GA4 Audiences: How to Segment and Use Google Analytics 4 for Better Insights

[00:00:00] Amanda Webb: Welcome to the show, Dana.

Dana DiTomaso: Thank you so much for having me.

Amanda Webb: And I have to say, you have been my analytics hero for a long time, and I’m always just, I get annoyed because I see something that you say, and I go, I wish I was as clever as Dana. So I just want to start there because I think I’ve got the smartest analytics person I possibly could on the show. So I’m delighted to have you.

Dana DiTomaso: Thank you. Thank you.

Amanda Webb: But I want to know, I want to know, where did this all start? Where did your journey into marketing into analytics start?

Dana DiTomaso: Yeah, I actually have a good one for this. So I actually have a degree in geography. I went to school in the mid nineties and so the internet, like the first web browser came out my first year of university. I think that if I been slightly younger, then I probably would have ended up in the internet from the start. But there was no internet yet. So I was like, I like computers, but I don’t know what. And so I ended up going to geography because I also really enjoyed physical geography specifically. And I have a degree in geography and I focused on what’s called fluvial geomorphology. Which is the focus of how rivers evolve and change over time.

[00:01:00] Dana DiTomaso: And then it did another year after my degree, in ecosystem restoration. So like restoring rivers that have been damaged from human processes back to a natural state. And so that was totally going to be my career, 100%. But when I graduated, a really conservative government had just come to power in Ontario where I was living and they had cut back all the environment jobs, and so there was really no jobs out there and I couldn’t afford to like, do a free job to actually get a paid job. So I interviewed at a software company, that made environmental management software for American companies to help them report to the Environmental Protection Agency. Like, okay, it’s computers and environment. This will be a good match. Did that job for a while, ended up working at a CRM company for a couple years after that, and then I got laid off and I had six months worth of severance. And in the meantime, I’d been sort of fiddling around with making websites on the side, like, just for fun. And then when I got laid off and I had the six months of severance, I thought, well, now I can, you know, do my own thing and try to make a go of it working for myself, because I’m not a great employee.

[00:02:00] Dana DiTomaso: I’m a much better business owner than I am an employee.

Amanda Webb: Yeah, yeah.

Dana DiTomaso: So I ended up doing that and took that six months to like, learn how to do things properly, get my first client. This was in, 2000. And I, my very first client, I think I launched her website and maybe like a couple months later, she messaged me and said, how do I get higher on Google? And I said, I have no idea. I will get back to you. And so I did some Googling because Google was still like relatively new at that point. And found Rand Fishkin’s blog, which was not called Moz yet, it was like pre Moz, and ended up meeting him in 2004 at a conference, just followed the Moz stuff, and it turns out I was way better at SEO than I was making websites. So I focused entirely on SEO until about 2016, 2017, when we started getting more questions. Google Tag Manager was recently out, and I started getting lots of questions about like, how do I track this, how do I track that, and like, okay, I think it’s time to learn this stuff. And it turns out I’m way better, well not necessarily better, I think I just, I think I was getting bored of SEO is I think what it was.

[00:03:00] Dana DiTomaso: It wasn’t necessarily capturing my attention the way it used to. And analytics has been so interesting and so really it’s, my full time thing now, like I still keep up on SEO. I still do technical audits. I have a course on technical SEO on LinkedIn learning. But I would say like my day to day work is analytics with a little bit of technical SEO, just because I find that part of it fun. So that’s where I ended up. And then it was, you know, pretty calm until GA4 came out. And then GA4 just took over my life. And I can’t imagine, I feel bad for people where it’s like, Oh, something analytics is something they do off the side of their desk, because there’s no way you could keep up with everything that happens in GA4 and Tag Manager and Looker Studio. Like there’s too much and there’s too many changes and there’s changes that Google doesn’t publish. You just find out about them when you log back in. Like I went on vacation for two weeks and I came back and like There’s all this new stuff. The UI looks totally different, right? And it’s, it’s so frustrating.

[00:04:00] Dana DiTomaso: And as someone who keeps up on GA4, like, my job is to do this. I find it frustrating. I can only imagine what, you know, normal human beings whose job it is not to be in analytics all day, right? How frustrating they must find it. And so it’s, I would say a lot of my time is spent, like, explaining decisions Google’s made and, like, how analytics works and the intricacies of measurement. So. That’s where I am now.

Amanda Webb: Yeah. I think when I left the world of social media, which is where I started my whole journey, I like I was going, Oh, thankfully I’m now doing analytics well. This was many years later, but I’m kind of like, now I’m doing analytics, there’s going to be less change.

Dana DiTomaso: Yeah, right. No, no. I would say, like, I think probably the, the field that keeps up with analytics in terms of changes is, I find, PPC. Our ads team at Kickpoint is always like, oh, they changed the interface again, or this is a new change, or that new change. So I think, like, that’s a field that also changes quite a bit. Although SEO is going through a bit of a moment, too. But I would say that, for sure, the, the pace of change in some fields is faster than others.

Amanda Webb: Yeah, but that’s part of the fun for me. I know it’s a pain for other people, but it’s what keeps me interested like you said you kind of moved away from SEO and I was the same with you know a lot of that like I’ve done a lot of marketing things and finally landed on this because this is the one that keeps me interested But enough about me. That’s not what we’re here for. Let’s talk about I think we’re going to talk about audiences and really because I’ve been watching a lot of your stuff and about audiences recently. And I also think they’re so underused and misunderstood by people. So I think this would be a really good opportunity to talk about why we would, we would create audiences in GA4 the sort of audiences we create and how there are some, a few little hacks that can help us get around using things like Tag Manager. So let’s start with what our audience is. Why should we use them?

Dana DiTomaso: Yes. Yeah. Audiences are groupings of users, sessions, etc. based on characteristics that you set. So for example, an audience could be everyone who views this page or everyone who did not view this page. For example, so a common audience I’ll often set up will be people who viewed the careers page or people who’d never viewed the careers page because people who are looking for a job at your company probably aren’t going to convert and become your customer. They just want to work there. So excluding them, especially if you have like a lot of people visiting that page is probably a good idea. My classic example of audiences really thinks about one of our clients who’s a convention center and they have two different properties. So we’re managing both their websites, but they have the same basic idea where they have people who book events and then they have people who go to events. And the people who go to events are a very different audience usually than the people who book events. So when we’re thinking about things like their conversion rate, or I guess their key event rate, we’re really thinking about, you know, who are the people who would realistically convert?

Dana DiTomaso: Because if we look at the website as a whole, 60 percent of the people who come to the website are people who just want to know What time is the show going to start? How much is parking? What’s the transit to get there? Right? They’re not thinking about booking their wedding. And so what I would rather do in their analytics is segment out all these people who only care about the shows and instead have a segment of people who just are looking at those booking pages and then segmented further by people booking conferences and weddings and whatnot.

[00:07:00] Dana DiTomaso: And then when we’re looking at things like key event rates, especially from paid, we’re only considering the people who could realistically be in that audience. So it can take a conversion rate from something like 1 percent that looks really bad to something like, you know, 10 or 12%. And that can really make a difference when you’re evaluating campaigns, because , If you end up with so much noise, it means that those little changes that you make in the campaign, there’s too much other data. And you can’t tell if a campaign change is actually being effective or not. But if you segment it down to just the people who could realistically be interested in the service, then you can see, you know, are we making changes? Is this more effective in terms of driving customers? Would we make changes to the website? Is this helping drive up conversion rates? And so that’s been really good for them to just be like, here’s the audience of people going to shows and They look at them, too, but for different reasons. And then here’s the audience of people who could reasonably book events. And so I think that that’s, you know, and yeah, of course we use that for remarketing in some cases, but a lot of it is just, let’s just look at the data that’s relevant for the question we have at hand, which is not usually, how does every single person in the world use our website?

[00:08:00] Dana DiTomaso: And I think that that’s where people can really make a mistake in terms of looking at their traffic as a whole, as opposed to just the parts that really, really matter.

Amanda Webb: Okay. That makes total sense. And as so many businesses would have more than one target market or more than one product that would appeal to different groups of people. So that makes sense. That makes total sense. Are there kind of like any audiences that we should all create? Like you should definitely have an audience for this.

Dana DiTomaso: I have a standard set of events. So in my course, Analytics for Agencies, I have, I give out, like, this is the plan that we use for our own analytics. So I’m setting up analytics for people. And I’m always adding audiences to it. I actually just had a great conversation with Brittany, who’s our PPC lead on our team, about the kinds of audiences that she always wants for paid. So that goes in there too. And so whenever we’re setting up a new GA4 property or working with a new one, we have this standard set of audiences.

[00:09:00] Dana DiTomaso: So one of them is anyone who’s converted in any way. Okay. So just grouping all the conversions together into an audience, and then people who have never converted. So I really like creating positive and negative audiences, is what I call them. And in GA4, to create those, in case you’ve never created a negative audience, when you’re going to create a new audience, just remove the positive part where it says include people where just hit the trash can next to that and then click the exclude and then say people who have never you know or people who have done this exclude those people and then it’s everybody except for those people and I find that that’s also really effective so converters not converters careers pages not careers pages and then people who visited your site multiple times in a set time period especially for people who visited your site multiple times say from paid or from organic search or from social media, either paid or not paid. Because I find as well that people who visit your site a lot from organic social media are probably already fans. They might already be customers. And so you wouldn’t expect them to necessarily purchase. You might want to exclude those from any sort of conversion analysis you’re doing. Key event analysis. I’m never going to change that word. It’s

[00:10:00] Dana DiTomaso: just going to be conversions forever, right? I can never remember that. Never. Another mistake I see companies making a lot is like if you have a login on your website, if people visit the login page or they actually log in, exclude them from the audience that you evaluate when you’re thinking about conversion rate, right? You already sold to them unless you do repeat purchases, right? For e commerce, maybe it’s a repeat purchase thing. Great. But if you’re in a SaaS business, you know, they bought. They’re probably not going to buy again. So why are you using them when you evaluate your overall site conversion rate, right? Like those people are going to skew your data. So you could exclude them entirely from, you know, to have an exclusion audience and people already logged in and please use that exclusion audience and paid like, it’s not perfect, but like, don’t, don’t advertise to people who already bought the stuff. It’s annoying. So people visit a lot from different, places, people who’ve logged in, a way to exclude people. So for example, we have an audience of people who, have access courses, like the logged in parts of the courses, and then we don’t include them when we’re thinking about people who’ve bought, because we want to make sure that we’re not, yeah, we are. They already bought.

[00:11:00] Dana DiTomaso: If they buy another course, great, but like that’s not the primary purpose of we want to evaluate their engagement instead. So I would say those are like some standard audiences to set up. And then the other thing too is that create your audiences as early as possible. Audiences don’t backfill. They only fill from the moment you create them going forwards. And the other thing to check too is after you create an audience, Give it a couple days, go back in, look at the audiences list. If it says less than 10, that means that either the thing that you’ve set it for is, has not happened, or you’ve set up the audience wrong. So definitely go back and check a couple days, don’t wait like a month and be like, oh, that audience didn’t work. But if you’re having trouble figuring out what audiences are, this is where Explorations, which I talked about actually at your event, so if people haven’t watched that video, you should, but setting up Explorations is great because you can do a segment and exploration, which is a great way to test out an audience before you go ahead and create an audience, because segments are backward looking audiences essentially, and you can create audience based on segments once you have something that you’ve put together that you’re really happy with.

[00:12:00] Dana DiTomaso: So just experiment with audiences. It doesn’t hurt to try audiences. They don’t negatively impact your data. Just create stuff and see if you’re getting the kinds of people in that audience that you would expect to. One of the things I used to love about audiences or one of the things I used to do most with it was I used to use them in comparisons and then of course, like not long ago, Google rolled out comparisons, which I’m freaking delighted about instead of having like this bookmarks bar full of things that I can reactivate comparisons. They’re there, but I think people get confused now because you can create comparison. You can create an audience. So when should you create a comparison like one that sticks?

Dana DiTomaso: Yes. So you can just, for anyone that hasn’t tried them, you can just create one that’s just there for now, but you can have them. Oh, but I do want to say that if you want to create a comparison that sticks, this is where I don’t like the UI. Because if you’re on a report and you create a comparison from there, it doesn’t save. But if you go into the admin and click on comparisons and then save it there, then it does save. So keep that in mind. I feel like that’s a UI fail.

Amanda Webb: Yeah. Oh, yeah. Lots of people make that mistake. Yeah.

[00:13:00] Dana DiTomaso: Anyway, yeah. So it’s I do like the new comparisons for sure. I do wish you could save them with reports though. I feel like that’s a feature that needs to happen. Like you can save a filter with a customized report, you can’t save a comparison. My rule of thumb with comparisons versus audiences is like, is there more than one action that a person could take? Because comparisons are really just like one, dimension, this is the thing. And also, you have to create an audience if you want to look at any metrics at all, because comparisons and filters are just for dimensions. And so if people are like, what’s a dimension? What’s a metric dimension is a grouping of data like pages viewed, and a metric is how many times people viewed the pages. So if you want to say I want to look at people who’ve viewed more than five videos, or people who’ve looked at at least 10 pages, you can’t do that in a comparison, you’d have to create an audience. So that’s something to keep in mind as well is just the the type of evaluation you’re doing will determine if you’re doing a comparison or audience, but I mean, honestly, I just like creating audiences just because it’s handy to have.

[00:14:00] Dana DiTomaso: And then you can get them out into Looker Studio too, which you can’t do with segments and you can’t do with comparisons. So creating an audience is always good there. And of course, you can make a comparison, you can make an audience a comparison because I went obviously into the dimension. Yes. So, yes, that works really well. So it does save all that time that I used to have of trying to create these complex reports.

Amanda Webb: Okay. Yeah. So I know there’s some magic we can do with audiences that I totally don’t do. And that’s what I’m talking about when I watch your stuff and I get annoyed that I hadn’t thought of that myself. So the first one was I watched a video you had or it was a post, I can’t remember, but it was about, how you could use audiences to do things. That you would normally need to use Tag Manager for.

Dana DiTomaso: Yes. Yeah, that was, so that was actually a lesson that we released from my Practical GA4 course. Because a lot of people who take my Practical GA4 course have been handed GA4. And they’re told, I don’t know, you figure it out. And they don’t have Tag Manager or, like,

[00:15:00] Dana DiTomaso: Tag Manager is like, this is a step too far. Which is totally fine, not everyone has to be a Tag Manager expert. And I, and I appreciate why Google really is pushing Tag Manager hard when they pushed out GA4. But, like, Again, if Analytics is off the side of your desk, you learn GA4. Good job. You don’t have to learn Tag Manager as well. And there is a lot you can do within GA4 that doesn’t require Tag Manager. So one of the examples, that I showed in that video, there’s a few different things, but that video specifically is talking about how can you record clicks on different items without using Tag Manager. Because typically if someone says, well, I want to record a button And this button specifically, you’d have to use Tag Manager, but what if you can’t or won’t? So one of the things I recommend in the video is adding a parameter to your URL. So let’s say your website is, you know, my website, kpplaybook. com. And you want to record people clicking on the, you know, signup button, for example, for newsletters. And the one specifically that maybe is in the header. So in that case, you can add a parameter that would be, you know, question mark, Page underscore location equals top nav, for example. Make sure that

[00:16:00] Dana DiTomaso: URL will actually work. But then what happens is when you click a URL on your site, it does record in the full page URL. So this isn’t the page path dimension, this is the page URL dimension. It will record the entire URL, including that parameter that you added. And then you can make an audience of people who’ve clicked on that. URL specifically, or like viewed that URL that includes that parameter. But then the magic part is what if you want an event that’s recorded every time someone does that? Then that’s where you would use the audience, the event trigger in audiences. And you can trigger an event every time someone enters an audience. So if you make an audience that the duration of it is one day, that’s the shortest you can make an audience, then you create a trigger and say every time someone enters this audience. Record this event. Typically, I pre penned audience created events with AUD underscore so that I know where they came from. Otherwise, you’re digging around going, where is this event coming from? What is that? It’s, that’s my way of naming it. Come up with your own, but definitely have a way to delineate, like this

[00:17:00] Dana DiTomaso: is an audience created event, so you’re not wondering later where it is or document it really well. And then what you do is you also check the checkbox to readmit people to the audience. when they re enter the audience. So like, if I click the button today, and then I click it again three days from now, you’ll want to record that event twice. So make sure the duration is one day, and that means, you know, if I click it three times in one day, it’ll only record one event. Oh well, like, there’s the limitation of it. But the reality is, you know, most people probably will only click it once, and then every time they click it, even if they come back and click the same button again seven days later, they will still record a second event. So the trick is to To fire an event when they enter the audience, make the audience duration one day, and then make sure that you, fire another event, fire, I think the checkbox is called fire an additional event when the audience membership refreshes, is what it’s called, and that’s when you create the event, and that’s how you can record multiple times, yeah, and that post is on, my website at kpplaybook.com/resources, I’ll send people there. I’ll

[00:18:00] Dana DiTomaso: make sure there’s a link in the show notes. Are people engaging with X? Are people clicking X? And that’s a free lesson from my Practical GA4 course. Yeah, and I just really like creating that kind of thing based on weird questions people ask me. So like, how can I do this? And I can’t do this. And I really love making analytics solutions for those kinds of, like, really limited cases. And sometimes the answer is, like, you can’t, sorry. But I’m stubborn and I always try to find a way to make it happen.

Amanda Webb: Yeah, I love that kind of digging around that you get to do in analytics. But yeah, I haven’t made any genius, genius moments like that yet. One day, one day I’ll get there, right?

Dana DiTomaso: Well, I’ll say the other thing too that I use audiences for quite a bit is, the funnel exploration is one of my favorite explorations in GA4. And the easiest way really to make steps in a funnel is if you have events for it. So for example, it can be an audience based on just viewing a specific page or viewing a specific page and taking an action. And then if you have an event that fires when

[00:19:00] Dana DiTomaso: someone becomes a member of that audience, you can then use that event as a step in your funnel, you can even use those audience triggered events as key events slash conversions. So, for example, if your ads team, if you’re saying, well, we only want to, you know, create, have someone convert when they, you know, for example, for one of our clients who has Google ads grants, one of their major goals is people really engaging with their content. So instead of a purchase or a donation, we have another key event that’s actually people consuming their content, which is a Tag Manager recipe we have available on our website, which means that they spent long enough on the page to read the page and they got all the way to the bottom of the content. And so by recording that, we can then send that off to Google Ads, for example, and then say, have an audience of people who’ve consumed at least three pieces of content, fire an event, make that a key event, and those are people who are extra engaged, for example. So, that’s really where, thinking, I, I always tell clients, like, approach this like you’re not thinking about the limitations of the product. Tell me in an ideal world what you want, and then my job is to figure out whether or

[00:20:00] Dana DiTomaso: not that’s possible. And that’s really the best way to approach analytics. Don’t approach it from, like, what you, what you think you know, or what your limitations are in terms of like, I don’t know if we can do this, I don’t know if we can do that, to say in a perfect world, this is what I want. And then see if there’s a way to make that happen.

Amanda Webb: So getting the right questions from is, is it asking the right questions? Is it really, I want to know this. And as long as you’ve got a good reason to, cause I would quite happily jump down a rabbit hole to find anything, but really to use my time wisely, you’ve got to have a good reason for me to find out. Something.

Dana DiTomaso: Oh yeah. And I find too with clients, like, don’t say to me, well, how many sessions have we had per channel? Like, tell me what your question is, which is, what is our most popular source of traffic? For example, right? Like, phrase it, like, Jeopardy. It’s like, the answer is in the form of a question.

[00:21:00] Dana DiTomaso: And I find that this is where, this is really helpful for them to kind of take a step back from the dimensions of the metrics and just think about what is the question you’re asking of the data. Because really, especially when you’re putting together reports, reports are a story that answers these questions. And so usually when I’m making a report, everything is a question. How often does this happen? Do people who come to the site via different devices actually, you know, convert the same proportion that they come to the website? Like, those are the kinds of things we want to know. Is this audience engaging with our content? How many people bought this thing, right? Those are all questions. It’s not, you know, purchase conversions or anything like that. Like, I think we need more plain language in analytics reporting. In Google, in general, I have this phrase that I use when I’m teaching, Google de gook, because they honestly, they never phrase anything in the most simple way that you can. And it’s like, it’s really busting through that and getting into like the plain English.

Amanda Webb: Absolutely. Kind of going into a different topic, because we were talking about this before we started recording. And I think it is important, which is, how accurate is our Google Analytics data? And I’m laughing as I say that.

Dana DiTomaso: Yeah, yeah. It’s not accurate at all. I mean, how accurate it is kind of depends on your own limitations. So, for

[00:22:00] Dana DiTomaso: example, you know, you’re in Ireland. There’s this little thing called GDPR, which some people in North America are vaguely aware of, and some people think it doesn’t apply to them. It does. And, You know, if you have to have if you’re in Europe and you have consent management on your website, and most of your audience is coming from Europe or the European economic area, technically, that means that you probably don’t have a lot of people saying yes to being tracked, right? A lot of people love clicking that deny button. And if they say deny, you can’t track them, you know, and there’s this cookieless tracking in GA4 where you’re sending still data to GA4, it’s just not personally identifiable. Is that GDPR compliant? I think the jury is still out on that one. We’ll see. But even tools like Matomo, for example, which they say is GDPR compliant, they also have cookieless tracking. So that is, like, a technique that these tools use, but the problem is that, again, even if you track them cookieless, you’re still not getting the same kind of data you would if everyone said yes to being tracked. And sometimes the data is kind of crap, and I don’t necessarily want to

[00:23:00] Dana DiTomaso: include it in my reports because it’s Google kind of making it up as it goes, this Google model data. I don’t, I’m not a big fan of it. So in that case, you know, you maybe you only have half of your traffic, maybe even less. I’ve seen some cases where only 20 percent of users to a site said yes. To being tracked. And so a lot of consent management tools. My personal favorite is cookie bot, which is just, I think, rebranded user centric, but cookie bot is my favorite. And they have a chart showing you percentage of people who said yes to different kinds of tracking. And then you can see, Oh, only half of our users have said yes. So if I’m looking at my analytics and I see these numbers, that’s only half of the overall users. And even, you know, on top of that, there’s people use ad blockers, which I know that there’s this new version of Chrome coming out, which is going to make it less, easy to block. Things you know what people are going to find a way to not be tracked. And you kind of have to respect that if people say no. So it’s where even for, for example, my own website, the KP playbook website, where we sell courses, marketers love ad blockers. We can only track half of our sales. And of the half we can track.

[00:24:00] Dana DiTomaso: Half of those show up as direct because also people remove my UTMs that I put on absolutely everything. Like, I am dedicated to UTM tracking my URLs, but people remove them all the time, and then they just show up as direct. I’m like, that’s great. Thanks so much. I have no idea what’s working half the time. And so, you know, that’s kind of the reality of it. So if I looked at this data and I said, oh, well, this is 100 percent of my tracking, It’s not. I can see in our actual, like, you can’t hide that you bought something. So we can see the number of courses that people have bought and the amount that they’ve spent, but a lot of it just drops into a black hole of we have no idea where these people came from. And so people will look at that and say, well, your tracking must be broken, like and. This is inherently how the internet works these days. You can say no to being tracked. You can hide your steps. And Sometimes too, like systems don’t talk to each other. For example, if I’m on your website on my laptop and on my phone and you don’t have a login in any way, so I’m not logging into the site, I’m going to show up as two users, there’s, there used to be Google signals in GA4 that would tie that together,

[00:25:00] Dana DiTomaso: but that’s not showing in GA4 anymore, so I show up as two individual users, even though I am the same person, it doesn’t matter. It’s like, If you think about, say, a user is like, you know, your driver’s license, and then you think of your devices as like individual cars, like, the government knows you’re you, you have these two cars, but all I see is a few separate license plates and GA4. So I think that’s something really to keep in mind is like you, how people engage with you really makes a difference in terms of, you know, the relative accuracy of your data. And so what I’ve been advocating for a lot, and there’s a post that I’ve sent you that will also be in the show notes, called Marketing Analytics Data is Wrong, Can It Be Fixed? Spoiler alert, no. There’s a lot of stuff that you can do in terms of how you report that focuses more on what I call directional reporting. You know, are things moving in the correct direction? Proportionally, are we converting? So, you know, I can only track this 50 percent of people. What I want to know, you know, if this 50%, are they doing the things I want them to do? Proportionally. And then can I extrapolate that out to

[00:26:00] Dana DiTomaso: the whole? So those are the kinds of things you can do. Just like, you can only, you can’t track everything. So grab what you can, but really evaluate. As much as possible, the data that you can grab and then just let go of the stuff that you can’t because the amount of effort it would take with, say, server side tracking, which gets around ad blockers, but isn’t necessarily accessible for a lot of smaller businesses. Or, you know, you just say, screw you GDPR, I’m going to track what I want and then risk a fine. Good luck. You know, is it really worth it for that extra percentage of people you might capture? Probably not. Like, just work with what you have and spend your time pulling insights from the data you can capture.

Amanda Webb: I think that’s such smart advice because we can get obsessed by that and I know, you know, every time I start working with a client and I’ll go, your cookie consent isn’t working properly, like that needs to be fixed. And it’s just like people don’t want to know, like, and I’m based in the EU, you know, I’m based in a place where it really matters. Those fines for GDPR are not kidding

[00:27:00] Amanda Webb: around amounts of money. It’s based on gross revenue. Like, don’t, don’t mess around. You’re going to spend a lot of money. And have you ever, so here’s my question. Have you ever worked with a client who had their consent management set up properly before you started working with them? Cause I never have.

Amanda Webb: I’m kind of thinking there’s one I just did some work for and I had a pretty like hot marketing agency. So they’d actually, it was the best analytic setup I’d seen. Usually I go in and there’s 10 different. Properties, you know, three different Tag Managers all sending different things. I think that was the only one though. So, actually they did most of the work rather than me. So yeah, absolutely. I would. Like, I managed to break my own analytics testing out my, my cookie consent. So, you know, it’s, it’s not easy.

Dana DiTomaso: Oh, totally. Like, yeah, I break my own stuff all the time, for sure. Like, that’s how you learn, too, right? So, another thing to make sure you remind me. Yeah, break your own stuff, not your client’s stuff. I have, I have three GA4s running on my website. One is just like a test property that I try weird stuff

[00:28:00] Dana DiTomaso: out in. Another one’s a backup of another experiment I’m doing. And the other one is production data where I do not fuss around with that? So have a second GA4 running on your website that can push off to like and do your weird experiments there so you’re not playing around with production data. There’s no real appreciable hit in terms of speed. It’s very quick. So I would say like set up a few GA4s experiment. Just don’t touch your production data, unless you’re sure that the thing you want to do is working properly.

Amanda Webb: Yeah, that’s really good advice. I probably should have done that, of course, instead of now looking at my analytics and going, you know what, should just install it again, start back from scratch my consent. I did like totally kill like 90 percent of my traffic and all my key events when I messed around with it for a little while.

[00:29:00] Amanda Webb: So now I look at my key events and go there with that. So, yes. I should have a second property. I’m going to do that. That’s the first thing I’m going to do when we finish talking. I’m going to put a second GA4 account on there so I can’t break it. Yes. Okay. Great advice. So, it’s not accurate, but we can still use it to see trends, for example, to see, you know, are we doing better this month than we did last month, better this year than we did last year. Okay. Yep. Great advice. Yep.

Dana DiTomaso: And yeah, and I would say that post as well like something else that I talk about is one of my favorite charts. I know people make fun of pie charts as being like bad. Data design, but I really like doing two pie charts. I have a pie inside a donut. It’s like a turducken, but the donut is percentage of traffic By channel, for example, and then the inner pie is the percentage of key Key events by channel and then you look and see do the slices match up and if there’s a weird you know out of whack like you have way more organic traffic than you have organic traffic converting.

[00:30:00] Dana DiTomaso: Then that says, you know, these people are coming to the site, but they’re not converting. Where are they going? Why aren’t we converting them at the same rate? Whether we’re converting visitors from, say, paid or other channels. And that is, really, there’s no numbers on that chart. I don’t show percentages. I’m just looking at the size of the slices. And that’s the kind of chart that is Anyone can look at and immediately see if there’s issues. There’s no numbers involved. You’re not getting hung up on like, well, this says 19 form fills, but I have 22 form fills in my CRM. Your analytics is wrong. You know, those three people had ad blockers, you know, or consent management or whatever it might be. So it’s, I really like those kinds of charts because it takes away the focus from the numbers and really thinks about the proportions of the data. And if they may, if they match up. Not all pie charts are bad. I just thought I’d throw that out there. No, I was just looking at Measure Camp from, yeah, Measure Camp just happened in Chicago and there was like a whole session of people making fun of pie charts. I’m like, no, sometimes pie charts are good. If I had been there, I would have defended them.

Amanda Webb: I think it’s just a case of, and I know I used to do this as well, they look pretty, right? Cause they’re nice colors like on the screen. So like before I really got into understanding data visualization, I was the, the person that was doing them all wrong.

[00:31:00] Amanda Webb: So, but yeah, a good, a pie chart can actually be effective if it’s used well. And that one sounds cool. I’m going to check that out.

Dana DiTomaso: Yep. There’s an image of that in the, in the post as well, about the great, great. So for an example of a great pie chart, everyone have a look at that.

Amanda Webb: I’m going to move on to the quick fire questions, unless there’s anything you want to add in there. Okay. All right. Hit me. So, data, data, or data.

Dana DiTomaso: Oh, data. Data. Yeah.

Amanda Webb: Well, that’s what I’d expect. I’m, I’m data.

Dana DiTomaso: I’m data too is that a Canadian thing is because obviously you’re in the U. S. So the Canadian accent is like this weird mix of the U. K. and America. And it does vary based on where you are in the country. Like, I’m from Ontario originally. Definitely there’s different ways we pronounce stuff in Ontario than like where I live now, which is British Columbia in the far west coast. So it’s, it does depend on where you grow up. People know I’m from Ontario sometimes when I speak because of how I say things like farm or car. Apparently sounds very. more east coast than west

[00:32:00] Dana DiTomaso: coast. So, yeah, it’s, I don’t know how they say it in the States. I’m sure that’s maybe data.

Amanda Webb: That sounds very British. They say data a lot. And data is the guy who has the loves data, YouTube channel. I don’t know if you’ve ever seen him. I didn’t realize there was another way of saying it until I watched his channel. And then I was like, okay, good. Right. The second one, I almost bounced in earlier when we were talking about this. And then I thought, no, I’ll leave it to the end. It’s a quick fire question. So cookies, do you dunk them or do you delete them? So as in a cookie banner, is it a yes or a no?

Dana DiTomaso: Most of the time I just say yes to get it out of the way, but I have an ad blocker so I know it’s not doing anything. So, I am, I am the problem. I understand that. Marketers are the problem. I swear. But if it’s a company I really like, I do say yes to cookies. And sometimes if I’m gonna like buy something from an ad, I’d also say yes to cookies. I’m like, that ad company should get credit for this ad because it was a good ad. So, I really do think about it.

Amanda Webb: Yeah, I think, We are the worst. We are totally the

[00:33:00] Amanda Webb: worst for this. I actually do accept cookies most of the time. It surprised me. People who listen to this podcast are sick of me saying this because I always say, surprised me that so many people don’t. Like, people who aren’t marketers don’t either. When I started seeing my stats coming up in my, cookie blocking, cookie banner, Whatever you want to call it. Okay, third one. I’m just looking at how it’s auto corrected that for me. AI, yay or nay?

Dana DiTomaso: Yay, absolutely. I would say AI has been up to a point. So for example, I teach a course at the University of Alberta. At the beginning of the year I submit my questions for assignments into the ChatGPT just to see what it gives me, and that’s my banner to be like, okay, is a student just like, ChatGPT’d it? I don’t care if someone uses ChatGPT to help them write their stuff, for sure. And I think it can be valuable, especially, you know, if you have English as an additional language, that it can really help, like, tidy up your grammar and stuff if you’re really struggling with that. But, don’t have the ideas come from ChatGPT. It’s either to help you build an outline or to

[00:34:00] Dana DiTomaso: help tidy things up at the end. It’s not where the ideas come from. And what, what I want to see from people is like, what are your ideas? And then use ChatGPT to clean it up. Because I think of it like a calculator, you know, in math, we’re allowed to use calculators now. And the same thing for grammar, right? There’s no reason that we can’t use something like ChatGPT to help tidy up grammar. If that isn’t your strong suit, you know, like doing math. No, I’m not going to figure out like 24 times 56. I’m going to use a calculator. The same thing for like, what’s the correct way to write its use chat GPT for that sort of thing, you know,

Amanda Webb: yeah, actually, that’s really good. I hadn’t thought of the calculator analogy. But yeah, who wants to be there kind of working out like complicated math problems when we can just do it.

Dana DiTomaso: I gave a talk to a teacher’s association. They were very stressed about GPT. about ChatGPT is when it first came out and like, look, this is eventually it is going to be like the calculator and I don’t think there’s anything wrong with that. Like, you still need to teach people the basic rules of grammar and structure and everything else. But at some point, it’s like when you learn long division in school, if they still teach that, I don’t know if they do. But when you learn long division in school, you know, that was okay. Now you know how to do it.

[00:35:00] Dana DiTomaso: And now we’re going to use calculators forever. The end. Just learn the structure, but then there’s a faster way.

Amanda Webb: Okay. So true. Okay. So that was, that’s my three questions, but I do have an additional one, which is what one thing, and you’ve probably got a massive list, would you change about analytics if you could? What, maybe Google analytics, like what one thing would you like that it shows you or one thing that it doesn’t do that you would love it to do?

Dana DiTomaso: Annotations. Honestly, if I had to pick one thing, it would be annotations. And if I was talking to Krista Seiden, who people might know is used to work at Google on the analytics team. And I said to her like, what are annotations coming? She, she knows a little bit of like behind the scenes at Google. And she’s like, nobody used them. Like, that’s a lie. Everybody used them. Everybody I know uses them and misses them so much. And yeah, there’s tools you can use that will overlay it. Like one thing we do is we have a Google Sheet that overlays annotations on top of charts in Looker Studio, but like annotations in GA4 should have been there.

[00:36:00] Dana DiTomaso: And I miss them. And sometimes it’s easier just to put it in one single place instead of remembering to open the sheet. People forget about it. If they’re in analytics already looking at the impact of something, it’s so much easier to add annotations there. So hopefully, They’re coming. I mean, like, I heard a little rumor that maybe, but yeah, I, I would love to see annotations. There’s so many other things I would change, but I feel like for a quality of life update, annotations feels important.

Amanda Webb: I kind of think they’re bringing the features back a little bit at a time because we’ve got the anomaly detecting circle. Which is, like, very exciting.

Dana DiTomaso: Yeah, which is so funny, because, like, but we had that in Explorations. I don’t think I want it outside of Explorations, only because it adds a lot of visual clutter, and what I worry about with the GA4 interface is I feel like they’re adding, like, this plot row stuff and everything else. That’s helpful, but it also adds a lot of visual clutter, and I’m always an advocate for, like, make your stuff as tidy as possible, only show what’s important, and I worry that with the new stuff that they’re adding in, it’s just gonna make it, like, more and more

[00:37:00] Dana DiTomaso: cluttered and make it harder for people to parse what’s going on. So, you know, don’t go too crazy Google, but please annotations. Thank you.

Amanda Webb: And in fairness, that was one of the problems with universal analytics because they kept bolting on new stuff. It just became this like intense menu filled interface. So.

Dana DiTomaso: Oh, I can only imagine the volume of technical debt they must have had in universal analytics. Like it was based on a code base that came out in 2003. It’s it was creaky.

Amanda Webb: Great. I have, yeah, I have a long list of analytics things, but we, we could be here all day talking about that. So I am going to say thanks a million for giving us your time today. Where should people go? I know you have a few courses, so tell us about those, who they’re for and then where they should connect with you.

Dana DiTomaso: Yeah, definitely. So my website is kpplaybook. com. I’m mostly on LinkedIn. There’s only two Dana DiTomaso’s in the world and the other one got married and hyphenated her last name. So now I’m it. So it’s not hard to find me if you can spell my name right. And the two courses that I have on KP Playbook are Practical GA4, which is just GA4 in depth,

[00:38:00] Dana DiTomaso: start to finish. I assume that people know nothing and I walk them all the way through from how does analytics work anyway, generally all the way to advanced exploration work. And so there’s about six and a half hours of courses in there. So you could do it in a single day if you just sit down and watch videos. And then the other course I have is analytics for agencies. This one talks about Tag Manager. And GA4 and Looker Studio and has all of our templates and tools that we use at my agency Kickpoint when we’re delivering analytics solutions to our clients. And when I make updates, I update what’s there because I use those tools in the course when I’m doing analytics updates for my clients. So that’s, yeah, and it includes an office hours and both courses include access to a community where you can ask me literally anything. So it is like having a personal analytics helper when you’re stuck with different questions. And, yeah, I would say that that’s definitely. The place to go. And then the other thing, too, is I do have four courses on LinkedIn Learning, one of which just came out in August, the new improved Advanced GA4.

[00:39:00] Dana DiTomaso: So if you’re just getting started in GA4, you’ve watched the basic GA4 stuff, check out my Advanced GA4 course on LinkedIn Learning. And if you don’t have LinkedIn Learning Premium if you have a library card, a lot of libraries have free access to the LinkedIn Learning courses. So check through your local public library if you don’t have, you don’t have LinkedIn Premium. Or message me and then I can just share it for free for 24 hours. There’s that too.

Amanda Webb: That’s I didn’t know, I kind of feel like I did know about that, and I think Irish libraries do have it, but it comes back to the back of my head, so that’s definitely something I should check out too.

Dana DiTomaso: Yeah, and like if there’s students listening too, if, check your university libraries, well most of them also have access to LinkedIn learning courses.

Amanda Webb: Great. So thanks a million for being on the show and do check out your courses. It is on my wish list for me to do your agencies course. So hopefully by the end of the year I would have done it or started at least.

Dana DiTomaso: Well, I was going to say the other thing I did for analytics for agencies is I have, so right after we’re finished recording this, I actually have study time because I always buy

[00:40:00] Dana DiTomaso: courses and I never actually take them. So I do an hour Friday mornings where we can just hang out on Zoom and people watch the courses and I’m here if they have questions. It’s not like a dedicated office hours, but that way people actually take the course because I really want people to finish the course, not just pay for it.

Amanda Webb: You know what? That’s weird. I do exactly the same with my Analytics Ace course. We have like co working Q and A. Yeah. So it’s two hours where everyone can just sit together and work. And actually hardly ever any questions come up. It’s just that space where people dedicate to doing it. So that’s a fabulous idea.

Dana DiTomaso: Yeah, I love it. And actually last week I had someone from Switzerland who it’s the end of her day. It’s the beginning of my day for me. And she just pops some popcorn and watches analytics videos for an hour. Like it’s a movie, which is. So hilarious and cute. And I, and I love that. So yeah, , fabulous.

Amanda Webb: Great. I’m, well, yeah, like I say, that’s on my wishlist to do, so hopefully before the end of the year I’ll be in there taking that course.

Dana DiTomaso: Awesome. And I’ll see some of our listeners in there too. Thanks so much.

Amanda Webb: Mm-hmm .

Dana DiTomaso: Thank you for having me.