Workloads: Data Warehouse, Data Science & Engineering, Operational Database
Components: Apache HBase, Apache Impala
Stolen Vehicle Tracking
Delivers industry leading driver scoring models, pricing algorithms, crash alerts and claims reconstruction to improve insurer services and reduce costs
Decreases claims management process from weeks to one hour
Supports goal to reduce innovation cycle times by 50 percent
Doubled business volumes
Big data scale
11 billion data points analyzed daily from five million connected cars
Octo Telematics has transformed how insurers assess risk, deliver crash and claim services, and manage customer relationships with the ability to analyze 11 billion data points daily from five million connected cars.
Octo Telematics is the leading global provider of telematics and data analytics solutions for the auto insurance industry.
By collecting and analyzing data from connected cars, Octo Telematics gives insurers insights to more effectively assess driver risk, deliver crash and claim services, and manage customer relationships. “We utilize every type of data—contextual data, driving data, behavioral data, and crash data—to forecast driving habits, improve crash notifications and response, evaluate crash dynamics, and detect fraud,” said Gianfranco Giannella, COO, Octo Telematics.
As Octo Telematics grew, executives sought to replace a custom-made data platform with a more scalable, next generation data management platform. “We wanted to rapidly expand the footprint of our services,” said Giannella. “We needed a platform that would support a growing volume of telematics and IoT data and enable us to prototype services and products much faster.”
Octo Telematics today powers its telematics Internet of Things (IoT) solution with Cloudera Enterprise. The platform stores, processes, and analyzes data on more than 170 billion miles of driving and approximately 400,000 severe crashes from five million connected cars. In all, Octo adds over 11 billion new data points from connected cars daily to the platform. Internal and external data sources, such as traffic and weather data, are also incorporated to provide additional context.
Using machine learning, the company can make more accurate predictions and risk models.
The same behavior, such as a certain acceleration pattern can be normal in certain weather or traffic conditions, but not in others. The granularity and self-learning capabilities of our algorithms provides context to improve risk forecasting and crash reconstruction.
-Gianfranco Giannella, COO, Octo Telematics
Additionally, modelers can quickly test new ideas, try different modeling techniques, and refine models on the fly to produce the best results—using data volumes never before possible. “We can experiment and introduce new products to market at a faster pace,” said Giannella. “Our target is to reduce our innovation cycle time by 50 percent.”
Octo Telematics runs its applications on Cloudera both in the cloud and on-premises. “We are in the business of providing Platform-as-a-Service solutions to our customers, and cloud is an essential piece of this,” said Giannella. “However, sometimes for technology or regulatory reasons, we also need on-premise services.”
The insights gained help insurers deliver a completely different customer experience. “We are utilizing the power of IoT and data analytics to transform the insurance industry and improve customer safety and experience,” said Giannella. “For example, in the case of an accident, because the technology understands the type and severity of the impact, if the policyholder can’t pick up the phone, the insurer can contact authorities to get help. Additionally, the technology can help reduce the time to manage the entire claim process from weeks to one hour. This includes analyzing liability, assessing damages, and steering the car to a repair shop.”
For Octo Telematics, these capabilities translate into greater business success. “Since we’ve gone with Cloudera, we have doubled the volume of our business,” said Giannella.