The hottest topic in the mobile advertising industry right now is fraud. Over the last year or so, we’ve seen a massive trend toward identifying fraudulent activity and the need to prevent further spending with channels driving this type of traffic. Paying for fraudulent clicks and installs is costing marketers a lot of money, so identifying this activity as soon as possible is critical to your overall ROI.
The TUNE Marketing Console continues to take steps to protect advertisers against fraudulent installs. Some of these steps include rejecting installs due to jailbroken devices; identifying unmatched iTunes/Play receipts and preventing attribution on clicks from known bot databases.
In building our next evolution of fraud tools for marketers, we wanted to think about the best solution — not just an easy, short-term fix. Today we are releasing two fraud reports for marketers that help identify suspicious activity so traffic sources can be held accountable for the users they are driving.
Lag Time Variation Report: Identifying and eliminating click fraud
A trend we look at is the time difference from click to install. This number can identify two broad classes of click fraud:
- Click injections: Malware that detects an install and sends a simulated click
- Click spamming: A variety of techniques including click spamming (cookie stuffing), impressions treated as clicks, click stacking, etc.
An install that happens too quickly after click could indicate a pattern of click injection across the traffic. (A click injection occurs when a new install begins on Android, and the existing apps on the device are notified of this install through broadcasting, an inherent behavior in Android apps.) When a fraudulent app on a device is notified of a new install, that app will send the device’s advertising ID to the publisher before the new app can finish downloading. The publisher will fire off a click that contains the advertising ID of the device and thus take credit for the organic install.
If the time from click to install is significant per source (and would display a linear curve if you were to model it out), it could be an indication of click spamming or cookie stuffing. In this case, there is no direct correlation between click and install, and publishers are taking credit for organic installs that were going to occur regardless.
Our new Lag Time Variation Report tackles both of these scenarios, providing the mean time to install and the percentage of installs coming in prior to a threshold, such as 10 seconds.
Using these tools to move spend from click fraud dead ends to productive traffic paths can help drive growth for your app in the market. In many cases this might be minor tweaks with the partner to discontinue 5% of the overall traffic by eliminating a specific affiliate, or even traffic in a given country. Continued refinement of spend can make great partners even better.
Install Validation Report: Understanding rejected installs
Another trend to look at is how many installs are being rejected per channel and understanding the reason for these denials. Gathering this information manually, of course, is easier said than done. That’s why we created the Install Validation Report.
Within every rejected install record, we provide a reasoning code that helps the marketer understand what specifically caused the rejection, down to the device level. The Install Rejection Report then aggregates rejected installs by source, allowing you to see at-a-glance the number of fraudulent installs received per partner down to the sub publisher level.
We’re just getting started
The new Install Validation Report and Lag Time Variation Report are just the beginning of a suite of fraud tools soon to be released by TUNE. Please reach out to your Customer Success Manager or TUNE Support if you have any questions regarding these newly released reports.
Never miss a thing!
Want the goods delivered straight to your inbox?
Sign up for our blog recap emails to stay in-the-know about digital marketing, analytics, and optimization.
Author
Sandor is a Product Marketer at TUNE. Prior to this role, he was an account manager with the Tune Marketing Console product for 3.5 years. He received his bachelor’s degree in biology and worked as a R&D scientist for 4 years at a biotech company. After receiving his MBA, he transitioned to tech and has loved it ever since. Outside of work he is learning to golf and loves laughing at memes.
