White paper: Cracking the mobile code

Best practices for a mobile telematics program

Insurers today are eager to leverage the predictive power of mobile telematics to produce more accurate insurance rates. But even small mistakes in models for mobile driving behavior can have costly consequences. Read our white paper to ensure the right knowledge and experience is applied to reap a true return on your telematics program investment.

Key takeaways

  • Smartphones are rich mobility data sources
    Mobile telematics programs leverage smartphones to collect granular mobility data—such as GPS, accelerometer, and gyroscope readings—that reveal individual driving behaviors, including distracted driving. This data is more personal and behavior-specific than vehicle-based data from OBD-II devices.
  • High-quality mobility data is essential for accurate modeling
    Effective insurance risk models require large volumes of high-quality mobility data matched with actual claims. Challenges include imperfect GPS signals, operating system discrepancies, and the need to distinguish benign phone movements from risky driving behaviors.
  • Mobile data outperforms proxy-based auto insurance models
    Models built on actual mobile telematics data and insurance losses are significantly more predictive than those relying on proxies like simulated collisions or OBD-II data. Proxy-based models can misprice risk and reduce rating accuracy.
  • Mobility data is the key to curbing distracted driving
    Mobility data from smartphones enables insurers to detect distracted driving, a behavior not well captured by traditional telematics devices. Arity’s research shows that the most distracted drivers have a loss cost 1.5x higher than the least distracted ones, making this a vital variable for pricing accuracy.
  • Continuous mobility data collection enhances reliability
    Ongoing collection of mobility data allows insurers to track changes in driving behavior over time, improving model reliability and enabling proactive customer engagement, such as feedback on driving habits or emergency assistance.