The importance of a view into the future

Which customer is likely to terminate its insurance in the future? That is one of the questions Rolf Mertens deals with as Head of Advanced Analytics at ERGO. In our interview, Mertens explains why prediction models are so important for an insurance company.

What is your personal background at ERGO?

Rolf Mertens: I studied mathematics and computer science at the RWTH in Aachen and joined ERGO in 1992. I started off in the IT department, where I was a part of the Software Development Team for seventeen years - specifically as Head of IT Coordination Composite. In 2009, I took over the role as Head of Private Customer Liability and Property Insurance Line Management. I really enjoyed the product development, portfolio management and the intensive cooperation with our various sales forces. Obviously, we were very pleased when we were even able to win multiple innovation awards.

During the development of our smart home product, I got in touch with Mark Klein (Chief Digital Officer at ERGO). He wanted to restructure the Advanced Analytics unit to consistently integrate the mathematical forecasting models that are developed into the operational business processes. By doing so, we would be able to generate real economic benefits. My knowledge of mathematics and IT, as well as the insurance knowledge I have accumulated over many years, led to my current role as Head of Advanced Analytics beginning 2019.

What exactly does a Head of Advanced Analytics do?

Mertens: Generally speaking, our team develops prediction models that we implement in various areas of the ERGO Group's value chain in order to improve them. For example, in both national and international Claims and Benefit Management or in Customer Care. My job is to work with my highly qualified team of Data Engineers, Data Scientists and Enablement Managers, together with the specialist departments and IT, to implement the relevant projects from the idea to practical use. The division was created in 2016 from a strategy project and meanwhile contains 27 employees.

Is it all about artificial intelligence?

Mertens: AI is the buzzword that many people know. However, it is turning into a marketing term. Not everything that has AI written on it also contains AI. Our department develops predictive models, where self-learning neural networks play an important role. But there are also a number of other mathematical methods to model predictions.

However, we don’t occupy ourselves with finding out why something happened, but with what is likely to happen and what we can do about it. To explain it with an example: If we see a decline in premium income in a business segment and one of the reasons is that we have a high termination rate, then this is the result of a classic cause-analysis. A look in the rear-view mirror, so to speak. Advanced Analytics, on the other hand, looks at the windscreen and predicts which of our customers will probably terminate their insurance contract in the future. We can then use that information to implement customer loyalty measures.

Why is this area particularly important for ERGO?

Mertens: It is in the DNA of an insurance company to look at the future and future events. Actuaries (Insurance Mathematicians), for example, derive the right premium level from this. In our division, we use these methods to "understand" incoming client e-mails for example and forward them to the responsible department to process. This eliminates the need for internal forwarding and the repeated reading of e-mails. Our customers benefit faster processing. Same mindset, multiple application possibilities.

What are the plans for your team in 2020?

Mertens: This year, we primarily want to use our analytic skills to classify incoming mail and optimize service processing in health insurance. These are large mass processes in which we want to automate routine work. That way, our employees save time they can invest in more complicated and important responsibilities. Real-time integration into our existing IT systems will be a major challenge that needs to be solved. Finding and preparing the right data in sufficient quantity, developing the business models on it and integrating it completely into the business processes so that it become beneficial for our customers, our sales partners, our employees and the company itself - that is the goal we want to be measured by.

What do you like to do in your free time?

Mertens: I enjoy spend time with my family, love carnival and like jogging, golf and travelling.

By Benjamin Esche