In the insurance industry there is around £1.3 billion worth of fraud every single year. With regards specifically to temporary car insurance policies, the overall percentage of fraud that occurs is much worse in comparison to annual policies. This was a huge hurdle that we needed to overcome when starting out, considering we wanted to sell policies 24 times shorter than what currently existed.
There are different ways by which people try to commit fraud in temporary car insurance, some examples include:
To prevent against these we use a few services and solutions in order to minimise fraud as much as possible.
After adding in your first name, date of birth and sex, we automatically generate the majority of your driving licence number except for the last 3 digits, which are random.
After entering the remaining 3 digits, we run a check through a service created by the DVLA, called MyLicence, and from this we extract your driving licence details. If you enter in false information, such as your age, then these checks will fail and you won't be able to move forward.
To prevent people from neglecting to declare any incidents that they have been involved in, we run a check against an individual's declared incidents.
If this doesn't match up with their declared incidents, they will be unable to go ahead and purchase. One thing that is slightly different about this process - as opposed to buying normal insurance - is that we need to know about all incidents that a person has been involved in in the last five years; not just at-fault claims.
When you first sign up for Cuvva we ask you to take a selfie, as well as a picture of your driving licence or passport, in order to verify your identity.
The method by which we catch out those who have crashed a car is surprisingly easy now thanks to mobile phones. We simply get the customer to take a photo of the front of the vehicle before purchasing a policy, so we can see the condition of the car they're looking to get insured on. Magic.
Our payment processor, Stripe, handles all of the sensitive data such as card details, and all of the transactions that are made through Cuvva. This enables us to have a very high level of security, because we don't have to store anyone's payments details internally. Stripe also has some smart rules that must be passed in order to be able to purchase a policy.
On the backend, we use a machine learning system that takes into account thousands of different data points, to then determine if someone is potentially fraudulent or not. It rates someone on a scale of 1-100, and if their score is too high it stops the user from purchasing a policy (though we can override it if it makes the wrong call!)
By having these systems in place, we can help prevent fraud and provide the best possible pricing.
That's a quick overview of our systems and how we help prevent fraud 😊 If you have any questions or feedback then let us know via the in-app chat.
Follow us on...