Protect360 Fraud Protection

Fraud Protection
Fueled by the
World’s Largest Scale

Powered by unsupervised machine learning
and a unique scale, Protect360 blocks
known and unknown fraud at every
level of the conversion flow: before, during,
and after install.

Mobile Advertising is a Magnet for Fraud Attacks

Fraud targeting mobile app marketers is evolving faster than ever. The bad guys have gotten smarter, and it shouldn’t come as a surprise. Fraudsters follow the money and there’s a massive amount of money in play in mobile advertising. 

The threat of mobile fraud is rapidly escalating, draining marketing budgets at an alarmingly high rate and the bad news is: it’s not going anywhere.

Mobile Fraud is Pervasive

 

Eats your advertising budget

22.6% of global mobile media spend is wasted on mobile fraud

Pollutes your data

Encourages re-investment in the wrong sources and of users you have already acquired

Wastes your time

Reconciling invoices and trying to understand complex fraud reports drains your human resources

Trusted by The World's Best Companies

A Layered Approach for 360° Protection

Real-Time Detection and Blocking

Install authentication Secure, post-compilation SDK, hashing and additional security measures block fraudulent install events, replayed events and rehashing instantly, before a single install or event is recorded.

Clustering-Based Anomaly Detection Clustering data and analyzing behavioral patterns to identify non-human behavioral patterns in real-time. Sub-publishers trafficking in this non-human traffic are automatically cut off.

Validation Rules This proprietary platform allows marketers to define when installs should not be attributed based on a number of parameters, including fraud data. Includes the ability to define custom CTIT thresholds per app, region, partner or campaign.

Attribution Calibrator While other attribution companies will report fraudulent installs and event as organic, AppsFlyer removes fraud from the reported customer journey, showing a frictionless journey and attribution installs and events to the right channels.

Post-Attribution Fraud analysis doesn’t stop at the attribution level. AppsFlyer continuously analyzes and reports your data even after installs and events have been attributed. Once behavior is identified as fraud, any future installs with the same attributes are blocked in real time.

The State of Mobile Fraud 2019

Everything you need to know about the state of mobile fraud:

  • How fraudsters evade detection
  • Why the fraud rate jumped 60% in Q2 2019
  • Which regions and verticals are hit the hardest 
  • The extent of the impact of post-attribution fraud
  • Which new fraud tactic is not (and should be) talked about
Get the report

Fighting Back

No matter the fraud type, we’ve got you covered

Install Hijacking

Analyzing multiple signals including short CTIT, referrer mismatching, multi-touch distribution patterns and more to identify and block install hijacking attempts in real-time

Click Flooding

Analyzing CTIT anomalies, click-to-install conversion rates and multi-touch contribution rates to identify aberrant trends and block click flooding at its source, at the publisher and sub-publisher level

Device Farms

We’re maintaining active databases for 98% of the mobile devices on Earth, automatically blocking device farms. Sub-publishers trafficking concentrations of devices “new” to the database are blocked in real-time.

Bots

Our unique scale allows Protect360 to identify both highly targeted and widely distributed bots operating at both lower and higher volumes, blocking all bots and botnets in real-time

Fraudulent in-app events

Our machine learning algos keep analyzing traffic even after installs- following the users along the full customer journey and preventing in-app events from fraudulent sources

Fake Purchases

Receipt validation is a mechanism used as a secure way against the app stores (iOS and Android), validating that in-app purchases indeed occurred as reported

Behavioral Biometrics for Game-Changing Bot Protection

Bots are mimicking human behavior​ that appears to be completely legitimate. Using our unique knowledge and expertise in human behavior and mobile devices, Protect360 analyzes hundreds of mobile sensors simultaneously: battery levels, changes in the movement of the position (angles and rotation) and device tilts to determine the authenticity of the user – differentiating between humans and machines.

Adding the Post-Attribution Layer

Contrary to some widely-held beliefs, 16% of app install fraud is impossible to stop in real-time. The main reason for this is that some advanced fraud schemes cannot be identified based on individual actions of a script or bot. They are only identifiable later on, as patterns emerge at scale and contradict standard post-install user behavior.

AppsFlyer is the first attribution company to add this missing layer, streamlining the reconciliation process, so that advertisers get a report that can be used for settling up with media partners based on fraudulent installs – making it easier to detect fraudulent activity before they pay their partners, publishers and sub-publishers.

AppsFlyer’s new Post-Attribution reporting layer uncovers fraudulent activity that would, with other attribution providers, go undetected, and provides reporting tools to assist in reconciliation efforts – saving our customers millions of dollars every day.

With Protect360 we see a huge difference in our user quality and only pay for verified, real user-driven installs. This saves our team a tremendous amount of time, so they can focus on growing our suite of apps.

 
Arpit Patel | Head of User Acquisition