• Homepage
  • articles
  • Why do the number of installs in Facebook and the tracker differ?

Why do the number of installs in Facebook and the tracker differ?

autor
banner

View the table of contents

Content

Who it is for

  • 1. webmasters
  • 2. affiliates

From time to time, webmasters encounter a situation where installs of webview applications are not displayed in the tracker. Let's explain why this happens, where to look for issues, and how to prevent the loss of statistics. Disclaimer: there is a lot of technical information below, so grab a coffee and focus.

 

 

Challenging path of traffic and postbacks

Let's first take a look at the scheme of traffic and postbacks passage through challenges, meaning from the advertising platform to the offer and back. If everything is set up correctly and working well, the scheme operates, the web is making a profit, and enjoying its tequila on warm islands. But then we’d be living in paradise, so it's necessary to figure out where the failures could occur.

 

When a lead installs the application, the install is only shown on Facebook. The installation is displayed in the mobile app tracker and for the buyer after the application is launched and the offer opens inside. If the application opens but the install is not displayed, there are several possibilities:

 

  • the user was cut off by the cloaca;
  • the user was not attributed by AppsFlyer/AdJust. The attribution model is a set of incoming data in a specific format that is unpacked by the attribution system – a program that unpacks the data for use in the next system – TDS, and thanks to this, it enters your flow in the tracker;
  • if the application includes AppsFlyer, during the data transfer between the service and the unpacker, there may be a difference of up to 5% between the number of installs visible on Facebook and the number of installs visible in the TDS system. If up to 5% of traffic is not attributed, everything is normal, but if the difference is greater, then perhaps you have used naming incorrectly;

 

In our practice, there was a single case where the maximum discrepancy at its peak was 15%, but this isn’t the norm. A difference of 15-20% indicates a possible shave or errors in linking the application.

 

  • malfunctions with the application itself. We need to get a grip and go talk to the support to find out the causes.

 

Let's move on. After installing the application, the web awaits “registration and deposit”. These actions are performed on the offer at the advertiser's end. Thus, if you don't see registration and deposit on your end but they exist for the advertiser’s end, there are two possibilities:

 

  • you have incorrectly configured postbacks;
  • a malfunction is on the advertiser’s side;

 

If you have registrations and deposits, but they are not reflected in Facebook, the malfunction can occur at three points. In this case, you need to contact mobile applications support to retrieve logs and determine the cause. Perhaps there's a flaw in the postback configuration from you to the TDS of applications. Typically, this is checked during integration, but it won't hurt to double-check.

 

If you notice that you have more registrations and deposits than on the advertising platform, and the figure isn't decreasing, you can pass a batch of ClickIDs and event names, and the rental service will then shoot postbacks and investigate the occurrences. An additional check to identify the issue involves comparing the numbers between TDS of rental services and TDS of arbitrageurs. In this case, a check is conducted to verify the correctness of the postback processing in Facebook based on ClickID.

 

If everything is fine with the applications, as evidenced by the logs, then the issue lies in the processing of incoming data on the Facebook side. All questions should be directed to Zuckerberg.

 

If no issues were found, it means you didn't have registrations and deposits because the advertiser doesn't have them either. There could be several reasons for this too:

 

  • Players are not opening the application after downloading. A low conversion from click to install indicates that the app may be too heavy for your GEO: leads likely minimize it to the system tray without waiting for the download and forget about it. Or they download the app in parallel with other activities. In this case, the question turns to your creatives, perhaps they didn't hook users from the start. Players might have seen a better “creative” and decided to play in a different app and on a different offer. Double-check the app's “weight” and the average connection speed in the GEO, and test a different creative.
  • The app crashes for players with outdated phone models. If a player installs the application but has an old phone that may not support certain features, the app might crash or not open at all.

 

And the most unpleasant part. Discrepancies may indicate a shave by dishonest advertisers and rental services, and here the question arises about how your relationships are built with partners, where the offer came from, and whether you are confident in the app rental service. If you see that the conversion from registration to deposit has dropped below the norm, then the question is for the advertiser regarding the correctness of the payment methods’ operation.

 

All traffic initially goes to the advertiser's black box, and the information from this box is returned to you. What it’ll be depends on the product's conversion and its settings. The industry is built on trust.

 

Stay vigilant for the consistency of displayed data and build warm relationships with all players in the market so that peace, friendship, bubble gum, bids, caps, and wild love prevailed.

Subscribe

Share

comments ....(0)

Leave a comment

You must be in to leave a comment

  • Homepage
  • articles
  • Why do the number of installs in Facebook and the tracker differ?