App Install Ad Fraud

Paid install fraud is a growing problem for mobile app developers, with fraud rates often exceeding 5% for some ad networks. Detecting these fraudulent installs is becoming ever harder as fraudsters camouflage their installs with increasingly sophisticated techniques like location and device spoofing, click jacking, and simulated user activity. DataVisor’s Unsupervised Machine Learning Engine is uniquely capable of detecting these fraudulent installs because analyzes all accounts and events simultaneously, uncovering the hidden connections between fraudulent installs. This allows it to detect entire rings of fake installs at once, even when each install is not suspicious when analyzed in isolation.

Request Trial

Common Attack Techniques

IP obfuscation

Fraudsters utilize proxies, VPN, or cloud-hosting services to hide their tracks from IP blacklists and reputation-based systems

Fake retention events

Mechanical turks fake post-install activity in an attempt to appear more authentic and evade rules-based detection

Device obfuscation

Fraudsters utilize mobile device flashing, virtual machines and scripts to synthesize massive numbers of unique devices

Arm and swarm

Attackers test techniques at a small scale to stay under the radar, and launch massive campaigns after they identify detection gaps

DataVisor Online Fraud Report

Learn about the techniques fraudsters use to evade detection

Download the report

Why Unsupervised Machine Learning

Mobile app installs are a relatively new target amongst fraudsters, and the attackers are also some of the most sophisticated. Every day, fraudsters test and apply techniques such as device emulation, cloud hosting and mechanical turks to appear legitimate and authentic. Unlike rules-based or supervised models, DataVisor’s unsupervised approach is able to identify the complete set of fake installs in an attack by correlating subtle patterns in the data, including never-seen-before techniques that take advantage of detection gaps. This means that your protection against fraud can always stay a step ahead of the fraudsters – not trailing behind.

Unparalleled accuracy and coverage

Catch all members of the fraud ring without negatively impacting good users

Unknown threat protection

Automatically uncover evolving attack campaigns without any training data or labels

Continuous detection

Maximize detection by analyzing all data points, including post-install app launches and in-app events

Key Features

Easy integration

Configure your attribution provider such as Adjust or Appsflyer to automatically stream data to DataVisor in a few simple steps

Fake install reports

Generate compelling reports of fake installs by ad network to collect make-goods that can be re-invested in growth

User linkage visualization

Show linkage and shared attributes among fraudulent installs to identify fraud patterns and discover new attack techniques

Ad buying decision intelligence

Create custom charts to identify which ad network has the lowest fraud rate and provides the best ROI

Custom report view to determine which ad network or campaign provides the highest ROI

Bringing Real Value to Our Customers


DataVisor Joins Forces with Top Mobile Gaming Company to Fight User Acquisition Fraud

Learn how DataVisor helped one of the most well-respected gaming companies in the world, with a massive install base of more than 300 million users across 180+ countries, to fight user acquisition fraud such as fake installs using click farms.

Read Case Study

Getting Started

Want to save millions in marketing dollars per year? Request an mobile ad fraud assessment today to see how much you could be saving.

Request Trial