Episode 96: Going from WTF to YAY with GA4
Announcer:
You are listening to Drive and Convert a podcast about helping online brands to build a better e-commerce growth engine with Jon McDonald and Ryan Garrow.
Jon:
Hey Ryan. So one of the biggest events from 2023 into 24 has been GA4 being forced upon us. So thank you Google for forcing Google Analytics four on everybody. But I have a hard time blaming them overall for this because Apple have a whole bunch of privacy moves they made a couple years ago and tracking has just had to change and really change quickly in a lot of cases, especially for e-commerce businesses.
So it turns out that to put it nicely, very few people are happy about it. Some are even angry. I am interested to see where you fall in that spectrum because both of our teams are in GA4 every day. But I have a feeling that across your thousands of clients, you guys probably see a lot more angry notes about this than I do, but I'm interested to hear your thoughts.
Ryan:
I'm old enough that I don't necessarily change. So regardless of what it is, I'm probably going to push back a little. But especially when it comes down to something as important as to tracking your traffic sources and what is or is not working, that's really when I bristle, when something changes so dramatically that impacts businesses of all sizes. But likely impacts smaller businesses to a higher degree because they don't have as many options or.
Even as many resources internally to dive into it and learn it as quickly or pull different reports as well. So I would put myself on the frustrated spectrum. Would be the best, nicest, most political way to put that. It's just that GA4 was pushed out and it's so different and was no, it feels like there was no thought to the transition whatsoever within the Google walls. We had some system that we had all used for, feels like close to a decade. There was the move to Universal Analytics but that was a phenomenal transition compared to what we're experiencing now and that was more just a Pixel change but a lot of the ways you see the data in the platform was the same. We got some cool reports and some hidden areas that were really cool.
GA4's like, "Hey, here's a completely new system that we're not going to be able to look year over year at all. So you have to have a lot of really weird ways you're going to try to compare data", which we've been comparing period over period in digital marketing since you started in it really. "How do we do last Black Friday, Cyber Monday?" Well, let's just look at the change the dates and do the compare. You can't do that anymore, at least for a year. And then you can start doing some stuff. But if I was in charge everything, it would be better of course.
Jon:
All right.
Ryan:
Just kidding.
Jon:
So you're suggesting changes to how businesses track online traffic, is that right? And not just traffic, but all of their digital marketing metrics I assume.
Ryan:
Yeah. Things are changing. We have to begin to live in this new world where you can't track the way you used to. And I might even place more of the blame on this on Facebook. They're the ones that Apple really pushed back against and forced this and it probably, as with most things, comes down to money. Like Apple needed money to look the other way and Facebook didn't want to pay whatever. But Facebook was probably overreaching in some of their privacy areas and Apple pushed back on that, which okay, privacy is a good thing.
Jon:
That is one of Apple's core tenants, right? They market on security and privacy, so I get that too.
Ryan:
Yep. So good on them for standing on it bad on the analytics world for not being prepared for Apple to actually do what they said they were going to do. And so Google, it feels like, had to pivot really quickly in what used to be called Firebase and was really designed to track apps and downloads and some cross-platform things, which is good. It was just nobody liked it before and then nobody likes it when it now is called GA4. So it is where we're at and so we have to work within the confines we've been given.
Jon:
Well, I have another option. Why don't we just throw GA4 out and not even put it on our site?
Ryan:
Well, in reality I'd like nothing better. But as of this moment, we're recording this right in the middle of holiday season 2023, there's not a better free analytics tool out there that comes close.
Jon:
Oh, okay.
Ryan:
Google Analytics, Universal Analytics, free. That tool might be the best free tool that's ever been given out. There was no charge at all and you could just set it up and use it and it would track wonderful things, create great reports. I can't think of [inaudible 00:05:19]
Jon:
Let's be honest. The end result for marketers was great, but what's the old adage? If you aren't paying for it, you are the product.
Ryan:
Of course. But as a marketer or a business owner, it's like use the data, okay, but at least I can use it to drive my business or inference what is working or is not working. And what works really well for a decade now is very different. Not that it can't work well, but if you're not willing to start seeing... And I will say a big asterisk on this is it coincides with really the pinnacle of cross everything attribution. We already knew your Google Ads didn't operate in a silo and say everything that Google Ads says it drove, did it all by itself and nothing else helped. New attribution tools such as GA4 is really trying to show that mess of spaghetti saying, "There's lots of stuff going on and here's what we say is responsible for this traffic." And so there's going to be frustration even going to look different in GA4, pulling the same report that you would've seen in Universal Analytics.
Jon:
Okay. So I'm hearing that you can't get the same kind of data out of GA4 that you did with Universal. Is that true or partially true or?
Ryan:
It depends on how much time you want to invest in learning the platform.
Jon:
Let's say zero.
Ryan:
It's rough. It's rough to get the same reports. You can get some of the same data but it's going to look different. It's going to need to be read a little different and I can probably make an argument that I kept Universal Analytics analysis on my end fairly simple. And so you would think the transition should be easier for somebody that didn't get into the minutia of all these nuanced reports within Universal Analytics. But it's still fairly challenging. I spent most of my time jumping into analytics, going to my favorite report, which was acquisition, all traffic, channels, and then I would view my channels between timeframes. Because I was usually trying to find out where something broke or something happened, either good or bad. And then try to diagnose what worked really well? That caused this. So that was where I would start as a high level. And then I would very easily be able to click on a blue link, dive into that channel grouping and add a secondary dimension like a landing page. It's just super easy and it kept all my timeframes in place. GA4 does not do that easily now. And trying to make GA4 work that way will frustrate you to the nth degree. If you've already tried to do it, you know what I'm talking about. But just don't, it just doesn't work there.
Jon:
Okay, so there's a lot of these reports, they're not the same as what they were. I'm hearing a lot of words likely, some. I'm not hearing a lot of, well, there's a one-to-one between what you had and what you get now at all. I'm hearing a lot of, some of it's there. Likely you'll be able to do as you want. It's very different. Ease my mind here. Please.
Ryan:
There's probably no easing of it. There's a lot of power in GA4 and a lot of things you can do with it but the vast majority of business owners don't even need to use a lot of those tools that are in there probably. And the massive change that's impacting this really comes down to attribution. So Universal Analytics last non-direct was the attribution model that it defaulted to. Which meant the last channel that was there when the sale happened that wasn't direct, so if it went Google Ads direct, it went to Google Ads. And that was easy to understand after you've spent enough time in it, you're like, okay, I know what I'm looking at.
GA4 defaults to data driven, which is Google's answer to, "It's a black box, but here's what we're telling you." And like, well, I don't want to understand that. You can't understand that. Compounding the complication in that is it's always going to change. So it's going to use data to get an AI to get "better". My air quotes here saying, yeah, it's getting better. Well, how do you know? You don't know it's better. It's just different and it's like Google telling you it's better.
Jon:
Yeah, that's fair.
Ryan:
That aside-
Jon:
We were at a Google event a few months back where they very clearly said, "Hey, prepare for this because we're going to have a black box of an algorithm and we're not going to be able to tell you one for one on your data like we used to." And I thought it was interesting because a lot of people in that room, some large e-com brands were there too, but they didn't even understand that unless you were paying for GA360 previously, you weren't getting a one-to-one of your visitor traffic analyzed anyways, you were getting a sampling of that and then they would extrapolate what should be happening to get numbers. So every analytics platform is a little different. So you always want have a source of truth, go with that as a source of truth, but know everything else needs to be trend lines. And you're really looking at the trends. And that's what bothers me about GA4 is that I can't look at the trend anymore.
Ryan:
Yes, and it's for some brands like you were really getting throttled. For 90% plus you were getting a large portion of your data just because you were small enough that Google's like, "Whatever, it's such a small data set, just have it." But Google's goal is not even [inaudible 00:11:00]. It's just their goal is stockholder value and shareholder price and make it as easy as possible for brands just to throw something out there and have Google tell them it's working. Performance Max campaigns, there's an algorithm behind that that just does the work and it lets you get shopping campaigns up even if you've never done them before and spend money. I'm not necessarily saying you're going to get sales but you're going to spend money, which is what Google wanted.
And so GA4, again, a lot of power, a lot of stuff there, but you have to understand some tweaks to be able to get it usable if you're stuck in Universal Analytics like I am. And so the easiest I've found to get to the channel report, I'm going to give you some, we're going to talk through it and you can rewind and try to go back through it. But step one, understand that the search bar in GA4 is probably going to be your best friend. We love the search bar at Logical Position. I may not know how to navigate somewhere, but I can be like, okay, I know I'm trying to get to this, so I go to the search bar and it's actually really good. That's probably the best part of all GA4.
Jon:
Well, everything moved too, right? Yeah, everything moves.
Ryan:
Yeah.
Jon:
Well that's sad but true. Yeah. Okay.
Ryan:
Search bar, just start typing model comparison and it's going to come up with advertising attribution model comparison. And then you have to make sure when it picks that report, that you set the report to conversion time, not interaction time because that'll impact how the data and when it tells you the conversion happened and the numbers you're seeing in your report. This is your channel report that I talked about before, that was acquisition, all traffic, channels. And you can see last click interaction and then you can see data-driven numbers. So I ignore data-driven right now because I'm trying to compare generally to... I did this for example on my Black Friday Cyber Monday analysis for a lot of brands. Let's pull that report here so I can then have on another screen my Universal Analytics from last year side-by-side and see, okay, what changed? How are the reach reports today? And so this is where wide screens are pretty helpful. My cousin has one of those, I don't know, 40 inch curved screens that operates like six screens for him. Those are really helpful now in analytics attribution.
Jon:
So what you're saying is Google has become part of the monitor lobby.
Ryan:
That's true. They probably have some shares of Samsung and they're like, "Hey, we need these big curved monitors." And yeah, Samsung and LG are doing really well in these new monitor spaces. There are also some good deals during Black Friday Cyber Monday. If there's still some deals, you might want to give it into those. I may have done some of that.
Announcer:
You're listening to Drive and Convert, a podcast focused on e-commerce growth. Your hosts are Jon McDonald, founder of the Good, a conversion rate optimization agency that works with e-commerce brands to help convert more of their visitors into buyers, and Ryan Garrow of Logical Position, the digital marketing agency offering, pay-per-click management, search engine optimization, and website design services to brands of all sizes.
If you find this podcast helpful, please help us out by leaving a review on Apple Podcasts and sharing it with a friend or colleague. Thank you.
Ryan:
Then there's one other report that I found pretty helpful to help some of us, what I would call old-school marketers, get into some of the comparison to find issues. I like reporting engagement landing page. And in that you click up at the top add comparison, and it pulls up this little widget on the right side. And it's session default channel group, exact match to pick, and that gives you your options. Otherwise you have to type something and it picks. But if you do exact match organic search, this is where I start finding problems or finding opportunities on SEO. So if something broke, I can be like, okay, which landing page struggled? Okay, this one did. I can go to way back machine for somebody I'm not working with and figure out, okay, you changed content here on this landing page, traffic went down or went up. Either fix it or expand that onto other landing pages. But that's a really powerful one, especially if you want to look at some of your product page opportunities within shopping. Like, hey, let's look at some product stuff, see which one's really driving a lot of sales and analytics. So two main reports I keep myself on if I'm forced to use Analytics 4.
Jon:
Okay.
Ryan:
It's not perfect, but it's getting closer.
Jon:
So okay, I'm still hearing words closer, still not there. Are you supplementing GA4 then? Are you using other tools? What's the game plan here to make up for these wishy-washy, it's not there yet, but maybe hopefully one day it might get there.
Ryan:
I would advocate for most companies to be paying for an attribution and analytics tool, most companies. Some of you listening are of a size or you can't or just won't pay for analytics attribution and you're likely going to be supplementing GA4 with a platform analytics. So you still have to have it, it's free.
Jon:
Like Shopify or something, right?
Ryan:
Yeah, whatever platform you're on, BigCommerce, Shopify, Meva, Magento. There's some backend system that says, "Here's where your revenue came from." And I guess a couple of important points if you're going to use that as your attribution stack, GA4 plus platform. Platform and GA4 will almost always be different just like they were before. Like UA was never going to match Shopify. Taxes and shipping are going to be handled differently. And then if you thought GA4 was struggling with attribution, the platforms in a privacy world and UTM issues, they're struggling even more.
And so you're largely going to be looking at totals on those. Okay, last month we did 100 grand in GA and we did 120 in Shopify. Directionally close enough to compare period over period, things like that. And we look at transaction ID comparisons. So if we've got a client on Shopify that only has GA4, we're going to say, "All right, let's look at the directional data between transactions to see what's not being tracked in GA4 or Shopify. Let's see what the issue is. Shopify is reporting on the taxes and shipping as a transaction and GA4's not." And then we have data driven, which is going to make sure that some of that... Just gets complicated.
So it's going to be different and just understand that. And if you're using just GA4 and platform, I will tell you you need to have the Google Ads Pixel on your ads channel. Make sure that that is how you're using your tracking for conversions and Performance Max and everything else, because there's a lot of things that go wrong when smaller brands in particular have implemented GA4. Some of them have revenue in the system and maybe not in ads. There's really dumb problems I see. Or vice versa. You have it there and you don't have it there. And it's using a data-driven model which may give less credit to ads which may not populate as much revenue in the ads platform to get it as aggressive as you may want to.
Quick story on that, we have a larger brand that migrated from one URL to another and it caused a whole host of issues. And their previous agency had said, "We're just going to go with the GA4 Pixel and let that be our attribution." Well, because of how people were interacting with their site, there was so many touch points. It was a B2B primarily with higher price point items so more research was done. And it was causing the ad system to think it was getting a much lower return because it wasn't getting credit within the GA4 system that maybe it would've gotten in the Ads Pixel. So we made the switch to the Pixel. And we have seen since that, and since the system has finally caught on to that Ads Pixel, it's showing more revenue, we've actually been able to increase site traffic and revenue almost 2x on a fairly large client just because there's the Performance Max campaign believes it's doing better. And it actually is, but it's seeing more conversion data with the Ads Pixel than it did with GA4.
Jon:
Okay.
Ryan:
And I like redundancy. So you can see both. But again, I like to have a Pixel that's feeding the Performance Max campaign be a little more aggressive so it can see more conversion data. And so use that.
Jon:
Okay, so if you don't want to use GA4, what should we be looking at? I know there's several options. We've talked about a handful of them. But I think it'd be great to recap those here and get your thoughts. Because there's probably some I haven't heard of as well.
Ryan:
Yeah, a lot of it's going to come down to your platform. So some of the biggest players out there in this brave new world of attribution modeling and analytics tracking stuff started on Shopify because that's where the vast majority of e-com brands are. And so Triple Whale arguably has the best user interface reporting system for e-com brands. There can be a lot of debate on that, but we found it very intuitive and what they've done with some of the comparison areas, just it's very simple for somebody that hasn't had a lot of experience.
Jon:
It was founded for the owner operator, right?
Ryan:
It was.
Jon:
Right.
Ryan:
And they've gotten a lot more complex too. So when you start with Shopify, you're starting 2 million or so brands using Shopify I think. The vast majority of them are under a million dollars in revenue. So they're small businesses and they've-
Jon:
I'd argue under $100,000.
Ryan:
Yeah, the Shopify+ number is not a large percentage of their overall number of merchants. So great. It's a very simple system to use. And Triple Whale was invested in by Shopify so they created just a very simple interface that has been getting fairly complex and they're building some really cool integrations coming forward. So if you're on Shopify and you're going to pay for something and you're a smaller brand, I'm probably going to tell you Triple Whale is it. That's what you need to start with.
A close, not a horribly expensive to that is going to be KnoCommerce, K-N-O Commerce.
Jon:
Okay.
Ryan:
They're platform-agnostic but they have a really simple integration to Shopify. The other ones have an extra step or two. Still not complicated, but they are a post-purchase survey, which has been around for a long time but we really ignored it because Universal Analytics was so good at telling us where traffic came from, we didn't need to go ask a converter, "Where'd you come from?" Now we're blind in certain areas. We have to ask them, "Where'd you hear about us?" And KnoCommerce has a really good system for that. It integrates in with Triple Whale. Triple Whale does have their own survey system built in but even they'll tell you it's pretty basic. KnoCommerce will integrate in there and actually give you quite robust series of questions and reporting on this.
And we actually used it on TikTok, well, not on TikTok, but because of TikTok attribution issues. We had a beauty brand that they were influencer driven and it seems like how you have to start a beauty brand now. Really good influencer, she starts a beauty brand, it starts growing. We're like, "Hey, you're really big on TikTok. You should spend money on TikTok doing some boosted posts, spark ads. Let's do some audience stuff." Okay, great. We moved, I think $5,000 of her budget a month over to TikTok. GA4 showed all of $100 in revenue. We're like, "Oh my gosh, sorry, we suggested that. Not good. Let's take that back, put it back on Meta and Google. Did that. Saw a slight decrease in revenue." We're like, "That's odd." Knocommerce was in our partner's stack and we're like, "Hey, we should talk to KnoCommerce and see what they have options." They're like, "Yeah, TikTok loves us because we can actually open up the eyeballs of what's there." I was skeptical, to be honest. I'm sorry. Barbara, she's the partner manager over there, great guy. I was skeptical at first. Like, "Okay, can you give it to this client for free for a couple of weeks so we can see it work?" Of course, they're very generous with that.
Put it on there. We started spending again on TikTok, found out zero people leave TikTok to go buy something, zero, not even... And this was an impulse buy, like less than $50 beauty brand. And almost four weeks later, people [inaudible 00:23:12] back on average to buy. From the first time they saw the ad on TikTok to when they purchased was between three and four weeks.
Jon:
That's a delayed attribution.
Ryan:
Oh, man. I never thought it would take people that long to take action to buy a $40, $50 product. Never. Turns out you got to hit them a few times and it takes time. But that was our first eye opener, like, holy smokes, TikTok doesn't suck for ads. Despite all the data we saw in GA4, it's actually really good at that upper mid-funnel. And so KnoCommerce is really good. And TikTok believes in them so much they actually give you a pretty hefty ad credit just by plugging them in. And even if you're already spending, they're like, "You already like us and you're spending on TikTok. Great. Use KnoCommerce, we're going to give you more money to spend because you need to have this tool to show how good we're doing for your brand."
So those two kind of go hand in hand. I would say Triple Whale first, then add on KnoCommerce or some could even argue vice versa depending on where you're spending your money first.
Larger brands on Shopify, the Shopify+, if you're over three, 4 million in revenue, you need to start looking at maybe some more complex things. Prescient will actually feed Pixels more data. There's actually a couple other companies that do that as well that help those Pixels work better and get more data to optimize. Since there's so much AI going on specifically on the social channels, Google as well, feeding them the data they need can help. And then SourceMedium is a great tool for larger companies that are trying to tell stories from the marketing team to the executive team. It operates phenomenally as a source of truth that everybody can look at and operate in very easily, arguably. You can't get as much analysis done into the minutia of an ad platform or an ad specifically like you would be able to do with Triple Whale, but it tells a very good story to everybody that they can all understand.
Jon:
That's a new one for me. I have to remember to look this one up because I've never actually used SourceMedium. I've never heard of it before. Interesting.
Ryan:
Yeah, it's a smaller company. They're hitting larger brands, for sure. They're not going to take a small brand. You wouldn't even need them probably on most small brands. But really cool, very simple. I say simple. Every developer that develops anything doesn't like [inaudible 00:25:24] I think it's simple. But very useful tool for everybody in the organization, from finance to marketing to executive team to look at and be like, "Okay, this is what this channel did. Love it."
Jon:
That's great.
Ryan:
When you get off Shopify onto the big commerce's, the [inaudible 00:25:39] commerce's, Magentos, Triple Whale in Q1 will probably be available in most of those. But if you don't have that, you're going to be looking at probably a Northbeam, [inaudible 00:25:51] or RockerBox for, again, I'm partnered with all of these companies, so I know all the partner teams that are all very good. But I would say they're all doing something very similar. They all have little nuances between them that are different, pricing differences or the way they price things. But don't take my word for it, talk to them. Let them sell you. I'm not here to sell you on any of them.
But they are, Northbeam, for example, is $1,000 a month to get attribution. So you're not going to be likely a $10,000 a month brand and paying for that instead of marketing, for example. So you need less expensive tools to get to that size. So they're really trying to go upstream anyway. And the complexities of the Northbeam system, for example, is massive. Their interface probably isn't as intuitive as Triple Whale, but their backend analysis and data, it'll make even very smart people's brains hurt when they can see all of this stuff.
So I think if I had to sum it up, most companies need to get beyond GA4. Keep it there. Don't throw it out. The baby out with the bath water. But you need to probably find something that is more usable for more teams, rather than just having a GA4 expert that can find all this stuff. You need something that a head of marketing can get into, a specialist in Meta or Google Ads can get into and see some valuable data, and then somebody in finance can go in there and start seeing some data to help the, forecast. So it's not fun when we have to go through changes. But I do believe we're being forced as a group of marketers to really engage our brains and think through what's going on and how are we pushing brands in ads and in marketing to the direction that they want to go, rather than just defaulting and being comfortable. We're going to be uncomfortable for a while.
Jon:
Well, look, I appreciate you being a glass half full here because it could have gone the other way pretty easily, where you're just basically like, "Look, Google screwed this up really and just jump ship and use something else."
But I understand why this is happening. There are alternatives, like you've mentioned. They should be used in conjunction. I think there's no reason to take GA4 off of your site because they're only going to continue to refine it and make it better based on user complaints. Because Google wants that data, they need that data. So they're going to make this tool better. And you want that data to have been there from the beginning so that you can start to do year over year, et cetera, as you get back more into it. So that's my opinion, is honestly, it kind of sucks right now but I hope it gets better. I see a path towards it getting better. But we'll have to see. Yeah, this is really interesting and wow, I feel for you and everyone else who's in here and my team who's in it every day. Glad that I'm not in it as much as y'all.
Ryan:
Yeah. And Google's got a lot of smart people. They will get there and I'm confident of that. It's just the process was frustrating. I would be a failure if I didn't mention the company Littledata to companies that just need GA4 to start working. If you haven't even had it work so that you can do the reports that I talked about, you should probably reach out to Littledata, great company. It's going to cost you a hundred bucks a month but it's worth it if you haven't been able to get revenue it your GA4 to start using your analytics. And they have some other bells and whistles to add on. But for sure a really cool company, really good owners that have a great heart to help businesses get the data.
Jon:
Yeah, they're a nice team. They were at the last LP event that we were at and it was nice to meet them.
Ryan:
Yeah, they're very good company so check them out too.
Jon:
Awesome. Well, thank you, Ryan. I appreciate the heads-up on what to do here on GA4 and yeah. Well, we can get off your soapbox now.
Ryan:
Thanks, Jon.
Announcer:
Thanks for listening to Drive and Convert with Jon McDonald and Ryan Garrow.
To keep up to date with new episodes, you can subscribe at driveandconvert.com.