- So on July 1st, 2023,
Universal Analytics, which is the analytics that you know and probably don't love currently is gonna stop receiving data. Doesn't matter if you still
have the tag in your websites, it's no data's gonna go in there. (birds chirping)
(elephant trumpeting) - Today I'm very excited to
be joined by Dana DiTomaso. If you don't know who Dana is, she's an analytics expert
and co-founder of Kick Point, an agency that helps businesses
improve their marketing. Her courses include Advanced GA4, migrating from Universal to GA4 and analytics for agencies. Dana, welcome to the
show, how you doing today? - I'm good, thank you for having me. - I'm super excited that you're here 'cause we're gonna talk about something that the whole world needs
to kind of pay attention to. Which is Google Analytics
four for marketing and it's coming people.
- Yep. - And it's coming fast. But before we get into this, I would love to hear a
little bit of your backstory. How in the world did you get
into marketing and analytics? Start wherever you wanna start. - Yeah, so I started
in this field in 2000. So I've been doing this a very long time and I came into it from a weird way. So I actually have a degree in geography. I was originally going to be working in ecosystem restoration as
a fluvial geomorphologist So if people don't know what that is. Is someone who does river rehabilitation. - I have no idea. Okay.
- Yeah. I was gonna be working on rivers to bring them back to a natural state. And so after I graduated there wasn't a lot of jobs in that field because the government
that had just come to power where I was living cut
all the environment jobs. So like I don't know what I'm gonna do. So I ended up getting a
job at a software company and that led to a job at
another software company. And then when I got laid
off from that second job, they gave me really nice
separates and I thought, I've always been interested in websites, I'm gonna learn how to make websites. And so that's what I started doing. And then my very first client, 'cause I was freelancing,
when I launched their website, she said to me, "How do I get on Google?" And I said, "I don't know,
I'll get back to you." This is 2000. And so I did some searching and I figured out that
I actually liked SEO a whole lot more than I
liked making websites. So I did strictly SEO for a long time. And then I got interested in
analytics just because SEO is, It wasn't really capturing
my attention so much anymore. But also, as an agency owner clients you're asking for results and they want you to prove the value for the money that they
give you every month. And so it was a matter
of digging into analytics was really interesting
and now it is something that I think really
separates good agencies from poor agencies. It's not just having analytics but like what you do with that data. And so that's led to where we are now. - Okay, so around the two thousands you were developing websites
and fascinatingly enough I started in the nineties, the nineties and ultimately did have a web development creative agency. And I remember when we
used to have to code HTML and then these visual editors. - Did you use Dreamweaver? - I did use Dreamweaver, yeah. - Yeah, me too. - Yep, yep. And Cathy Wyman's book
or whatever, I can't. - Yeah, I have that one. Yeah, it was great. Yep.
- Yep. So along the line you became a search engine optimization
expert if you will, but somewhere along the line
you pivoted into analytics. So what year was it approximately that you started focusing
more on the analytics and was it Google Analytics at the time or was it something else? - Yeah, yeah it was actually Tag Manager that got me into analytics. So I think it was 2015 or 2016 and I said, "You know what? I'm just gonna have to sit down and learn this Google Tag Manager thing." I've been putting it off 'cause I didn't wanna learn programming. So I thought it was
quite programming heavy and I guess the first version of it was more than the version that we have now, but I'm just gonna learn this. And then as I learned more
of the things you can do with Tag Manager and the stuff
that we weren't capturing that we could capture, then I
got really excited about it. And so that that led to
getting into analytics 'cause you gotta put the data
somewhere and. (chuckles) - So let's talk about
like the last year or so. Let's bring it up to the present. So somewhere along the line you get into doing courses and- - Yep. - Tell us a little bit more about what's been going on for
you in the last year or so. - Yeah, so LinkedIn Learning approached me just over a year ago actually and said, "Would you be interested
in making courses? You're recommended by someone." And so the first course
I did was technical SEO since that's my background. And then I said to them, "A really good course to do would be transitioning
from Universal to GA4 'cause I think this is
gonna be a big deal." And at the time I recorded that they hadn't actually announced the sunset date for
Universal Analytics yet. It was just at some point
you might wanna move. And now that they've announced the sunset and there always seems to be
a Twitter poll, it's like, "Hey do you think Google's
gonna move the date?" They're not gonna move the date everybody, I'm really sorry but like get
your stuff off of Universal and put it in GA4. And so that's been a lot of our work now is really helping people, getting them to understand the differences between Universal and GA4 and building a lot of reports
in Data Studio actually, or pardon me, it's now
called Looker Studio. I can't possibly bring myself to say that it's still a Google Data Studio to me, but it's building a lot of reports 'cause the interface in GA4 sucks. So it's just for clients- - What was it like? Because you're the only person I know that actually has worked with LinkedIn, it's called LinkedIn Learning. What's it been like working with them? How's that been for you? - They are awesome to work with. I really like their team. They really understand
how to create courses, it's what they do. A lot of the people who work there, the original linda.com
before LinkedIn bought them and because they
understand adult education and how to structure a course really well, it means you have a
really like tight schedule but at the same time it's very manageable 'cause they have a rule about, a course should be about an hour long. That means you're gonna
have five minute videos, you're gonna have X number of videos and then some courses are longer. I think my advanced GA4 is
maybe an hour and a half because there's just so much to cover. But that way it's easily
digestible for people and they really get, what you're gonna fit
into those five minutes. - So tell us just a little
bit about Kick Point, like who do you service, what do you help them do specifically? - Yeah, so we work with
clients of really all sizes from tiny places. We have a Fortune 100 client as well, but what we mostly work
with are in-house teams. So we work with a lot of in-house teams to teach them how to do this stuff 'cause it's not rocket science, right? It's marketing. So a lot of, especially
SCL people are like, "Oh no, you can't do it." Or there's some sort of
like special dark magic that goes into it or something. It's really not. And so often from an agency perspective, it's cheaper for some clients
to do some things themselves than it is to pay an agency to do it. So for example, we'll teach our clients, how to do a better job
of managing Google Ads or Facebook ads, how to keep up on SEO and then we're always in touch with them. So something changes like GA4, we'll email and say, "Hey this is coming, we can set it up for you. PS you need to know about
this, this, and this." And so because we're a small agency and we didn't wanna stay relatively small, we don't wanna have retainer clients who are just there forever. So we work with much shorter
timeframes for clients and figure out over timestamp, doing the same thing every month forever. Let's do this for six months. Okay, now we're gonna do this
other thing for six months. - Well first of all, thank
you for sharing your story. I think it's really useful for people to understand your background and the fact that you
really heavily focused for quite a few years here On
the analytics side of things, there are a lot of marketers out there, including many of them that
work for my company (chuckles) that are like burying their
head in the sand right now and do not understand the urgency here of why there's gonna be this big change from what's known as Universal Analytics, which is Google Analytics
to the rest of us into this new GA4. So I would love you to give the case to everyone who's listening, why they or their boss
or someone on their team or their clients really
need to pay attention about what's gonna happen here. The changes are coming in
the world of analytics. - So on July 1st, 2023,
Universal Analytics, which is the analytics that you know and probably don't love currently is gonna stop receiving data. Doesn't matter if you still
have the tag in your websites, no data's gonna go in there. So you basically will have
no idea what's going on. You have to be using GA4 by that point and then by the end of 2023, although that date might get
pushed a little bit further, Universal Analytics data
will be gone, like gone. You won't be able to access it anymore, which means you're
gonna have to get it out of Universal Analytics
into a tool like Big Query, which is Google's database
software for example. I don't have a lot of good tools yet to export from Universal to Big Query. I think those will come
out closer to the actual, we need to get our data
off of your deadline or Google might provide a tool as well, but that's what's gonna happen. And then with GA4, you have to have it set
up because the problem is even if you only set it up today, you're not gonna have
year over year data in GA4 at this point, right? So if you're past July, you're not gonna have year over year data, which means you're gonna have
to compare year over year with Universal, which means you're probably
gonna have to learn a tool like Data Studio in order
to do that comparison. So you're already a bit behind the ball and you don't have to do
anything fancy with GA4. Like there's lots of things, things that we talk about setting up, but the basics is just get
the GA4 code on the website, worry about enhancing it later, but just having things like sources and channels and page views. Like just get the code on there and just worry about the rest of it later. - Okay, so just to summarize
what you said in July, probably some July- - July 1st.
- July 1st, 2023, which is coming very fast,
Google analytics as we know it, which is also known as
Universal analytics, will no longer work and by December, the end of the year it
will essentially disappear. - Yep.
- Now, the bigger question that a lot of people are
asking themselves is, "Why the heck is Google doing this?" Can you try to address the reason why this is coming to the end of life? - I mean Google's obviously
not saying this directly, but it's because of privacy issues, particularly regarding GDPR. Google's in a lot of crap, simply with the European data authorities and there's always these
posts that are put out saying, "Google Analytics is banned
in Italy" or whatever it is, it's not banned, there's
security problems with it. And one of the security
problems, and I can't quite, I'm not a privacy expert so I can't get into all
the intricacies of it. And this is not a podcast on GDPR, although that might be useful. The issue is that when you go to Google, when you go to a website that
has Google Analytics on it, the information is sent to the states if you're a European citizen. And that is against the rules essentially, because any data that's in the states can be accessed by American authorities for security purposes and
that's against EU rules. So GA4 has much better data
controls and so that's why they're not gonna push the
date of Universal Analytics. They have to get people off to GA4 and I mean GA4 is a
better product in general. It does some things a lot
better than Universal does. I like it. I understand why they're pushing people and the timeline is too soon, but you will like it
eventually, (chuckles) just not right know. - That sounds like something that a totalitarian state would say. - Right?, it feels a little
bit like Stockholm syndrome, like, eventually you'll be
like, "This is great," it's not. - Yes, yes, we'll like it. - It is great. It just looks horrible. I wish they'd work more
on the user interface side of things. - Do you feel like beyond
GDPR there's also lack of data at play here a little bit. And this is part of what, maybe this is because there's
so many different apps and software companies
like Apple for example, that are essentially
blocking data flow, right? So do you also feel like maybe the data that's coming into Google Analytics, Universal Analytics is
becoming less and less and as a result of it it's
becoming less accurate and therefore there's going
need to be some modeling and AI stuff that maybe
the old one can't do? - Oh yeah, totally. And there's machine
learning in GA4 for sure and that's been built in for a while. And that's part of their
like it's so great. I mean the machine
learning learning is fine and it's neither here nor there, it's just a fact of marketing technology, in the year 2022 and beyond. There's gonna be machine
learning involved. But I think the biggest thing is that it is those privacy rules
and it's that storage of data that is the big concern. And I mean the reason why GA4
started in the first place was actually to bring together
app data with web data in a single place. So you could track users
across multiple platforms and now it's the default
web analytics tool. But in terms of actually tracking people and like looking at
those numbers and saying, "Are they accurate?" They haven't been
accurate for a long time. You can't track everything that happens. About 20% of the internet
uses an ad blocker. Those people are never gonna show up. Other people have privacy
rules as you mentioned Apple, although hilariously they
have their own privacy rules, but now they're launching
their own ad network. So clearly they were just
like kicking everyone else off so they could get the money. - But even browser, and browsers are blocking data too, right? - Oh yeah, totally. - FireFox, Brave, that
kind of stuff, right? - Yeah, so if you wanna for example, look at attribution and say, "Where did my most purchases come from? Is it organic, is it paid?" Then you can only really look at people who use Chrome browsers
and trust that data as much as you can. People on Firefox don't trust it. People on Safari don't trust it 'cause the cookie's clear every single day as far as your website is concerned. You are a brand new
person every single day if you use Safari, even if you're not. So if you have a lot of Safari
visitors on your website and you're looking at
those attribution reports and treating them like they're true, you're incorrect unfortunately. - Yeah and that's on the
mobile side of things, that's a huge audience, right? Because most people on their iPhone, they don't bother to install Chrome, they just use the native install app. - Because why would you? - Right, exactly. - Yeah.
- Okay. So we now understand a
little bit more about why the move to GA4. It's all sorts of things
we just talked about. What can Google analytics four, what is it solving for, what can it do? Let's talk about the benefits
if you will, of utilizing it. Things maybe that the
old version cannot do. - Yeah, it can track a lot more things than the original one could or you could track things there but it was much harder
to get the data out. So in the old, and I say old, the existing analytics
Universal, in my mind it's old, I have to do everything GA4 now. So I don't mess myself up but it's like translating
between two languages, it's like do I speak English or French? So in the old GA, the Universal GA, there were things called scopes. So something for example, there was a session scope which would be, I came to the website and I came to the website
through organic and I converted. Those are all session scope metrics and then there's page scope metrics which is I visited this page, I scrolled 25% of the way down this page, I played this video and those
are like hit level metrics is what those were called. And you couldn't mix the two so you couldn't, for example say, "I wanna know what pages someone looked at on their way to convert" without doing a whole bunch of data blends and stuff like that. It wasn't something you
could get at by default. So now in GA4 everything is a hit. So there's no separate
page view event type, there isn't a transaction event type, a transaction is a hit,
a page view is a hit, somebody scrolling 90% of the
way down the page is a hit, somebody hitting play on a YouTube video on your website is a hit. And so because everything is a hit, it's a much more level playing field in terms of how you can
mix your data together. You can also record way more stuff. So in Universal you have events, you have the category, action, label, those are the three
things you could record. Now you can have an event
like playing a video, you can have up to 50
individual parameters associated with that event. So for example, for one of our clients, they have a a form that
people will fill out when they wanna purchase their services, in that form they have several
dropdowns of information like how big is your company? Which location is closest to you? What industry are you in? So if we were gonna refer that Universal, we would have like a
super long action or label that'd be like all this
stuff mashed together. But in GA4 we can have a field for how big is your company? What office do you wanna talk to? And then we can create
a really nice dashboard for them where they can slice and dice these different pieces
of data together to say, "How big are the companies who contact the Manchester
office," for example. And so that ability to
capture so much more data I think is the biggest mind
shift between Universal and GA4. 'Cause you're so used to
being limited in Universal and now it's like just forget
what you know about Universal, think about, in a perfect world, what would you like to track? And that is something that
is a lot more doable in GA4. - Okay so I heard two things. I heard that you called everything a hit, which really means a hit means like a piece of action probably, right? And you can record more
parameters for example. So one of the things that I wanna ask is, some people that aren't super technical may not understand that Tag
Manager and Universal Analytics typically play very strongly together. Do you need that strong integration in GA4 as well to do these kinds of things that you're talking about? - You can run GA4
directly in your website. Google does recommend you
run it through Tag Manager. But something that's nice about GA4 is events that you would have
to record through Tag Manager. Like I mentioned video views for example, if you were gonna have those
events show up in Universal, you would have to set up
a whole bunch of things in Tag Manager to make that happen. It's built in now. Even with search on your
website that's built in, people downloading a file
like a PDF that's built in, 90% scroll depth is built in. So there's a bunch of stuff that you had to do in Tag Manager that you don't have to do there anymore. But because we can capture so much more, actually our containers in
Tag Manager are bigger now, they work, it's like, "We can record all this
stuff, let's do it." Things like, how did they
navigate to this page? Did they use a menu, did
they use a body link? Was it a footer link? Like I never would've
recorded that Universal 'cause it would be too
difficult to get the data out in an easily reportable format. But in GA4 because it's so flat, it's so much easier to get that out and better understand how
people are using websites. - Okay so the big
advantage that I'm hearing is really the fact that you
can record a lot more things and you can track a lot more things. That's the big advantage. Is there any other
major advantages to GA4? Does it make it easier to do reporting or anything like that? Or is it really just
all about the data input that's really the key advantage? - Yeah, reporting is not
great in GA4 currently, there's a section called
Explore it's okay. The real thing to do is, and this again if you don't feel like you're a technical person, first of all don't tell yourself
you're a technical person. Everyone can be a technical person. I have a degree in geography, come on. So if I can do this, you can do this. So one of the things that I
always think to say to people is make sure to export
your data into Big Query. There's a free export in GA4 where you set up a Big Query database, you push the data over to there because there is this thing
in GA4 called Thresholds. So if they feel for privacy reasons that you're capturing personal information or like information,
only one person did this and you'd be able to identify that person based on that action that they took. They don't show it in the default reports but everything makes it out to Big Query. So make sure to set up a Big
Query connection right away and there's a help article, it is fairly straightforward to do it. I recommend doing that just because, and then you have control over your data as well, which is amazing. And something again that
we didn't necessarily have with Universal unless you
wanted to pay a bunch of cash. The other thing. - The question on the Big Query is this some sort of a Tag Manager thing? - It's a database tool. - I meant the ability to
connect the things together. - Oh yeah, no it's in
the settings for GA4, you click on the Big
Query option and admin and then it's you just walk through the- - So one time setting and then even if you never use Big
Query, you've got it someday. - You've got it. - What I'm supposing you say. And the other thing is
you were about to say. - Oh audiences, audiences are huge in GA4. So in Universal, and if
you've ever used Google Ads, there's audiences that
you can set up and say, "I wanna show this ad to
people who view this page or people who went on my
website and didn't buy." But in GA4 you can actually
have time based audiences. So instead of saying
people who came on my site and didn't buy, you can say, "I want people who came on my website, watched two videos over
the course of a week, or people who came on my website, watched a video then in that same session added to the cart or
came back a week later and added to cart." And there's lots of different layering and things that you can do with audiences and you can create events based on someone entering your audience. So for example, for one
client they have a login link, they're a SAS company so it's software that you can log into. So we wanna make sure that
we're only thinking about the conversion rate for people
who haven't signed up yet because we don't care if you've signed up, of course you're not gonna sign up again, you already gave them your money. So we have an event that
puts people in an audience. So when you click the sign
in link on the website, we tag them as being a current customer. We send off an event and then we just say, "Ignore everybody who has this
event of current customer" 'cause we don't care about them when we're thinking about
that conversion rate, that kind of thing was
difficult and a pain in the butt to do in Universal, it's so much easier now and
the targeting too for ads is gonna be significantly better. - Okay, let's talk about
engagement a little bit and how it's different and why it matters because when we were prepping for this, you were mentioning to me
that there is some differences between how engagement
is between these two Universal versus GA4. - Yeah, so I mean a metric
that people report on from Universal and you shouldn't, so stop it now is bounce rate. And the reason why I say like
don't report on bounce rate. So bounce rate in case you
don't know what bounce rate was, you just saw it and you're like, "This is people who came on my site and then didn't do
anything and then left." Possibly it also probably was people who came on your website, spent 20 minutes reading your stuff, maybe watched a couple videos, but because they didn't
go to a second page because you weren't recording events when they watched your videos, they could have filled
out a form on your page. If they didn't go to a second page and you weren't recording an event when they filled out the form,
they would still be a bounce. So your bounce rate is
meaningless and when people say, "Oh well I want my
bounce rate to be lower," like we can make it zero today, I'm manipulating the data in Tag Manager bounce rate is meaningless. - So that is no longer a
metric in the Universal. - So there is a metric in
GA4 called bounce rate, but it is not anything
related to the bounce rate in Universal and I wish they
called it something else. What you'll see in GA4 by default is a metric called engagement rate. So engagement rate by default is if somebody had your website
open in their active tab. So all you people who are tab hoarders, like not in one of those tabs you've had around for six years, right? An active tab for at least
10 seconds by default you can up that to at least 60, 10 seconds is a pretty short time. I would say you probably
wanna up that based on what you're doing. So for example, like for your website, Mike for example, you want
people to read, right? So you would probably put up
for the longer engagement rate or if someone converted or if they visited two pages on the site. That's the engagement rate. I think it's way more effective. And then bounce rate is actually just the
negative engagement rate and what's gonna happen, I
know what's gonna happen, people are gonna look at that and say, "My bounce rate doesn't match
between Universal and GA4. It's not measuring things correctly," it's like apples to oranges. They are not the same thing. It's like apples to like leaks. Like not even in the same universe in terms of what they're measuring. - So on the engagement rate, what I'm hearing you say is it defaults to 10 seconds, is that right? And you can change that in like a master setting kind of thing where you can have it be longer and then somehow you can look
at the audiences for example that are more engaged versus less engaged. Is that kind of where you could
go with something like this? - Yeah, so for example
like display traffic, I mean people spend money on
display and they're likely, people have really low expectations for display advertising in general 'cause it's usually trash. But if you have for example, like one ad network
you're paying money for and you see that everyone who comes in has an extremely low engagement rate, that's a sign that's maybe bots or it's people who miss clicked. I see this a lot on like
traffic from newspaper advertising networks and it's like people who are using the newspaper
app on their phone and they didn't mean to tap on your ad and they hit back right away. Those people would not come up as engaged. So it's something where I
think that that part of it is really useful but
it's more a diagnostic than an actual like reporting tool. Don't use it as something you're gonna measure success against, success is the things
that make your money, bring your organization
closer to its goals. Not everyone, making money
is not necessarily the goal of every single website out there, but you do have a goal and engagement rate is probably not that goal unless you're like
selling ads like Buzzfeed. - What about time on site? Is that still a metric
that's in GA4 because. - It's better for sure. There is an engagement
time that's measured, but you do have to keep triggering
events for it to happen. So it's like if you just
sit there on the page, even if you're reading
it, you scroll down, you don't necessarily get to 90%, you may not have an accurate time on site. So something that we've actually built is a metric called content consumption. And so in content consumption it's run either as a WordPress plugin or through Google Tag Manager. And the way you set it up is you say, "This is my container of content." So you get on the website,
here's an article, this is the container for the article. When it loads up it says this article will take 10 minutes to read. This other article will
take five minutes to read. This one will take two minutes to read. It starts a timer. And then when you've
hit the amount of time it would take to read and you've had the tab open for that long, it says "Yes, this person
dwelled long enough on the page to read it." The second metric is it looks for the bottom of that container. So you'd say the footer is the end and then once the footer
enters that user's view port, it's what it's called when it
gets into your browser window, then it fires a second event to say that they scrolled down far enough. If both those things are true, then the content was consumed. And if none of them are true, it means that they just abandoned ship. If they dwelled long
enough but didn't scroll, then they hoarded the tap for later. If they scroll down but they
didn't spend long enough, then they skimmed through the article. And these kinds of behavior metrics can tell you how people are
engaging with your content 'cause you can know if you've got a video, you know the people played it, you know how long they
read, they watched it for, you know where they stopped. But when it comes to content you don't necessarily know that. And so that's where content
consumption really helps. And there will be a link
in the notes as well that will take you to the blog post, which walks you through
exactly how to install it step by step. - Okay, so just so we're clear, this is some sort of a
custom Tag Manager thing that you built that's
kind of open to the world. - And a WordPress plugin, yep. - And a WordPress plug, oh either one. - Yeah, so if you have
WordPress site, use the plugin. If you don't then you can use
the Tag Manager container. - Got it. - It works on any site,
the Tag Manager container. - And does it send some
new information into GA4 that might not normally be in there? - It is how that works. It's like this is brand new information. - Got it, okay cool. So alright let's talk about
audiences a little bit. I know we've mentioned
it a little bit already, but how are they different
than, is there even an audience? There is some sort of audience's thing, I don't know if they call
it audiences in Universal, but what's different
about audiences with GA4 versus Universal and what do we need- - Yeah, so the big thing is
that audiences in Universal were really just used for ads. You didn't necessarily do a lot with them. Instead you would use
something called Segment. So if you wanted to say show me people who are just organic
traffic, you'd use a segment. So when G4 audiences are like segments and audiences now stronger and (chuckles) like the more improved version and so you can use them
for ads absolutely. But because audiences can contain really whatever you put in them, you can use them in
the place of a segment. And as I mentioned earlier, you can actually trigger an event when someone enters an audience. So for example, as I mentioned the example
like when someone logs in or clicks a login button under your site, you can say, oh they're a current customer and then you can remove
them from consideration in your conversion reports. Another example is a client of ours who's a convention center
obviously had a couple slow years, but one of the things
that we measure for them is they have two distinct audiences that come to their website. It's people who are going to shows who wanna know things like when
is the band going on stage? And it's people who wanna book events. There's more people visiting the site who want to go to shows than
people who wanna book events. So if you only looked at the
site wide conversion rate, it would look horrible, because it's those
people aren't converting, they just wanna know when
their band's going on. So instead we've created audiences of specifically like wedding
planners, meeting planners, convention planners based on the pages that you viewed on the site and what you haven't viewed on the site. And then we're only
showing the conversion rate for wedding planners, convention
planner, meeting planners. And that really helps them know, this ad was really great
for wedding planners, this ad was not so great,
we're gonna shut this one off. It really removes the noise. - So one of the questions
that comes to my mind like this feels a lot
like the email software that we use where we can assign tags based on pages people hit, and then we can kind of say like, "Here's someone who's
interested in our conference for example, because they
visited our conference page. Or here's someone who's a customer. These audiences that we create, it sounds like they're not
automatically created for us. We have to create them ourselves. - Yes. And is this something we can do in GA4 or do we have to use
Tag Manager to do this? Like help us understand how the logic of that
all works a little bit. - Yeah, so there's an audience's section in the configuration and then
you just set up the audiences. The audiences only start populating from the moment you create them. They're not gonna go back in time. So make sure to set them up
as soon as you think of them. And you can't edit an
audience once it's published. So make sure you actually want to, you can edit the name of it but
you can't edit what it does. So, but you can always
create more audiences later if you messed up something, in more than once I've
published and be like, "Oh I forgot to click that button" and then you have to delete
it and start over again. But I think what's useful about them is really thinking about, what do you consider for your buckets for people who come to your websites? You mentioned for example,
your conference, right? People interested in the conference, you know that's gonna
be an audience, right? You're gonna have different topics. So for example, people
who have visited like more than five articles
specifically about Facebook, that's a Facebook audience, Twitter audience, right? And people can also be members
of more than one audience. So you can really start
to create audiences based on specific topics
and then you could say, "You know what, we're only
gonna show remarketing for people when we publish
a new article about Facebook to people who've already shown
an interest in Facebook." - Well and for those that are
on the organic side of things, I also could think creatively about how you could try to get
some demographic data on what these audiences might look like. For example, let's say we've got a bunch of
different products, right? And we have different audiences
based on different products or just the example you mentioned, like people that read articles that contain the word
Facebook and the URL, they have an interest in Facebook. We could discern whether or not people who are interested in Facebook tend to go visit our social
media marketing world page and convert into a customer, which will help us maybe in
some of our organic marketing to produce more Facebook
articles for example, right? Is that the kind of thing we could do with something like this? - Yeah, for sure. So we have one client who wants to know what pages people start on on the site that are more likely for them to lead to another section of the site. And so I can't say the names 'cause you'd know who the client is but, - I get it. - For them it's, we've created an audience of just everybody. And then we have an audience of people who specifically visited one of these target pages. And then we say, so show
me all the landing pages where people start on the site and then show me the proportion of people who end up doing this thing
that we want them to do. Where do they start? And there's some really clear patterns. Like this blog post is terrible at driving people to that section, this blog post is fantastic. So if you're gonna market
stuff to that audience, use this post, don't use that post. - In the past, oh by the way, is there a limit to how
many of these we can have? How many of these audiences? - That's a good question. I don't think so, but I would. - Okay. - I mean there has to be eventually. I'm sure there is somewhere
it's just like large. - The way I think of this is, the way I think about how we do things in Universal analytics
with UTM parameters, right? It's kind of the same except
it sounds more powerful because with UTM parameters we've got to send them there
in the first place, right? In order for those UTMs to trigger. Now we can track anybody
that hits those pages. And I would imagine, can we create even
audiences within audiences? For example, anybody who hits social media marketing world pages more than X number of times can
be a high interest audience. Can you do that kind of stuff? - Yeah, and if you fire an event when someone enters that audience, then you could create a new audience based on people with that
event plus something else. So it could just be like
audiences all the way down. - And do you need, in
order to fire events, can you do that outside? Do you have to have Google Tag Manager to be able to fire events or. - No, you can fire events specifically from someone
joining an audience. So you can say, anyone who's visited the conference page in the last 30 days and then fire an event
saying calling, "Audi," I usually start my audience
events with "Audi_," and then conference page for example. And then you can also say, then another event when
this person refreshes. So if they come back in the 31st day and go to the conference page again, you fire a second event. - Can you actually say if
they do not take action within a certain period of time to remove them from the audience. Oh really? - You don't remove them from the audience. - Okay, they're always in
it once they trigger it. - Yeah, but then you could see like, are they taking action, the
people in this audience. - Fascinating. I wanna ask you a little
bit about UTM parameters. I know this wasn't part of our
original line of questions, but so many marketers are so heavily reliant on UTM parameters for nearly all of their
organic and paid marketing to try to track where the
heck the customer came from or the prospect came from. Does all of that still work in GA4? Does some of it change? Can you talk to us a
little bit about that? - It's changed, yeah, so it works. Don't freak out. (chuckles) It still does work. There's some new channels have been added. One of them is paid social, yay. So the downside is you can't actually create your
own channels anymore in GA4. Maybe eventually you'll be able to, but right now you can't
create your own channels. There are new sources, sorry, new mediums, pardon me, that you have. So if you Google GA4
channel configuration, then you'll see the list
that Google's put together. And really it's based on
what URL they're coming from. But there's video now you
can have an audio UTM. So for example, you'll
notice in the document that is linked to as
part of the presentation, when you click link to
come to our website, you'll see the UTM contains audio. And so that's podcast stuff for example. So there's so many more
things you can do there. I think there's paid shopping, there's video as a
channel now, there's just, there's so many more
things that you can track. - So I want some clarifying like the source campaign and
medium is what we use in Gen- - And you'll still use those. Nothing changes there, it's just you have more
options that will show up as specific channels now in GA4 as opposed to just going
in that other work. Good traffic goes to. - Yeah, and channels, I guess I didn't even think about that. Channels are like the natural tracking of Facebook traffic and Google traffic. - Yeah, so a channel, is it taking the medium and
turning it into something nicer? So channels entirely based on the medium, the source doesn't actually
matter in some cases, but if you look at that list, there are a couple sources that matter to differentiate between different shopping networks for example. But go take a look at the list 'cause it is quite complete and well, and so what I have is a spreadsheet that we use for our clients
where we actually listed out, here's the mediums you're allowed to use, here's the sources you're allowed to use, here's how you should format campaigns. And that way there's good parameters. So they don't scrub the traffic too much. - Okay so to be crystal clear, we may need to change our
source campaign in mediums because what I'm hearing you say is you can't just put anything in there. You have to follow specific
guidelines or is that. - Yeah, so in Universal you'd have that channel called other, it was just like Google
doesn't know what's going on so they're putting it in other. - Right, right. - That name in GA4 is unassigned. So it does still have a
bucket for miscellaneous. You should never really have
anything in there though. You should always try to
have stuff make sense. The only one I think is
really missing from GA4 and where we're telling clients, you know what, just
put it in here for now. We'll sort it out later is offline. So for example, our local clients who use flyers to
advertise their services, they wanna have a URL in there. We wanna make sure we know if
people are coming to the site from this offline flyer,
there's no offline channel. So we're telling them
just use an offline medium and then we'll sort it out later. At least that way it's being tracked. - But generally speaking, we probably shouldn't
have to change too much about the way we're sending data for source campaign in Medium. - And there's gonna be some
new UTM parameters coming, but I don't think they're in use yet. But you should also see them when you take a look at that
channel configuration as well. - Awesome, all right. We talked a little bit about reporting. I wanna dig a little bit deeper. The native reporting inside of GA4 you said is not very good. Does it have the kind of reporting people are used to in Universal Analytics or is everything gone and you're just not gonna
be able to find anything? - It's just really basic. Like if you wanna know any
people came to the website from organic, sure. If you wanna know literally
anything more than that, it's hard to get at. (chuckles) - Oh wow, okay, so wait a minute. 'Cause like for example, let's just take our sales page, right? For our social media marketing world, like normally we'll go there and we'll look at the campaign parameters, you know what I mean, and try to determine where everybody was, all the traffic was coming from. And we'll also look at maybe
like if we're doing paid, we'll look at whether or
not those conversions came. Can we still do that
kind of stuff or is that- - You have to create an
exploration to do that. And what I hate about exploration, explorations are kind of like
the do it yourself reports, but the problem with explorations, you can't control the
width of the columns. - Oh wow. - Sometimes it's really hard to read. And so I mean I go over them
in my course on advanced GA4, I also complained about the column with quite a bit I think in those videos. So instead I think
you're gonna have to use Looker Studio slash Google Data Studio, which is a really important
tool I think to pick up if you haven't been using it. Just so I mentioned earlier, get your data in Big Query, then use Big Query as your data source to run reports in Looker Studio. So the nice thing about using Data Studio is that when you go in and if you wanna make changes to the data, so earlier I mentioned, you're gonna have an offline medium that's gonna go into unassigned in GA4. Well, in Data Studio you can
actually change it to make, you can make your own channels, you can re slice the
data in different ways and it doesn't touch the
original data source. It's like a presentation layer
that sits on top of the data. So you never harm your original data by going in there and doing this stuff. And it's a really powerful reporting tool because you can do so much with it. - Okay, so back when it
wasn't called Looker Studio, when it was called Data Studio, I remember you could
just one click integrate into Universal Analytics.
- Yep. - Can you not do that with GA4? - Oh yeah, you can. It's just the data that you get is gonna have that same thresholding that I mentioned earlier, so won't necessarily be totally complete. So that's why if you have it in Big Query, you use Big Query as your data source because you'll actually
have complete data. But if you just wanna run something quick and you don't need it to be, you're looking for trends, not absolutes, which is what I always
recommend anyway about data, then you can use GA4 as a data source. That is just fine. - Okay, let's spend a little bit of time talking about Looker Studio. Other than changing the name, I remember that it was
kinda like a blank canvas and you can drag elements in there and you could have
dropdowns and date ranges and create cool little graphs and charts. Has that all remained or
have they completely changed? - No, it just, it looks exactly the same. The only thing that's different is the completely ridiculous name. - And can you easily share that with staff members and stuff? Because goodness about Universal Analytics was that everybody could, well yeah, you'd have to share that too. - You'd have to share it. - Yeah, there were
customer reports in uni. Can you, okay, so there's customer reports in Universal Analytics and it sounds like they have a new name for it. You just called it
Explorations or something. - Explore, yeah, yeah. - And there's a limit to what you can do because of the tables
is what you're saying? - And you can share them but
not well and it's not great. But you can share in Looker Studio, it's way easier to share stuff and you can even create scheduled emails. So some clients like to
get an email once a week with their dashboard link for example. It's pretty simple. - Now if everybody listening right now wants to start kind of using GA4, what do you recommend the steps be? Obviously I think you
mentioned at the beginning you gotta go ahead and I
think there's just a button inside of Universal Analytics
to start collecting data. - Yeah, but I wouldn't use that. - Oh really? - That's tied, I don't know
how Google's gonna handle this. If you click that button, it's tied to your Universal
Analytics property. So what's gonna happen on July 1st when Universal Analytics shuts off? I don't know. I mean I hope they'll do something but knowing Google they won't,
they'll forget about it. So 'cause I mean we all, we've worked with Google for years, right? They do stuff all the time
and they don't necessarily, customer service is
never their top priority. So what I would suggest
is go into analytics, go into your analytics account and then create a new property. It'll be, when you go into the admin, there's a column called account, there's a column called property and there's a column called View, which you'll see only
in Universal Analytics. There's no views in GA4. And then at the top of the middle column you'll see a button that
says Create Property, Click that create a GA4
property, go through the steps. At the end of that you'll get
what's called a measurement ID and it's a G dash followed by a bunch of
letters and numbers. That is the same as that
UA dash number situation that you had from Universal. That's your measurement ID. And then that's what you
would put in Tag Manager or in some sites, you put
it directly on the site so that, that is your
new ID for that stuff. - Okay. There's still a lot of
people who are obviously like eyes rolling back a little bit because this sounds so darn complex and like you said earlier, you will eventually love it. What have you been able to, is there stuff that you've
been able to discover for your clients or for your own business that you just frankly could
never have known before by using GA4? - Oh yeah, so one of the things that we started recording is it's so much easier for us to report on content consumption now. We had to do some pretty wild
creative stuff in Data Studio to get the data out in a useful format. So now reporting on it is so much simpler. I can't believe how quickly we
can report on this stuff now. But things that I never would've
recorded as events before. So for example, if you have
different dropdowns on your site so people can access different sections like did they get to this page, the menu, did they get to it
through an in body link? I never would've bothered
capturing that before 'cause it would be too difficult
to decipher at Universal, you can go ahead and capture it, you can create what are
called user properties. So thinking about, for example, people who come to the conference or don't come to the conference, you can actually assign
them a user property based on what they've done and then that get one
user property per person or device particularly. So you could also use user properties. There's, yeah, it's just, it's, I've lost my train of thought. (chuckles) - No, it's totally. - I just got excited
about user properties. - No, I mean I was just asking you what are some the cool things
you couldn't have imagined, you know what I mean? And you talked about some of those things. - I mean I think the biggest thing is really looking at the site. When we approach websites, the first thing that we do
when we're doing analytics is we make what's called a Gold Charter. And we have a blog poster
site about that as well, which I can link to. And in Gold Charters we're
really thinking about what are the overall goals
for those organization and then how can analytics support us knowing whether or not we're
hitting those goals, right? And something like bounce rate page views is rarely a business
goal for an organization. Same thing with like rankings, right? Rankings are rarely a business goal. So really trying to break it down and then that helps you figure out what you need to build in the report, which also helps you look at the website and figure out what kinds of
things you need to record. Because I always tell people, "Approach analytics from a
question answer perspective," not the question is, what is the most productive
channel in terms of conversions? Not how many people came to my website from various channels, right? So I think that that's where
you're really approaching it from more of a storytelling perspective. Which again, unless you're
working for yourself and you're making reports for yourself, eventually you're gonna have
to show this to somebody else. And at that point, the
rules for reporting, according to me anyway, should be, you should not be there
to explain the report. Reports should explain itself, which means you have to make
it as clear as possible, right? And I think that that's one
of the most important skills as a marketer is being able
to create clear narratives that explain whether or not
you're doing a good job. And frankly, like even if
you're not doing a good job, even if, let's say you're
running LinkedIn ads and it's not going well and you don't know what
to say to the client, at least you can say, "Hey, we realized, we
ran these LinkedIn ads. People aren't even scrolling
25% of the way down the page," 'cause you're recording that. And so you can say,
"You know what, instead, I think we're gonna need
to move the stuff around, or this audience isn't working,
let's just shut it off." But you're coming at them with solutions instead of just throwing
a 50 page report at them and hoping that they'll figure it out. That's not a way to build
trust with a client, right? You wanna be clear and honest with them. - Well first of all, thank you so much for
answering all my questions. I know that there's people that are gonna wanna check you out and you mentioned a resource, so if you want to go ahead and we should mention that your
courses on LinkedIn Learning are very affordable and
United States dollars are talking like $40 or less. But where can people discover, where do you wanna send them if they wanna first of all
connect with you on the socials if you have a preferred social platform and then if they wanna
check out your company, where do you wanna send them? - Yeah, so I'm Dana DiTomaso on Twitter, which if you can spell it, good job. I've discovered through
Google Search Console that people spell my last
name all kinds of ways and people still find me. There's only two Dana
DiTomasos in the World. The other one and I are
actually friends on Facebook, although we are not related. And then, yeah, so that's
one place to find me. I spend a lot of time
on LinkedIn of course 'cause of the courses. So you can also search for me there and follow me on LinkedIn and then I'm gonna provide
a link with kickpoint.ca/sme and I'm gonna have a bunch
of resources there as well that you can check out. - And that the, what did you call it? The WordPress plugin is gonna
be in there as well, right? - Yep. - That we talked about
for the consumption thing and also how to set it up in Tag Manager. - Yep. - So it's kickpoint.ca/sme. Dana, thank you so much
for coming on the show. I really appreciate you
spending your time with us and answering all of our questions. And folks like you gotta get on this like do not wait because
the clock is ticking. Thanks again Dana. - Thank you so much.
