Presented by AppsFlyer.

Fraud is an ongoing issue in the mobile app marketing space, much to the industry’s chagrin. The Interactive Advertising Bureau (IAB) estimates that the cost of mobile ad fraud to advertisers is $1.3 billion worldwide – and a recent study from AppsFlyer shows that up to a quarter of this fraud is specifically related to app install and engagement.

As much as marketers try to keep an eye on mobile app fraud, there always seems to be a blind spot where fraudulent activity slips through. And as is the case with all blind spots, most of the time you are not aware you have one.

As marketing budgets increasingly shift to mobile performance-based campaigns, fraudsters are following the money and turning their attention to that channel. As a result, the entire ad industry is forced to quickly adapt, update, and implement anti-fraud systems to tackle the threat of mobile ad fraud on an ongoing basis. It’s a game of cat-and-mouse, and the fraudsters quickly adjust and change their approach. They’re clever — they know how this game works and seem to have a trap set at almost every touch point of the mobile customer journey.

The key to staying ahead of fraudsters with such diverse tactics and deep reach is simple: stop relying on only one or two sources of fraud protection.

That would be akin to trying to prevent a bank robbery by keeping all the gold in a safe in the lobby of the building. Instead, you want to employ a number of safety measures, which complement each other to offer optimal protection. Think of the same safe, but adding a vault with a high-tech alarm system, as well as a well-trained staff of security guards.

Moreover, it’s important to recognize you’re hardly alone. The fraud your campaigns may have fallen prey to is happening at a massive scale. Bad actors are attacking thousands of campaigns simultaneously, exploiting apps on a global level. Mobile advertisers cannot answer the fraud challenge on their own. Sophisticated mechanisms and fresh approaches are necessary to help marketers fortify their defenses – and we need very big data to cut these criminals off at the source. Specifically, we need to be able to identify fraud at the device level, and the only way to do this is with a massive, cross-publisher database.

So what must we, both as individual marketers and as an industry as a whole, do to keep fraud at bay? Here are three key points to consider:

1. Think layers

The most effective fraud protection systems will use a variety of signals to create big data and machine learning-powered insights across publishers. So, as in the bank metaphor above, additional layers of protection on top of rule-based and modeling components are necessary to enhance effectiveness. Consider device-level protection as one of the most advanced technologies in defending against fraud. Combined with massive scale and layered in with other anti-fraud features, this type of protection gets smarter as fraudsters adapt their tactics.

2. Own and apply your data

For advertisers, access to raw data insights is crucial to validate what exactly they are paying for. Raw data gives advertisers the power to detect the most sophisticated patterns of fraudulent behavior, especially if they can access granular insights on a device level. If a high percentage of a network’s installs are generated from new devices, this is an indication of highly suspicious activity and should be addressed swiftly. Without looking at the device-level, it is impossible to know which networks deliver such traffic. In the offline world, this would be comparable to thousands of $1 and $2 purchases occurring in a single store on the same day. Typically, your bank would notify you about such suspicious activity, right? As a marketer spending thousands to millions on mobile advertising, you should expect the same protection.

3. Go beyond rules-based protection

Prevention and detection utilities based on suspicious patterns that can be codified into a set of simple rules rely on lazy fraudsters. Since many evil-doers are fairly ambitious, they have little trouble getting around IP filtering or faking mean time to install. While these remain important components of fraud protection, they are not as effective as they were just a year or two ago. As an industry, we need to leverage big data in order raise our game if we’re going to stay ahead of fraud — and invest in mechanisms that are constantly learning and adapting to new forms of ad fraud.

Those three suggestions could save marketers a lot of heartache — and a lot of money — in a remarkably short amount of time.

Marketers need to protect themselves against fraud across the entire customer journey. While click-fraud is like death by a thousand cuts, in-app engagement fraud can be like severe blunt-force trauma. Fraudulent engagement grows more costly as it moves down the user funnel.

At the same time, the further into the funnel fraudulent activity happens, the more metadata becomes available for addressing it. For example, impressions (top of funnel) are often reported with the server IP and user agent, versus clicks (mid-funnel) that in most cases are reported from the device. Installs (lower funnel) are always reported from the device. Few advertisers have the scale to build meaningful metadata profiles of devices in order to block known fraudsters and flag suspicious devices for further investigation. It therefore falls to those with access to this metadata at scale to build solutions and watchlists that publishers across the industry can use to address their own mobile fraud.

As an industry, we need to take bold and swift moves to block known bad actors — where they are clearly in the black — and flag suspicious device-based trends. The time for baby steps is long past because the fraudsters are already sprinting. But while fraudsters may be clever, marketers can join forces to beat fraud. Armed with granular data, adaptive anti-fraud technology, and multi-layered fraud prevention that uses ecosystem-wide, device-level insights, we can remove those blind spots once and for all.

Dig Deeper: Learn everything that’s happening in mobile fraud activity with AppFlyer’s recent study — The State of Mobile App Install & Engagement Fraud.

Matan Tessler is Director of Product Management at AppsFlyer.

Sponsored posts are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. Content produced by our editorial team is never influenced by advertisers or sponsors in any way. For more information, contact sales@venturebeat.com.