Our approach revolves around leveraging SKAdNetwork (SKAN) and device-level data side-by-side to create comprehensive reports in a highly accurate tabular format. Some other tools available on the market use an “all-in-one” or “single-source” approach, essentially trying to calculate the overlap between installs attributed via the SKAN framework and installs attributed via the App Tracking Transparency (ATT) framework. While this sounds good in theory, it has some significant shortcomings. And, more importantly, you have to sacrifice a whole bit to get this insight.
The problem australia business fax list with iOS all-in-one campaign data
In theory, by calculating the overlap between SKAN installs and ATT installs, one can see which SKAN installs have already been attributed by ATT. This would mean that marketers could confirm, at an aggregate level, that Total SKAN Installs = Total Paid MMP Installs . One way this system could be used is to confirm unexpected differences between the two data sets. However, this troubleshooting process does not include any further steps. Therefore, a very important bit is wasted, just to demonstrate numbers that do not match.
Let's look at a couple of examples to illustrate this point.
First, we have the example we call False Hope:
SKAN installations: 100,000, where using the attributed bit, 80,000 of them were returned with a value of 1, meaning they were attributed
In this scenario, a UA manager might mistakenly believe that all of their SKAN installs were attributed, that their MMP shows the same number of paid installs, and that their reporting includes the correct number. This conclusion is false. While the numbers appear to match at a general level, we need to take it a step further. Of the 80,000 installs attributed by SKAN, let’s say 40,000 were attributed to Facebook and the other 40,000 were attributed to Google with very little information about the campaign. Meanwhile, the MMP shows 20,000 installs attributed to Facebook, 20,000 attributed to Google, 20,000 attributed to a web campaign (which currently cannot be run through SKAN), and 20,000 attributed to AppLovin.
