Data Science is The New DNA of Insurance Broking

Close up of young business people analyzing data using computer while working in the office

This latest Opinion piece is by Steve Wallis, Partnerships and e-commerce Director at Close Brothers Premium Finance.

In the last few years, data science has become much more important in broking, especially in personal lines. It is now an essential tool by which brokers stay competitive. Most personal lines brokers hold a huge amount of customer data and unlocking it can lead to many opportunities for growth in this intensely competitive market.

Data science is the technology equivalent of a 21st century crystal ball. By analysing millions of lines of data, brokers can assess at the point of quote how they expect a customer to behave. For example, will they carry a bad debt risk, how likely are they to renew, and what are their characteristics in terms of claims, paying by instalments, or buying add-ons.

The scientific techniques used to make these predictions are incredibly sophisticated, but the added advantage is that brokers can build their model and test it against the existing data set to confirm how predictive it is. This builds confidence in how much emphasis to place on the model as part an overall risk analysis.


Does this mean that insurance broking, at least in personal lines has become a different kind of industry altogether, and that traditional broking is no longer relevant?

What is clear is that new entrants into the market are set up to be digital, which includes data analysis, so the market is a mixture that include newbies as well as the established players. Many long serving broking people have adopted new ways of working and most, if not all understand the value in their data as it is data that is driving success for their brokerages.

Brokers who work in data science need a certain level of skill set. As a trusted partner, Close Brothers Premium Finance (CBPF) is able to help out, by producing models and fine tuning them, but the broker needs in-house capability, which means people who have some level of understanding to take the models, deploy them and refresh them periodically.

Therefore, it is not necessarily a heavy investment outlay in technology that is needed, but an investment in the right sort of people. Brokers require some analysis technology, and there are various programming languages and modelling techniques available; what our data science team does is work with brokers by tuning and adapting our modelling approach to match the capabilities of the brokers we work with.

At CBPF, we placed our data scientists with our broking partners to help upskill their in-house resources. This has been key to the success of a number of joint projects that we’ve run in the past and resulted in not only a functioning model, but we’ve also left them with a more capable team.

Data science works best for volume broking, and that is why it has been so successful in personal lines. At CBPF we have seen a trend for data science to move into commercial broking where there are volume lines, such as van insurance or small retail. Business lines with high volumes enable modelling techniques to come into play; more data equals more opportunity.


I predict that the trend for data science in commercial will continue, and brokers should embrace it.

Data science is not going to replace personal relationships, in fact it helps to improve your customer knowledge, and thus better service their insurance requirements. High street brokers with smaller books will continue to trade on their local community knowledge and experience, but volume commercial brokers are in the midst of a digital revolution and will inevitably follow personal lines down this route.

It may be that we’ll also see more personal lines brokers move into SME using similar data techniques they apply to volume home and motor. The purchase process is similar and the approach to pricing, being highly automated, is also the same.

However, there is no read across to larger value commercial and corporate broking. They won’t be affected.

At CBPF, we are looking at ways of leveraging our own data to help brokers generate advantage from our insights. Serving 1,600 brokers across the UK gives us an unprecedented volume of data which we are keen to analyse and feed back into the market.

Across volume broking, data science is ready to undergo the next phase of development. For the benefit of our brokers, and as a supplier and partner, CBPF is keen to do the heavy lifting required to bring our own data resources into play. Our data science team is working on a number of projects that will add considerably to the brokers’ armoury when it comes to better understanding their customers and we look forward to updating the market shortly.

About alastair walker 12109 Articles
20 years experience as a journalist and magazine editor. I'm your contact for press releases, events, news and commercial opportunities at Insurance-Edge.Net

Be the first to comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.