kasko2go: Making motor insurance more profitable through AI- and telematics based risk assessment
Commercial motor insurance has a loss ratio problem. The European motor insurance market is a huge market with revenues from motor insurance premiums of more than 135 billion euros in 2016. In the same year however, almost 104 billion euros were paid out in insurance claims. This means that more than 76% of income from motor insurance premiums had to be spent on claims, leading to low margins for car insurers. The biggest cost drivers for insurance companies are coming from a high loss ratio caused by bad risk in their portfolio. Motor insurance is seen as a low margin business, which hasn’t changed much over the years.
Traditional motor insurers continue to calculate risks based on factors like age, ethnicity, and gender. They also rely on motor vehicle records (MVRs) to understand specific driver behaviour, focusing on violations and driver eligibility. While those records are helpful, they don’t tell the whole story. Insurers can only assess actual risks in retrospect, meaning after the financial loss of claims has already been incurred.
Kasko2go takes a fundamentally different approach. They have developed an AI- and telematics based scoring solution, which enables a much more accurate assessment of the risk of motor insurance clients. Their mission is to make motor insurance more profitable, while making the streets safer. Their latest innovation is Normal Sigma, a solution based on empirical, behavioral and location-based information. This open-source solution enables insurers to identify and categorize risks. It allows insurers to try out telematics scoring without having to invest in expensive infrastructure.
How it works
Normal Sigma is a new approach to telematics insurance, which takes into account not only the driving behaviour, but also the road conditions, the weather, and most importantly, the reliability of the data, that can be collected from the driver. Normal Sigma applies a sophisticated scoring model, based on the customer’s personal driving style. This enables it to identify and separate good drivers from bad ones. This way, insurers can create a community of drivers who are least likely to cause an accident. They’ll be rewarded with a policy price reduction of up to 50%. With this approach Normal Sigma will substantially reduce claim frequency and increase the profitability of motor insurance.
At the heart of Normal Sigma is its in-house developed scoring model, which uses personal driving parameters as well as statical, statistical and dynamic parameters. Some examples of these parameters are phone usage, driving distance or time between anomalies, but also environmental factors like traffic, weather, driving location, driving obstacles, to name just a few.
Install the app – and start driving
To use Normal Sigma’s pioneering, cutting-edge technology, drivers only need to install the app and start driving. The behaviour of a driver differs based on his immediate environment. Especially in situations with high traffic density, a driver is more likely to get into an accident – this lowers his or her score. The reverse applies as well.
In traditional motor insurance, good drivers are disadvantaged, because they overpay for their insurance, to balance out bad risks. Kasko2go’s solution enables insurance companies to precisely identify their risks and categorise them. This makes it possible to offer good-risk-attracting policies without losing money for claims to good drivers and fill their portfolio with more profitable policyholders.
Tackling data reliability to improve risk assessment
The underlying assumption of the current telematics solutions is that getting more accurate data is the key to a correct risk assessment of each individual driver. Driving habits like acceleration, speeding and braking are at the basis of almost any telematics solution on the market. Unfortunately, there is no such thing as perfect data, nor is there any definitive research that shows the direct connection between braking, speeding and acceleration and the accident rate of a given driver. This is why kasko2go has taken a very different approach, incorporating into Normal Sigma mathematical models that are able to make accurate assessments, even when the data is not perfect. Instead of just focusing on speed, braking and acceleration, and trying to connect these to accident rates, the team has shifted their attention to road conditions, location and weather. They’ve analysed the parameters that actually affect the accident rate and created a system that can monitor, follow and predict them. This leads to much more accurate risk assessment, enabling insurers to better identify and categorise risks.
Car UBI Solution now available open-source, free of charge
Kasko2go believes there is a big potential for the motor insurance industry to fundamentally overhaul their system of risk assessment and improve their profitability. That’s why they are making their Car UBI Solution, which they’ve been developing over the last 3 years, available as open-source and free of charge to the motor insurance industry.
Who is kasko2go?
Kasko2go is headquartered in Zug, Switzerland. The company has around 35 employees of 10 nationalities, based in 3 countries. They employ more than 40 engineers and scientists with high-tech, scoring and telematics experience. Kasko2go has more than 5 years of research into the driving behaviour of more than one million drivers. Over 150,000 drivers have tried kasko2go’s products and more than 10,000,000 kilometers have been driven using kasko2go’s apps.
“Why not pay only for your own driving behavior? You pay as and how you drive. We want insurance justice for everybody. Safe drivers should be rewarded by paying less, no matter what nationality they are. Especially as they all contribute to safer streets.” Genadi Man, CEO and Co-Founder, kasko2go
Check out their DIA Prime Time video here: