I'll be doing a walk through of a live client ad account on Facebook, compare that with what we see through Google Analytics with their UTMs and also what we see through Shopify. So comparing all three platforms, see where it makes sense in the data and see where it doesn't.
Facebook Purchases
We're gonna get started through Facebook ads. So right now I'm gonna take a look at many different time periods and compare these different time periods. So obviously it is very different because the thing is the late attribution with Facebook may be coming in seven days later than it actually comes through. So that's why with UTMs, we can't really look at a specific date. So I'm gonna look for the month of January so far. So the month of January for that client, as you can see through Facebook ads, right now there's $1,980 spent on ads for a little over $20,000 back. So a 10.4 ROAS overall for these guys. So this account was doing pretty great on Facebook at the moment. And right now that would be equal to about 375 purchases attributed to Facebook once again.
Google Analytics Purchases
If we head over to Google Analytics, and from there, let's make sure that we select the same period of time. What we use in terms of UTMs are FBMKT, FB ads, as well as Facebook. So right now it's mostly FBMKT and FB ads that we use in our ads, depending on when we actually launch them. Looking at that right here, if we just take a look at Google Analytics, we would have about 147 purchases for these two campaign source. So looking at this right here, that would mean that it's about half of what Facebook reports, right? 375, one, let's round it up to 150. It's less than half. It's about 40% of what Facebook reports in terms of purchases, just by looking through UTMs.
Shopify Purchases
And now, if we bring it over to the Shopify dashboard, so from there, once you're on Shopify, you go to Analytics, Dashboard, Online Store Sessions by Traffic Source, and then let's view the reports. Actually, what we wanna do right there is take a look at Sales by Social Source. We add a filter, and then we just add a source for the UTM, and we put FB Ads, Facebook, and FBMKT, just like we did with Google Ads, or Google Analytics. We apply the filters, and that outputs 173. So again, if we look back, same time period right here, looking back at Google Analytics, 147 Google Analytics, 173. So a little more than 20 extra sales showing up on Facebook right there, or actually showing that right here. There's the NA source right here, so we could take eight, but let's just round it up. There's a discrepancy of about 15 to 20 sales between UTMs on Shopify and UTMs on Google Analytics.
Why Google Analytics is under-attributing?
There is definitely a clear separation between what's attributed on all three of the platforms. None of those are outputting the same numbers, and this is where it gets really tricky. Sometimes you would see for the same scenario, about 375 purchases on Facebook, but then you would take a look at Google Analytics and see like 170, and then it would be all over the place throughout all platforms. There's about a 40% of the purchases attributed back on Google Analytics, and closer to 50% attributed back on Shopify. And the question that a lot of people are asking is where is it actually accurate? Where do I look? What is the real stat? Is it Google Analytics? Is it Shopify? Is it Facebook? And the answer to that is that none of those are actually 100% accurate. The fact is that Google Analytics and Shopify UTMs are definitely not 100% accurate, and that always has been the case even prior to iOS 14. Because UTMs, basically people have to click through an exact link, and it's the URL that tracks the sale, would have to be pretty much a direct conversion for that to happen. For that saying, there is definitely a clear drop-off all the time with UTMs, so it is not the best way to track your ads results. So that leaves us with Facebook. Is Facebook the best way to attribute back? I would say still, Facebook probably isn't the most accurate form of reporting as we're speaking. There are other alternative ways to report. We've been using Omnitrack for some of our clients.
Why Facebook is over-attributing?
Facebook is still not 100% accurate. And there is no set thing. What's set with a UTM is that they always under-attribute. I've never seen UTMs over-attributing because it's pretty much impossible. People would have to click on the very specific URL containing that very specific UTM for that to be marked as a sale. It very rarely over-attributes. Actually, I've never really seen it, but it is clearly under-attributing, always has been, because not everyone is on that same exact URL that is containing, again, this UTM code. There is a drop-off. That's clear with UTMs. Those are basically under-attributing. But with Facebook, it'll depend. Some of our client accounts, we can see that they're clearly over-attributing some sales, not with that one specifically, but another one I have in mind is they do wholesale through their websites, and in no way can people actually go from ads and actually do a wholesale order. But sometimes ads will get attributed wholesale orders. Because the purchase conversion value for that day, it's just gonna be through the roof. ROAS is gonna be 50, and that's just plainly impossible for that client. So that said, we know for a fact that for some client-add accounts, it does over-attribute sales, but it's the complete opposite with some client-add accounts. As an example, mainly within the US, it clearly under-attributes. Some of the main problems we've seen within the US market specifically are that the purchase conversion value is just off. For some reason, sometimes, it doesn't report back the exact purchase conversion value. So we get for US clients lots of ghost purchases, which appears to purchase on Facebook, but there's just no conversion value associated with them. So that happens a lot. Or we'll get purchases come through, and it looks like purchases are 10, 12, 15, you know, $20, whereas the AOV that client is closer to 47 to $60, depending on the days. So that again, just looking at the Ads, it makes it very hard for us to optimize when attribution is all over the place.
What should we do?
From a business owner's standpoint, it doesn't really affect you in any way when it comes to revenue. When looking at revenue, if the ads, whether the ads work or not, you should see that through Shopify. You should just see, do you get an increase in revenue? You know, how does that compare to your regular rev when you turn on the ads versus when they're off? You should just see a trend that shows you, you know, at the end of the day, you get the same results. But from a marketer point of view, or just again, if you're a business owner that really wants to understand your stats and wants to be clear as to what channels brings that, how many sales, then attribution is very important. From a marketer standpoint, you wanna know what, conversions and what is the value depending on what Ad sets, Ads and campaigns, because you wanna know what to optimize. You wanna know what to turn on, to turn off, to increase the budget, to scale or to downscale. You wanna know all of these things. At the moment, if you do not want to invest in any of these, high-end tools, it's still the best ways to compare all three platforms. Look at Google Analytics, look at Shopify, look at Facebook and, you know, report back through all of the three platforms. It makes sense out of the data. Sometimes let's say the cost per acquisition may be pretty accurate on Facebook, but then the amount of actual purchases that you get is much more accurate on Google Analytics. And then the AOV is much better on Shopify. Again, you have to look up all three platforms. You cannot look at simply one platform anymore. It really has to be across three different platforms for you to make sense out of that data. But if you do have the resources to invest in platforms, I highly recommend you check out some of these. I've mentioned them on the channel before, but Omnitrack never gave it a try personally, so I can't really comment on that platform. Omnitrack, we've been starting to use that for the past two weeks with one specific client ad account. And from what we've been told by the platform owners, it takes about a month for them to start becoming accurate with the data. And we can clearly see right now that the data is not 100% accurate from what they come out with their platform.
In summary, your UTMs, so looking at Google Analytics and Shopify UTMs are always going to under attribute. That's just a simple fact. You're going to have less purchases associated with that, but the overall AOV is going to be much more accurate with these metrics. And then looking at Facebook, overall, it's usually kind of good for attributing the purchases amount. Sometimes purchase conversion value is going to be off. Definitely it's hard also to attribute these back to the ads. And that's why looking at the UTMs to see what campaigns, you know, are performing, following, again, the UTMs, will help you make some informed decisions about what to do with your Facebook ads campaign.