Promotion Abuse
Detect Promo Exploitation by Bad Actors
Promotions are an effective way of acquiring and retaining users of your platform. However, many seemingly straightforward promos are susceptible to unintended abuse by malicious actors, negatively impacting user experience or resulting in direct loss of revenue. DataVisor’s scalable machine learning models scan through millions of user events in real time, automatically identifying patterns and correlations in user activity. New anomalous patterns in promo usage and transactions are surfaced automatically, allowing platforms to identify and close loopholes before the damage spreads.
Common Attack Techniques

IP obfuscation
Fraudsters utilize proxies, VPN, or cloud-hosting services to take advantage of regional specific pricing or promotions

Location faking
Cybercriminals use software to fake their GPS location and take advantage of geo-specific mobile promotions and pricing

Device obfuscation
Fraudsters utilize mobile device flashing, virtual machines and scripts to appear as though the login events are coming from different devices

Multi-user abuse
Sophisticated fraud avoid detection by accruing promotional credits or currency on one account and cashing out on another
DataVisor Online Fraud Report
Learn about the techniques fraudsters use to evade detection
Why Unsupervised Machine Learning
Promotion abuse is one of the most difficult types of fraud to detect reliably. By its very nature, many promos are one-off events that do not exactly match any previous campaign, which means that potential avenues of abuse are also unpredictable and vary from promo to promo. Unlike rules-based or supervised models, DataVisor’s unsupervised machine learning approach is able to automatically correlate signals in user data and surface anomalous patterns without any knowledge of previous abuse cases. This makes DataVisor uniquely capable of detecting and providing a safety net against abuse of all future promotions.

Unparalleled accuracy and coverage
Catch all bad actors involved in an abuse case without negatively impacting good users

Unknown threat protection
Uncover unexpected abuse patterns for new promotions without any training data or labels

Early detection
Detect abuse before it spreads, limiting impact to your revenue and good user base
Key Features
User linkage visualization
Analyze linkage and shared attributes among bad accounts to understand the entire scale of abuse
Automated Rules Engine
Automatically create and sunset new rules to help understand how new promotions or features are being taken advantage of
Real-time detection
Limit financial loss by detecting and shutting down abuse cases at a formative stage using DataVisor’s real-time API

Account linkage view to discover hidden links among malicious accounts
Getting Started
Want to get started and find out how DataVisor can help find malicious accounts hiding inside your online service? Request a security assessment today!