Adjust, the mobile attribution and analytics firm with expertise in ad fraud prevention tools, just announced that “click injection” fraud is likely to become a dominant form of mobile ad fraud in 2017.
Click injection is a fairly new type of mobile user acquisition fraud. Fraudster app developers would publish their fraudulent app, which fakes a response to an ad campaign and thereby generates ad revenue for the developer. In other words, developers are injecting fake user clicks. These are generated from within the app and timed so that they happen within a second of the download of an app.
As of right now, it only works on Android, which uses install broadcasts to time a click. A fraudulent click perfectly matches the mobile user’s ID. Advertisers reviewing their ad performance will then notice that more installs are coming from ads compared to organic activity.
Andreas Naumann, Fraud Specialist at Adjust, explains:
“Since the fraudulent approaches we announced last year have now been proliferated by the latest tools and solutions, fraudsters are seeing their average takings plummet, and as such are looking for new ways to game the system. This new scheme is technically similar to ‘click-spamming’ we described early last year, but evades the tools that prevent click spam. We’re expecting it to roughly supplant and equal click-spamming activities in size, which accounted for an estimated five percent of ad engagements on Android.”
Adjust is currently running tests with different algorithms that may aid in preventing fraudulent conversions as part of its Fraud Prevention Suite.
It says that injected clicks are really impossible to distinguish from real ones. However, patterns within the whole campaign do offer a clue as to what’s going on. Looking at the length of time between clicks and installs offers one such clue.
Indeed, click injects result in visible bumps.
Paul H. Müller, Co-Founder and CTO, Adjust says that fraud prevention requires fine tuning and changing mobile attribution isn’t an easy task to accomplish.
“Rushing to repair datasets is just as likely to aggravate the inaccuracies that faked conversions cause. That’s ultimately the long-term problem for data-driven mobile advertisers: skewed conversion numbers and dirty data.”