to p pay a data scientist to create for you or pay for some tool. That's really expensive. But, and my, my mission statement is like, to make analytics accessible and actionable, and Google Analytics four is free, so accessible to everyone.
And it is a very powerful tool if you know how to use it. And that caveat is the biggest part that, that keeps people away from it is like figuring out how to use it can be tossed. I,
I think I read somewhere you, you. I might be getting the percentage wrong, but you said something about the effect of most people are only using about 20% of what is actually capable in the platform.
Is there anything, I don't know, a feature or I don't know, a filter that you would [00:35:00] recommend our listeners like add or create or dig into like tomorrow?
Yeah, I, the biggest one that I always run into that people haven't like explored at all is NGA four. There's like a primary navigation, right? It's like home reports, explore and advertising, and I tell everyone, you have to have the advertising section.
You have to have data in there, one, and to have data there, you have to have a Google Analytics or a Google Ads account linked to your Google Analytics for account, even if you're not running ads. So I have an account that's set up. I, I don't run ads. I've never run ads, but I have a Google Ads account set up and it's connected and you have to have a key event set up.
So if you don't have those two things, the advertising section's not even gonna work. So one, do those two things and then two, go to the advertising section. And in the advertising section there's like a secondary navigation and un there's, that's where your attribution reports live. So if you click into attribution, there's attribution paths and that shows you the equivalent of multi-channel funnels [00:36:00] from Universal Analytics.
Everybody thought that left and that was gone. It's not gone. It's there. You just have to go to it. My network's made up at this point, mostly of SEOs and so they're like. Never in that advertising section. And I'm like, no, you have to be there because organic is generally sandwiched in the middle of all of these of the paths.
And so you're not gonna see it attributed in, in the acquisition reports like first user acquisition or traffic acquisition. You have to go into that, that those attribution reports and see all of the organic touch points in the middle of the funnel. People just don't even know that it exists there.
I wanna just rename the thing attribution from, why do they call it advertising?
You don't need Do.
So now it makes more sense why they call it advertising because now they've, now they support a lot more like ad cost importing into GA four so that you can see like how much you spent on Facebook and how much you made. And the same with Google Ads, [00:37:00] obviously, but that didn't used to be there.
That's actually relatively new. You could always do it. It wasn't very easy. Now it's a lot easier. So that part's relatively new. Maybe they just knew that it was coming and that's why they decided to show, call it that. And then also just because they want people to spend money on Google Ads. Correct. I don't know.
I've been saying for years that they should have called that attribution, but nobody at Google wants to talk to me. I've tried. Haha.
A great use case of that. Actually. We were looking at that multi funnel, if you will, or multi-touch attribution screen, and we found that we work with a company that creates workspaces, like really high powered machines, and we found that one of the first stops on their solution pages for different types of software that need these types of machines was from organic.
And then most of the conversions then came through paid search, email, et cetera. So we actually pitched them on a, like an organic remarketing strategy where we tried to get. Better visibility for those solution pages. And then we use page search to pick up those, [00:38:00] those visitors that didn't convert on that first visit.
Super powerful. Yeah, it's a great tip.
Yeah. And it's just, it's one of those things that's so overlooked, so easy to overlook, right? It's overlooked, but that's just 'cause it's so easy to overlook. And that's, that's what I'm, I constantly talk to people about is you have to know your analytics because there is going to be a time when somebody comes to you and goes, what's the value you're bringing?
And if you can't answer that question, right, absolutely. Because you don't know where to look, that's gonna be a real shame. Knowing that's there. The other thing is, especially when it comes to SEOs, like I said, most of my friends are in SEO. A lot of my clients, we focus on SEO, so I have a bit of a, a bit more of a background on that.
But is if you're creating content for a client, whatever that might look like on on a website to create an audience, you can create an audience or you can create a segment of users that visit that page and then become a lead, or [00:39:00] then complete whatever that key event is, because that will give you a concrete number, right, of I can tell you that.
There were 34 people this month that visited a blog post and then filled out our lead form or, and then ended up purchasing. And that can get more powerful however you decide to set it up and use it and all of that. But it's like anytime you can create a concrete number, it's going to be helpful. And thankfully in GA four there are ways to do that.
So I highly recommend doing that as well.
Creating audiences that you can then use for advertising segmentation is really smart. Really smart.
Yeah. Have you found any useful ways to integrate Google Search Console data into GA four?
Yeah, so search Console data is available in GA four if you connect them.
It is tough though 'cause it's not like it aggregates weird. It doesn't show up very well. So I actually, I've built out a Looker [00:40:00] studio. That combines GA four and GSC data, but mostly I haven't been like in the weeds in SEO for a long time. And so mostly I'm just looking at like content, like what content's performing well, and that's based usually on landing pages.
Yeah. So I'm not getting like into the nitty gritty with like keywords and that kind of strategy anymore. Yeah.
I know the migration. Whether you were getting auto migrated or doing it yourself was cumbersome in terms of mapping everything up correctly. Are you still uncovering issues from those migrations?
Oh yeah. When you're doing your audits? Oh yeah. What's the, what's a couple of common things that you see that maybe people can be on the lookout for?
So the biggest one that will tell you right away if something was auto migrated, is if you go into the admin section and then under data display go into the events report, which is now key events and events, which I don't know why they did that, but whatever.
And then you toggle over to the events section and [00:41:00] then you click on Create events or manage events. It will show you events that were created in GA four, which most people doesn't even know. Don't even know that exists there. And most of the time, if you see that, and if you see in the custom DI dimensions, the event category and event action, guaranteed that was auto migrated.
Because what GA Ford, what they did was they would, oh, we'll auto migrate it for you. We'll just create, because remember, you could create goals in universal analytics, right? Whereas if they view this page, then it's a goal. There's something somewhat similar in GA four, but the GA four data structure is much more complex than universal Analytics was.
So it's very easy to mess up and Google was just like, oh, we'll do the right thing and we'll auto migrate all of this for people and it just does not work well. And so yeah, [00:42:00] that's like a telltale sign that it was just auto migrated, which is fine. But again, like that's why I do the audits 'cause. Most of the time it's just not tracking what you think it is or what it should be.
So yeah, I advise against using the event builder and GA four at all. If you can, just because you have to know the data structure really well in order to use it effectively,
your recommendation is to create the success what whatever the KPI is to measure it through Tag Manager and then share it with Google Analytics.
Yeah. So create an in tag manager and push it to GA four and then whatever other destinations you have, right. Whether that's meta, Google Ads, whatever. Because also something I'm huge on now is like documentation and making sure everybody knows how everything is set up so that we're not breaking things.
And if you have something set up in Tag Manager and something set up in Google Analy, like in Google Analytics, and then you're like, why can't I see that meta? It just gets really messy. And [00:43:00] so the cleanest way to do it, in my opinion, is to just do everything in Tag Manager, have like. Naming conventions so that everything's really easy to find and document the hell out of it and give it to everyone and say, if this stops working, look here.
And then
it's like an extra step and it's really smart. Super
important. Yeah.
I, I was, I was listening to an old podcast with you and Garrett Sussman, I think actually it was back in 2023, and there was a section in there I'll link to it in the show notes. There was a section there about the differences between how Universal Analytics and GA four used this metrics.
So for example, users versus active users, like very different between the old platforms. Yeah. Have they done any better about I didn't think so. No, no, no. I was kind of rhetorically, but
no, they didn't roll any of that. They haven't rolled any of that back and they really haven't. [00:44:00] They just, the biggest, in my opinion, the biggest flop around GA four.
Was the lack of education, and I'll leave it at that, but the lack of education that they offered people that have been using a platform for 10 plus years and that they just changed all of these things without giving anyone an explanation as to what it means or, sure. There's some documentation out there that's like, why was my, why are my user numbers different in UA than GA four?
And by the way, they still have that up, which is hilarious 'cause it's, you don't even have UA anymore, but whatever. So like that stuff was helpful, but like you could have just put that in the platform one, like a tool tip. That would've been great. But now it's because also we're, the users weren't educated, but then they were required to turn around and educate their clients.
They didn't have the necessary tools. But anyways, no, they haven't changed any of those or made them any easier to understand. Another key example is like we were just talking about [00:45:00] attribution. If you look in the first user acquisition report, it's gonna say one thing and then if you go to traffic acquisition, it's gonna say another thing.
And then if you look at it and explore, it's gonna say another thing if you use, depending on which, which one of those dimensions you use or if you use them together. And then there's this like third acquisition source, which is just called source. And that's what is only ever tied to key events because it's the one that got the.
Largest share of credit for the attribute for the conversion. And it can have like percentage of an attribution connected to it. So then the numbers are we too, and like how do you explain that to someone? The data here is aggregated. The data here isn't, this is looking at first touch, this is looking at last such.
And this one is just looking at whatever the model says and what's the model. I don't know, I can't really explain it to you. It's an algorithm that they came up with and I have an idea of maybe how it works, but you're just gonna have to trust me like [00:46:00] it's a little bit, it's tough. And I understand the frustration behind it for people
is, is there a view that you recommend in light of the filtering versus not?
Or it, 'cause it is hard to get your arms around and depending on how the client's pulling it versus how we're pulling it or, yeah, another agency is pulling it. Getting those numbers aligned are tough,
unfortunately. It really depends on what you're looking for as an org. Right? What do you value more For me.
Every one of Mo, most everyone I work with had some experience with Universal Analytics. And the closest reporting that you can get to that is to look at traffic acquisition, because everything was session based, everything was last click. I do a lot of reporting on session source, medium, uh, and session metrics and, and slash also slash 4:00 AM giving people a glimpse into the multi-channel funnels because it's still, again, they would've had some exposure to that because I've always [00:47:00] been, even before GA four, big proponent if you have to understand the multi-channel funnels because it all works together.
And that's just coming from my experience of I've worked in social, I've worked in paid social, worked in paid Google ads, worked in, you know, organic, like SEO world. So it all works together. If you take one piece of the puzzle out, then you're gonna lose out somewhere, especially if you're not doing it strategically.
And so. I was always like, you have to understand all the touch points. You can't just say, oh, we're only using last touch, or, oh, we're only using linear or We're only using this. It's like, yes, everything has some sort of an impact. And so I do think there is a time and place right for each of those reports, but you have to understand who you're talking to.
At the end of the day, you always have to know who you're talking to and what they're looking. So if they're looking for, okay, where did they come from before they convert, I can tell you that they generally, they come from a Facebook ad and then convert, but [00:48:00] before that, they've had a touchpoint with paid search and a touchpoint with organic search.
I
got a quick follow up. I know we're running low on time too, is like the referral traffic question. I'm curious. We're starting to see trickles of LLM traffic. Mm-hmm. Up for almost all of our clients. Very small, 1% of traffic. But there's a couple of trends that are starting to emerge this year. I wonder, are you seeing that same thing with your clients?
Is it still what sub 1% of traffic or is it different for different industries? Are you seeing anything interesting?
I've seen a couple clients that are getting. Decent amounts of traffic from LLMs. I know John Henry and his folks over at Growth Plays, they do some really good work around 'cause they do SEO, but they've apparently, 'cause I asked one of my clients, I was like, oh my gosh, you guys are getting a lot of LLM traffic and it's good.
Like it's converting. That's impressive. And she was like, yeah, John Henry and his folks over there are doing a good job. So I, as more agencies [00:49:00] start to look into what it looks like, we're gonna see more and more of it for better for worse. And, but I am seeing more of it. I would say most people, like you said, are probably gonna sit at that 1%, maybe even lower.
But I would argue, 'cause I'm thinking of another client now that I think about it, that they are getting engaged traffic. Like pe the people coming from LLM sign up for the newsletter, for instance. Like I know that off the top of my head, it's never about, you can never look at one number. I always say like any metric looked at a loan is a vanity metric.
So if you're just looking at the traffic numbers, yeah, it's still really low. However. A lot of that traffic because you have to think, I mean, we all use LLMs. When you interact with an LLM, there's a very low chance that I'm going to click on a link out. 'cause I just want answer to a question. If I click on that link, I have some sort of need that I think that you're going to fulfill.
And I would say, even if it is low numbers right now, [00:50:00] I've still seen very engaged traffic and even converting traffic. But I will say, just a quick note on that is you we're gonna see a lot of direct traffic for a while chat. GPT thought they knew what they were doing, adding a UTM to their links and they set them up incorrectly.
Always make sure that you're looking at your direct traffic. If you go into the explore section of GA four. Go in, bring in your direct traffic, like filter for just direct traffic and then add in the dimension page referrer. And a lot of times you can recover some of that direct traffic. It will tell you, I know a lot of Android apps, all that traffic comes through as direct.
Specifically LinkedIn traffic comes through as direct, but you'll uncover some LLM stuff there as well. Yeah, but most of them you can see just by the source. Most of them you can see just by the source. It'll say Plexity or Claw or whatever.
I was gonna ask about that specifically in terms of are you finding any ways to better make [00:51:00] that connection?
Sounds like that might be the best approach for now.
Yeah. And then again, just, I mean that as far as tracking that traffic, is it coming to our site? Yeah, just looking at the sources and then looking at page refers. Just to double check that nothing is getting lumped into direct because it does or that it's not getting direct lumped into the wrong thing because Google also uses a last non-direct touch.
Attribution. And so if it got lumped into direct at any point in time, it could still come through as direct. I don't know. It's a very strange thing. So yeah, definitely check out your page refers 'cause there's a lot of data in there that is in incorrectly thrown into direct traffic. But as far as looking at understanding that traffic goes, I always look at what pages those people are landing on because that gives you a good idea of what the LLMs are using your site for.
Right? What kind of data do they think is important from your site? And you'll see a trend. No, all these people land on these kinds of [00:52:00] pages. Those are the ones that are getting served up.
You know what's crazy is we started to go back 12 months, export all the URLs that have had a session, and we're finding like five to 10% are broken.
Yeah. Either just hallucinated pages or they're part of old training sets. And so that's actually become a little bit of a deliverable for us is like just doing a quick status code check of all those URLs and saying, Hey, can we get a redirect here? Or maybe we wanna bring that page back to life if the content still exist.
And so you
can create, you can create a custom report in GA four if they all have this same like title, like generally broken pages, have a title not found or whatever. So you can create a report in GA four where you bring in the page location, you have to do page location, 'cause that gives you the full URL.
So you bring in pull page location, and then you can, so in the, in the actual report section of GA four, you can create your own reports or you can edit reports. So [00:53:00] you'd have page location, you add a filter for page title contains, and then whatever that is. So not found unavailable, 4 0 4, whatever. Um, and then you just bring in the metric views and that will show, you'll have a report that shows you all of your broken URLs that have returned or not a 4 0 4 or not found
very nice.
So that way you don't have to pull 'em every month manually.
Yes. Good point. Google Analytics also just released their MCP server. Any thoughts about that in terms of good, bad, the ugly fighting?
I have mixed feelings. Garbage in, garbage out is the biggest thought process there is. Yeah. Like I said, well all my clients go through an audit process, so we need to make sure that we have the best data possible.
And also all of my clients in that audit process, I give them recommendations for events that are customed to their website and their users and all of that. And. The MCP [00:54:00] isn't gonna understand that, right? This is the argument for all LLMs at this point is they can't put those pieces together. They don't have an understanding of your website, who the person that comes to your website is what they've been through, that's led them to your site.
What things can be clicked, at least not yet, right? What things can be clicked on? Why would we want them to click on that if they spend X amount of time here? What does that mean? They have somewhat of an understanding, right? But especially something that's been trained on Google's like data sets and things of that nature.
They're looking for one kind of setup. Most people, if they've set their stuff up correctly, have some things that live without, like outside of that, right? So that's one part is. The chances of them understanding your full setup are slim to none unless you're using only the names that they gave you, the dimensions that they gave you, so on and so forth.
Um, and then the other part of that is if you didn't set it [00:55:00] up correctly Yeah, garbage and garbage out. If you have bad data going in, you're gonna be given recommendations based on bad data. And then understanding the data structure enough to be able to query it and know what's actually coming back to you is hard.
Like in the video where they launched the, where they were talking about like how you can connect to it, he, he does a query and he is like, how many users did I have yesterday? And it comes back and it's a 3,400 or whatever the number is. And my first thought was like, yeah, but is that total users? Is it active users, is it new users?
And is anybody gonna know that they should ask or that because an LL M'S only gonna know what an LLM knows and that they don't return. And it
always gives you an answer. Yeah. Even if you ask the wrong, you always get an answer, but you ask the wrong question. Yeah. What you do with that answer, you're,
yeah, so I'm a little bit nervous.
I know people are really, and then the other side of it is like, what are they doing with the data that's fed to it? Is everybody [00:56:00] else's question, so I don't know.
And where is it being stored? And we're almost right back to universal analytics. Right,
exactly. Yeah. They're just digging themselves another hole.
I don't know.
Yeah, exactly. Before we let you go, I wanted to get your thoughts on the future of analytics, like AI driven analytics, or I know other predictive models, like are you playing around with anything that you found useful or I don't know that you would recommend? As far as reporting or insights, even
I, yes and no.
People are very protective of their data, so it's not like we can just be putting people's data into to these models. Right. Because we. Know where it's going, who has access to it, all that kind of stuff, right? So that's one part of it. And then the other part of it is just, I play with it every once in a while.
'cause I just wanna know what my longevity looks like one and two. Like how do you use this to be become more efficient? Because ultimately that's what we should all be doing, right? [00:57:00] Like I've tried, I'm like, oh, here's this website. Can you tell me what kind of things I would need to track on this website?
And it just comes back with absolute garbage and half the things don't exist. They're just like, this is what you track on most websites and it's cool. That's not real. And by the way, all LLMs suck at GA four, and I know because I'm teaching college courses and get. Chat, GPT assignments back, and I go, you all said the same thing, and half those things don't exist in GA four.
Those were from Universal Analytics, so how about we try that one again? Oh, well
that, that's pretty amazing. Not surprising though, unfortunately.
Yeah,
unfortunately. Listen, this conversation's been awesome. Uh, before we let you go, we'd like to do a little bit of a prediction, so I'd love to get your opinion on if you'd go to Google Analytics in the next 12 months.
Could you foresee any, I don't know, landing page changes or features added? What, what do you think that experience will look like in a year?
Yeah. They just started trying to push out more [00:58:00] their AI assistant in GA four, which right now. It's terrible. It, it'll give you an answer and then you click to see the data set that it pulls from, and it takes you to a completely wrong part of GA four.
So that kind of sucks. So I suspect that will be better and hope it becomes better. I hope people are able to query their data in a way that feels natural, and I suspect that they'll be able to do that. I think that will be in the interface natively within. Maybe not 12 months, but I see it in the near future that people will be able to query their data more naturally like that just through a chat.
And by the way, like Google's been playing with that for a long time. Like a long time that you can like query your data in the chat. It's just never been good. So hopefully they, they figure that out. I suspect we'll see a lot more of that. I think it'll become a lot more chat based, just like everything else.
But I also think that all of this is a pendulum, right? We're gonna go really extreme one way and it'll come back and go really far the other way and then it'll [00:59:00] end up leveling out where we see like a nice happy medium. 'cause right now, in my opinion, most of this chat stuff's actually quite invasive because it's, if I wanted to chat, like I know five platforms where I can go and chat right now.
Like I just actually wanted to pull something myself. Yeah, I think that's what they're gonna be focused on and because they have to play the game. But I think like the biggest thing that we can do is. Hone in on our critical thinking and actually put that to work because they don't know. Yeah. In my ideal world, it would be like, so what did you try and do this month and what are where it would have more information and more background?
Because like you said, Joe, like it's gonna give you an answer. That doesn't mean that it knows what you were trying to do or has any sort of context around the data, like numbers by themself, don't tell a story. I would just caution everyone with that and say, remember to use your brains please.
That is excellent advice to end on.
Use your brain. Very good. Love it. Pree. Thanks again for [01:00:00] joining. Before we go, tell everyone where they can find you on the line.
Oh yeah. Just Google Brie e Anderson and you'll find me. I guess I'm on LinkedIn the most now because everything else kinda. Sucks. But if you find a good social platform, let me know and maybe I'll hang out there.
But I've got a lot of stuff out about GA four and I'm always happy to talk about data stuff and help any way that I can.
Fantastic. Again, if you enjoyed the show, please remember to subscribe, read and review. We'll see you next time. Bye-bye.