Affiliate marketing may not be a massive channel, but it is an important one for any online sales project.
Whether you run an ecommerce or sell training or services , you should not dismiss it.
If you already use it, you know this. If not, in my course on setting up your ecommerce business I explain the reasons in depth.
But of course, one of the important requirements is validating conversions. Because when you work on CPA, it is (very) important to understand whether a conversion meets the agreed criteria or not.
And this, when you use the attribution model last-click , is—or rather was—fairly easy to check with Google Analytics.
Specifically, with its Universal version.
If you had the tool and the affiliate programme pixels reasonably well implemented, Analytics gave you results similar to those from the affiliate platform.
And that was useful, because it made the validation process much easier.
Even when it was not last-click but any other model, using segments, it was more or less easy to obtain consistent results.
But we can no longer use Universal Analytics.
Now we have to measure with GA4.
And things change.
Because, even if you keep both the same session duration and the same attribution window configured in the tool, when it comes to measuring traffic, Universal and GA4 do it differently.
But before we start, take a look at this because it might interest you:
Allocate the budget and create different scenarios very quickly
Putting together the annual Marketing plan for an online store is a pain in the ass. Whether it is a new store or one with history.
And if you also have to make the budget fit the required investment, even more so.
That is why I use this template, where I enter a project's data and it then returns the required investment with the figures updated. Without having to change formulas and calculations.
It is not a bad way to start with your ecommerce analytics.
Besides the calculator, you will also receive a good tip or trick in your inbox every day to improve your digital business or project.
Now, let's get to the differences…
Traffic differences between Google Analytics versions
In fact, if we have the same user session duration (30 minutes by default), the same conversion window (30 days) and a similar implementation (with Universal and GA4 tags firing at the same time), the traffic differences will be limited to a single case:
That the user has arrived through different sources in the same session.
And although it is only one case, it is not negligible.
Imagine, for example, that you want to switch internet providers and search Google for one of their web addresses.
You know, the one the tech expert in your WhatsApp group has spoken highly of.
You browse around and see their rates. You think they are better than your current provider's, but you leave signing up for another time.
And you go off to read Marca.
Because that behaviour is very normal, the marketing people at the internet provider you visited know it. So they launch an aggressive retargeting campaign, with banners chasing you while you read about the latest galáctico signing in the newspaper.
The offer they show you is good.
Very good.
So you click the banner and you are back on the company's website.
And since the landing page is even better than the offer, what the hell, you sign up.
Everything OK, right?
Let's see:
- You, happy because you will have saved some money on your monthly bill.
- The new company's marketing team, happy because its direct-response strategy works.
- The company providing the retargeting (for example, Criteo) is happy too because, as it works on CPA, it will collect the commission.
- The only party that loses is your old mobile company, which was clearly ripping you off.
Sure?
Well, the answer is that it depends.
Because when you see that Criteo wants to charge you for that transaction and your GA4 says it was organic, you will have to decide what to do. Who you can trust.
You would not have this dilemma with Universal, because there you would indeed see that the transaction belongs to Criteo.
You may think these are rare cases, but I assure you they happen:
These are two conversions the affiliate (Blue) might not get paid for if we validate the transactions in GA4.
You may now need to do a deeper analysis than before…
Reasons for the attribution difference
The explanation is simple:
- In Universal Analytics, if a user entered from a different source to the site within the same session (as in the example), the tool assigned a completely new session. In our case, we would therefore have two sessions: the initial organic one and the second display-retargeting one, which is where the user converted.
- In GA4, because it is within the same session, the traffic reports will only show us the first source of the session, that is, organic. Therefore, the conversion will be assigned to this channel.
The difference makes sense, right?
Now, if we take it one step further and have the user in our example not convert in this second session, but instead return directly to the website a third time the next day and sign up then, both versions of Analytics will show a new display-retargeting session and assign the conversion to that channel.
Interesting.
Final conclusions
As you can see, little by little we are getting into the details of GA4 and looking at specific scenarios so we can try to provide solutions.
Maybe one day we will manage to have a tool as robust as the previous one.
For now, we will have to settle for understanding these quirks and knowing how to apply them correctly.
In any case, if you have questions about your GA4 implementation or want to review your measurement systems, let's talk.
Leave a Reply