Singular Fraud Index highlights most secure mobile ad networks & challenges for marketers

A new report from unified marketing analytics system Singular provides a look at the use of mobile ad fraud prevention by marketers and ad networks.

In its analysis, the company found that 63 percent of marketers don’t have fraud prevention tools enabled in their mobile marketing systems (often fraud prevention is a premium service that costs more). Nearly 70 percent of the mobile ad fraud attempts stopped monthly were attribution manipulation, in which the perpetrators try to take credit for app installs from organic and paid traffic sources. When fraudulent click attribution is successful, marketers can end up allocating budget to the scammers’ sources and away from the sources that are actually driving real installs.

“Fraud is still a major problem for mobile marketers — and part of the blame lies with the analytics industry,” said Singular CEO and founder Gadi Eliashiv in a statement. “Third-party analytics providers treat fraud prevention as a luxury, offering it to marketers as a ‘premium’ add-on rather than a free feature embedded in the platform. Equally problematic is the complexity of existing fraud offerings — for instance, attribution platforms that place the burden on marketers to run their own statistical analysis to detect fraud.”

One method of attribution fraud is via click injection, Eliashiv explained in an interview last week. When a user downloads a new app, a malicious app in the background on the user’s phone sends info to the newly downloaded app’s attribution system to make it appear that the install came from that app source by registering a fake click when the user opens the new app for the first time. “Some of these [malicious] apps have 100 millions of users,” said Eliashiv. “Once they’re downloaded, they listen and monitor the device for new installs. They know which app to send the click info to in order to steal credit between download and opening for first time. Attribution systems can only match up once the app is opened for the first time — that’s the window for click injection.”

Illustration of click injection attribution manipulation. Source: Singular.

Eliashiv said Singular’s open analytics platform offers a unique view into fraud prevention techniques across a wide array of mobile ad networks and systems, including its own. Marketers use Singular to unite their various tools and systems — ad networks, fraud prevention systems, analytics, email solutions, attribution systems and so on — in one platform to map and normalize the data in a common schema for attribution and analysis.

“We looked at all the mobile marketing channels and how many installs were attributed to the sources across various attribution tools to get a sense of volume across networks,” Eliashiv explained by phone. “Then we looked at what was being flagged as fraudulent on a per-network basis. It’s not fair to compare one ad network to Facebook, for example, but looking at it based on ad network’s own volume and size, we were essentially able to benchmark each network against their own volume.”

The Singular Fraud Index (registration required to download) highlights the 20 most secure mobile ad networks. Eliashiv said there weren’t necessarily any surprises on that list, but what was surprising is how poor the quality of some networks that didn’t make the list actually is. In no particular order, here are the top 20 ad networks in the 2017 Singular Fraud Index:

Source: Singular