top of page
< Back

AI helping cut the costs of motor insurance




Driverly is a new Cardiff-based insurance provider offering flexible car insurance through a monthly subscription-based policy.

They also provide a Driverly App that assesses driving behaviour and calculates a price based on how a person drives. Safer drivers are then rewarded over time with discounts and other rewards. Designed for safe drivers, Driverly aims to go one step further than current insurance providers that offer the typical telematics ‘Black Box’ and base it on more than just a demographic by looking at multiple factors impacting how you drive.

Rewarding safer driving

The aim is to disrupt the car insurance market, particularly for young drivers, by developing a more personalised method of calculating insurance premiums. The Driverly app already had the capability of assessing the risk of routes based on demographics and driving style. However, they wanted to add another layer to how drivers are being assessed, helping them save money by rewarding safer driving.

Not your average telematics app

Driverly envisioned using artificial intelligence and machine learning technologies in the app to calculate the risk of a particular driving route. By gathering data based on the actual driving style of the driver, demographics, vehicle data, roads travelled, accident record, weather data and traffic intensity data. These combined factors would allow Driverly to assess each driver on a deeper level and give a personalised car insurance premium.

How did we do it?

CEMET and Driverly conducted research and development into using artificial intelligence and machine learning to build the UK Driving Risk Map. The goal was to obtain a reasonable estimation of the risk of any possible driving route in the UK. Through collaborative workshops, CEMET and Driverly concluded that each journey made in a vehicle carries risk factors. The data was gathered through multiple sources, and risk was determined to be comprised of 3 components: The Driver, through gathering data in the background from the driver's mobile phone. The Vehicle risk, by analysing accident data and via the MOT API. And The Route, which is computed by the creation of a network graph that represents the UK road network. Each section on a graph will have a risk associated with it, computed by analysis of accident and weather data, among other data sources.

CEMET mainly focussed their contributions on the network graph, with some input into the other areas. Adding this extra layer of data analysis meant the way drivers are being assessed is more accurate, which helps the driver save money and rewards safer driving.

With big plans for their platform development, they aim to disrupt the motor insurance industry and deliver significant savings to their customers. CEMET is proud to continue to support Driverly Insurance through the early stages of its development.

To learn more about Driverly’s latest work, visit Diverly’s website or follow them via Twitter or LinkedIn.


Go here for full article.

bottom of page