Quago
Quago provides a behavioral intelligence platform that converts in-game device sensor data—including touch metrics, accelerometers, and gyroscopes—into actionable insights for gaming companies. The platform detects cheating, predicts player churn, and identifies post-install CPA fraud using machine learning models trained on over 100 billion sessions, all without collecting personally identifiable information. Its lightweight SDK integrates with minimal performance impact and delivers real-time detection and reporting.
- Gaming & eSports
- Data & Analytics
- Artificial Intelligence
- Cybersecurity
- Software Only
Funding
Funding not disclosed
Founders
Founder details are not available yet.
Product
Problem
Online gaming companies face significant revenue losses from in-game cheating, user acquisition fraud, and player churn. Traditional analytics tools fail to identify critical behavioral patterns, and existing anti-cheat solutions often miss sophisticated cheating operations, leading to skewed analytics, wasted marketing spend, and damaged player trust.
Solution
Quago converts in-game device sensor data into behavioral intelligence that identifies cheating, predicts churn, and detects post-install CPA fraud. The platform analyzes hundreds of data points from each player interaction—including touch metrics, accelerometers, gyroscopes, and device tilts—using proprietary machine learning models trained on over 100 billion sessions. Pre-trained generic models identify inauthentic behaviors immediately, then adapt to each game's unique player base over two to three weeks for customized detection. The solution provides real-time insights and evidence-based reporting, enabling gaming companies to recover lost revenue, optimize user acquisition, and segment players by predicted lifetime value and skill—all without collecting personally identifiable information.
Target Audience
Primary customers are gaming companies, including AAA game studios and mobile game developers, who need to protect revenue, optimize user acquisition campaigns, and improve player retention through behavioral intelligence.
Features
- Patent-pending sensor-based signal analysis capturing hundreds of data points per swipe, including touch metrics, accelerometer, gyroscope, and device tilt data
- Hybrid supervised and unsupervised ML models that adapt to each game's player base within two to three weeks, with continuous learning from new patterns
- Cheat detection with a 0.000000016 false positive rate and 10x recall improvement over comparable alternatives
- Real-time detection and reporting for immediate response to cheating, churn risk, and fraud events
- Lightweight SDK (~100KB compiled) with 0.07% impact on device performance, negligible battery consumption, and no new user permissions required
- Privacy-compliant architecture with no PII collection, fully compliant with GDPR and Apple privacy requirements
- pLTV and churn prediction capabilities available as early as 24 hours post-install
- Evidence-based fraud detection providing clear documentation for reimbursement claims