Add vehicle sighting data to insurance carrier predictive analytic models; you’ll be surprised at what you find out.
There is no debating the value of insurance carrier predictive analytics. Property and casualty insurers report that models built from predictive analytics help them to improve business performance. And like any business model, the better the data that goes into it, the better the model. A report by Willis Towers and Watson shows the top data sources in 2015 included usage-based insurance or telematics data, agent interactions, customer interactions, smart-home data and social media. All of these sources are expected to grow in the coming years. But our models show that auto carriers may want to add another type of data, vehicle sighting data, to their predictive models.
Vehicle Sightings Data Points to Fraud
Tests with carriers using vehicle sighting data are turning up new insights into garaging fraud, commercial use of a personal vehicle, radius and territory integrity, and final build fraud. All of these areas are rated because they are predictive of loss, and all of these areas are subject to fraud. But what happens when you add new data, vehicle sighting data, to the insurance carrier predictive analytics mix?
Let’s start with what vehicle sighting data is and isn’t. The data is gathered from license plate sightings captured nationwide all day, every day. The sightings include images of the plate, as well as the date, time and location of the sighting. Unlike telematics data, this data is not opt-in. Because all vehicles are required by law to have a publicly available license plate, all vehicles are opted-in. Capturing vehicle sighting data also known as vehicle location data, with license plate recognition (LPR) cameras simply automates a process that could be done using any phone or camera.
Personal Auto Garaging and Commercial Use
Now think about what this vehicle sighting data can tell you when added to predictive models. For Garaging issues the data could show:
• Vehicles seen at the garaging address have low relative loss ratios
• Vehicles seen at a new discovered garaging address have a higher relative loss ratio
• Vehicles not seen at all have the highest relative loss ratio
For Commercial Use issues, we know that vehicles used for commercial purposes have higher relative loss ratios. With the vehicle’s location, you now have data that can help you determine risk because vehicles operating in high-risk areas have higher relative loss ratios.
Using Vehicle Sighting Data for Radius, Territory and Final Build
For Radius and Territory, you know that vehicles operating outside of their Radius and Territory have higher loss ratios – vehicle sighting data can tell you if vehicles are operating outside of the Radius and Territory.
For Final Build if the vehicle location sightings show a final build with a higher risk than rated for, those vehicles will have a higher loss ratio.
And taking the vehicle sighting data beyond these cases, what else can it tell you and how could that data make insurance carrier predictive models smarter? The carriers we work with are often surprised to find out that they have much more fraud and data integrity issues on their books than expected. The vehicle sighting data gives carriers real insights about vehicle usage that they can’t get anywhere else. And the best part is, unlike telematics, the data can’t be disabled, policyholders don’t have to opt in, and there is no expensive IT integration. DRN is the leader in vehicle sighting data and analytics – we see over 5 million vehicles a day, covering 75% of the nation. What could that data tell you?