Mobile ad fraud – market status

To understand just how big the impact of mobile ad fraud is, we need to take a broader look across main industry verticals, and the clear difference between non-gaming apps and gaming apps.

Gaming advertisers are known as savvy digital marketers, very data driven with high awareness to every point in their user journey, which leaves very little space for fraudsters to operate.

Gaming advertisers are heavily focused on engaged users rather than playing the numbers game and aiming for high install volumes. This translates into significantly lower CPI rates that are complimented by a sophisticated in-app and CPA structure. 

This structure encourages publishers to provide quality users that engage with their apps and directly translated into a very low install fraud rate of only 3.8%.

However, this doesn’t mean that gaming apps are immune to fraud, as fraudsters have begun shifting their focus towards these CPA in-app events, with rowing in-app fraud presence measured year-over-year.

Cross vertical fraud

Cross vertical fraud

On the other end of the install fraud scale we can find several non-gaming verticals, specifically Finance apps. High install fraud presence in finance apps can be associated to several factors such as:

  • Large scale marketing budgets
  • Lower awareness to digital KPIs – established banks or investment firms taking their “first steps” in digital advertising
  • Highest average CPI rates in the market 

Travel and shopping apps are not far behind, and also suffer significant install fraud rates due to their relatively high CPIs and marketing budgets. However, these verticals are generally more familiar with online KPIs, having historically operated online since the early desktop days.   

The average install fraud rate for non-gaming apps currently stands at 31.8%, meaning almost one in three app installs is fraudulent.

For a full vertical fraud rate review see AppsFlyer’s State of Mobile Ad Fraud 2020.

iOS vs. Android

When examining app install fraud by operating system, Apple’s iOS is significantly vulnerable, while Android users suffer from over 6x higher install fraud rates.

Apple’s iOS takes the walled garden approach, which includes a strict vetting process for apps seeking to enter its store, and thus creating a safer environment for its users. However, iOS devices still suffer from click flooding fraud attempts, as this fraud tactic bypasses store regulated defenses. 

iOS vs. Android fraud rate

iOS vs. Android fraud rate

Android, on the other hand, operates an open-for-all OS, which attracts fraudsters seeking opportunities and loopholes to exploit. Unlike Apple devices, Android devices also allow users to download off-store apps. These apps can be found outside the traditional app stores (GooglePlay and AppStore) and offer the stripped APK version of the app shown to the user.

These stores are open to everyone, without ANY filtering process, and malicious apps often infect them. This accumulates to Android’s high install fraud rate, as off-store apps often service fraud operations by injecting devices with malware and adware, without the user’s knowledge or consent. 

Post attribution detection

Fraud detection attempts must carry on beyond the point of install attribution, as fraud attempts have been increasingly targeting in-app events over the past few years.

 

Post attribution fraud detection

Fraud methods constantly evolve and adapt to anti-fraud solutions - improving their ability to carry out install fraud schemes by avoiding install fraud detection logic. Additionally, fraudsters increasingly turn their attention towards more lucrative CPA targets rather than focusing solely on CPI rates, applying new methods that are specifically designed to bypass standard install fraud detection methods.

These new methods can only be identified retrospectively – after the install was attributed. 

By doing so, newly introduced fraud methods can be flushed out by assigning them to new fraudulent clusters and patterns that may have not been familiar at the point of their initial attribution. 

Installs that helped establish this new reasoning can then be denied after they were attributed to fraudulent sources. This can only happen retrospectively once their cluster reached sufficient statistical significance to accurately be labeled as fraud.

A long standing industry misconception claimed that all fraud attempts can and should be identified and/or blocked in real time. However, AppsFlyer’s unique post-attribution fraud detection solution discovered that at least 18% of fraud attempts on average can only be identified after the point of attribution – exposing yet another market blind-spot. 

These are fraudulent installs that would have gone unnoticed had there not been an additional retrospective protection layer.

Financial exposure

Accurately measuring the financial impact of mobile ad fraud is somewhat difficult, as the specific business implications per advertiser vary.

It is, however, possible to calculate the financial exposure to fraud during a given time period (the amount of marketing activity exposed to fraudulent attempts). AppsFlyer measures the definitive majority of mobile marketing activity worldwide ensuring its data is safe to rely on for an accurate estimation.

It is estimated that in 2019 about $4.8 billion were exposed to mobile ad fraud.

The current estimation for the first half of 2020 stands at: 

 

$ 1.5 billion

 

Global crisis

While some may associate fraud to be a region specific issue, this couldn’t be further from the truth.

Keep in mind this guide’s opening statement – Wherever the money is, fraud is likely to be there.

And there’s plenty of money to be made through mobile ad fraud worldwide.

Developing markets, who are less technologically advanced, are often less regulated than other more developed markets and are thus ideal for fraudsters to operate. However, through emulators, VPN proxies, and other technological tools borders become irrelevant, allowing fraud to easily infiltrate any country.

Global fraud crisis

Global fraud crisis

Some, more developed, markets like the United States, Canada, and Germany may present lower install fraud rates than others, however these markets often hold the majority of worldwide marketing budgets. This means that the financial impact on these developed markets with lower fraud rates could potentially be bigger than developing markets with higher fraud rates and less marketing focus.

Higher mobile device accessibility, developed economies, and global consumption culture are all key factors in making mobile ad fraud specifically and online ad fraud in general a global crisis.

Key takeaways 

  • Mobile ad fraud is an existing and ever growing industry issue, wasting billions of dollars annually.
  • Fraudsters are inventive and creative. Their methods evolve over time to adapt and bypass industry regulations and anti-fraud defence mechanisms. 
  • While mobile ad fraud is regarded under two main categories, methods under these categories can differ in methodology, technology and scale.
  • Fraud is a business. Fraudsters are ROI driven, and follow the money wherever it may be.
  • Mobile ad fraud is not limited to specific points in the user journey. Nor is it limited to specific verticals or countries. Wherever the opportunity is, fraud will be there.
  • An advanced and sophisticated mobile ad fraud solution is a necessity in today’s ecosystem.
    A solid and secure infrastructure, combined with an adaptive solution for identification and blocking of existing and new fraud methods is required for any online marketing initiative at any scale.

Click here to learn more about protecting your business from mobile ad fraud. 

Chapter List