Over the last couple of years, insurance carriers have been experimenting with sources for driving-related data.They have also been exploring ways to utilize this data to improve efficiencies in their underwriting and claims operations. With improved driving-related data availability,carriers are looking at influencing and improving driving behavior to reduce claims and improve loss ratios. However, carriers are struggling with the accuracy of the data generated;for instance, drivers can get penalized for over-speeding if the GPS generated from the app on a smart phone is not accurate.
While there is a trend wherein both the OEMs and carriers are collaborating to improve the reliability and accuracy of data sets, carriers are experimenting to unravel the complex co-relation between the data generated, interventions and improvement/deterioration in driving behavior – some going to the extent of onboarding professional psychologists to work with the analytics teams.
Shaping driving experience
Though theft detection and weather related alerts have been in place for some time now, drivers appreciate alerts to conduct proactive servicing of specific parts of the vehicle, and the avoidance of inconvenience after a breakdown. Drivers also like the concept of being incentivized by their carriers when their driving score improves, rather than getting discounts on the next premium. If the interaction is not perceived to be intrusive, drivers look forward to and evaluate the insights provided to them. They try to improve their driving scores to win incentives, and with intelligent gamification, enjoy friendly competition against their family members and friends. Gamification of the interaction with drivers is critical to sustain interest levels. To reduce fatalities in the young driver segment, some carriers are also creating interactions with parents, and measuring the changes in driving behavior post interventions.
Advanced AI capabilities help carriers to distinguish between hard acceleration on a flat road, and the same hard acceleration while driving up a slope, so that a driver does not get penalized on the drive score in the second scenario. The insights that can be generated are limitless and carriers are configuring and customizing the insights to suit their requirements for better underwriting and claims efficiencies. For example, if a driver has been driving for more than 3 hours at a stretch, at night, on hilly terrain in snowing condition, it is possible for a carrier to configure alerts to the driver, to stop driving, and advise hotel options in the vicinity.
In the unfortunate event of an accident, carriers are developing capabilities to alert emergency services, or if the accident is not serious, hail a cab service, or advise the driver about the nearest garage.
The possibilities in terms of the services that can be built-in are numerous. Data availability on G-force, direction and speed of the vehicle, before and after the collision, is helping carriers re-construct accidents and detect frauds, if any. If there is a dash-cam fitted, then forensics becomes even better. Advanced AI is now being used to predict future accidents and to identify whether a driver is at fault or not for the accident.
Most of the above examples are also relevant to drivers of vehicles of large fleet services, for example courier services. Today, a camera ported above the driver’s sun-visor, can measure the width of the driver’s pupils and eyelids, identify fatigue, send out alerts to stop driving, and can even go the extent of advising the driver of a safe place to park, considering there is valuable merchandise in the vehicle.
One will agree, that all the above, will ultimately help carriers improve their underwriting margins, and not only transform the claims process, but also enhance the experience for their customers. Implementation and achieving the goals is going to be challenging for carriers in each of the phases, whether it is acquiring data, creating insights, or acting on the insights. Carriers are realizing that there is a substantial element of “do, learn andmodify” in this journey roll-out, like any big data analytics-based program.
