The term “big data” has been thrown around so much that most people probably don’t even know what it means. “Data? Sure we have data,” they say. In fact, most organizations have too much data. There is a belief that the massive amounts of data now being generated goes mostly unused as resources and/or knowledge of how to use that data is lacking.
When trucking telematics systems started collecting data – and we’re talking complete vehicle data, not just simply location data – it opened up a whole new world of opportunities. Throw in digitized datasets that surround the vehicles, such as CSA and driver records to name just two, and that data suddenly has additional uses.
While more and more fleets are utilizing thousands of data points to make operational decisions, insurance providers are tapping into those same data points to help create more accurate risk profiles of carriers and their drivers to build improved insurance offerings for those fleets. And at a lower cost in some instances.
Insurers such as Reliance Partners are using available data to find markets that fit a fleet’s profile best.
“It gives us a summary of an account and we can take it to a market that is best for that risk,” Andrew Ladebauche, CEO of Reliance, explains to FreightWaves.
Using data such as CSA scores helps insurers identify the risk a carrier poses, but Ladebauche says the data usage goes beyond the obvious.
“They are able to track even down to the VIN number of vehicles historically, allowing you to identify chameleon carriers,” he says. “We’re able to track that all the way down to say that Company X that was shut down in 2015, now those trucks are being used by Carrier Y in 2017,” which allows an insurer to dig deeper and determine if that new carrier may pose a risk or is operating illegally.
From an insurer’s perspective, the use of the data is helping generate a more accurate pricing model, Ladebauche notes, although fleets may not have seen a decrease in premiums due to the size of awards from lawsuits.
Data is also helping insurers identify drivers that may have previous violations or issues and switched carriers to avoid detection. That, obviously, adds risk to the new carrier.
“[Data] gives us a summary of an account and we can take it to a market that is best for that risk.”
Insurers are also creating more dynamic pricing structures, as Ladebauche explains, and could potentially build more personalized risk pricing models, as noted by Master’s in Data Science, which looked at big data and insurers.
“Through a judicious analysis of big data, insurers have now been empowered to improve their pricing accuracy, create customized products and services, forge stronger customer relationships and facilitate more effective loss prevention,” it said.
Despite this potential, though, research conducted by Marketforce, the Chartered Insurance Institute (CII) and the Chartered Institute of Loss Adjusters in conjunction with Ordnance Survey, found that 95% of those surveyed believe insurance underwriting departments do not have the necessary tools to effectively utilize the data in making insurance decisions. Still, 82% believe insurers who do not utilize big data will become uncompetitive.
“This report provides food for thought for underwriting teams preparing for advances in analytics and real-time pricing, helping them to identify the necessary skills sets that will be critical in enabling underwriters of the future to survive and thrive to the ultimate benefit of their customers,” said Ant Gould, director of faculties at the Chartered Insurance Institute.
Insurance data scientists are collecting data from a number of sources, such as telematics devices, smartphones, on-board diagnostics, credit reports, government statistics and even social media.
The Master’s In Data Science report pointed out three future scenarios for insurance – personalized risk pricing, pay-as-you-drive (PAYD) and pay-how-you-drive (PHYD).
The personalized risk model utilizes historical data, combines it with analytical applications such as behavioral models based on customer profile data and real-time data from vehicle sensors, satellite data and even weather data, to create a detailed and personalized assessment of risk.
For instance, your fleet may be based in Texas, but data may indicate that 50% of your vehicles operate in Minnesota winters, creating more risk due to the conditions and potentially leading to more claims. This gives the insurer more insight into that fleet’s operations and helps it decide whether that fleet represents the right risk to insure.
The other two options, PAYD and PHYD, are becoming trends in automobile insurance. PAYD simply chargers auto users based on the miles they drive – the more they drive, the higher the premium costs. PHYD is similar, but it incorporates telematics data such as speed, acceleration, cornering, braking, lane changing, fuel consumption, geolocation, date and time and can allow insurers to recreate situations when an incident occurs. That, in turn, can help an insurer exonerate a driver when appropriate, also helping keep costs in line.
Progressive Insurance is predicting that by 2020, 25% of all automotive insurance premium revenue will be generated from telematics devices. With so much onboard technology on commercial vehicles, it will only be a matter of time before similar programs come to trucking.
Stay up-to-date with the latest commentary and insights on FreightTech and the impact to the markets by subscribing.
