Pricing data analysis: increasing importance of profitability improvement in the world of insurance

The client is a leading speciality insurer and reinsurer, part of a global top 10 insurance group. They are operating in and outside Europe through a range of different entities. The client provides insurance coverage for a range of risks to commercial enterprises, and reinsurance protection to other insurers around the world.

Initially, the increased workload and the short-term deadlines concerning the need for a pricing tool review, resulted in the client turning to TriFinance for support. The choice for TriFinance was based on the pragmatic approach and the knowledge sharing philosophy. The strong data analytics skills of the consultant combined with the insurance business domain expertise of the mentor contributed to a successful solution delivery which resulted in several project scope extensions.

This project consisted of multiple (data) analyses with two different profitability improvement objectives.

• On the one hand, we provided a review of the existing technical pricing models to offer a correct insurance premium to the final clients.
  Within an insurance company, technical pricing models play an important role. These are models that generate a premium based on a set of variables and parameters. Through these models, underwriters are quickly provided with a first indication of what target premium they can offer their clients. Due to the market constantly changing and the increasing amount of data available, it becomes more and more important to update these pricing models on a regular basis. The aim of this project was to assist the client with these updates by creating the first detailed analysis of the burning costs for a set of different Non-Life products and countries. Next to the creation of a well-structured and clear analysis, it was equally important to make the results accessible for further usage by the stakeholders. This meant creating dashboards in which the stakeholders could check the data, play with the parameters and align their own experiences with the results of the analyses. These analyses would be the first of their kind at the client and as such be used as a benchmark for future work.

• On the other hand, we performed a specific analysis in support of the launch of a new digital broker platform.
  Next to the update of the technical pricing tools, the client was also in the process of setting up a digital broker platform which would enable the offering of cross class packages to their customers. In order to support the decision-making process concerning these packages, we performed an analysis to gain insights on the cross-class profitability for several industries. Similarly, with the burning cost analyses, the results also needed to be accessible for real-time usage within the digital team. This meant creating a dashboard in which the digital team could perform simulations on the data.

On the one hand, the data analysis in support of the technical pricing tools resulted in three achievements:

• Increasing confidence in the quality of the data, creating accessibility to the results & allowing a modular use of it, and ensuring reproducibility in the future. First, the data quality had to be checked and improved in order to increase the trust in the data and in the results.
• Secondly, a set of different dashboards was created to make the results easily accessible. These dashboards enabled further analysis by the stakeholders, which in turn provided new insights and a better understanding of the burning costs in each Non-life portfolio.
• Finally, the analyses needed to be reproducible, this was ensured through a detailed documentation and a flexible SAS set-up.

On the other hand, the analysis supporting the broker platform resulted in three slightly different achievements:

• Aligning different segmentations, creating accessibility to the results and visualizing the impact of discounts. The first challenge was to align all activity codes and segmentations of the different classes into a single cross-class segmentation. Through a detailed analysis of the as-is situation, a uniform segmentation was created.
• Hereafter, an analysis on the different classes was conducted and a dashboard was created to make the results easily accessible. This dashboard enabled further analysis by the digital team and created the first insights concerning possible packaging propositions.
• Finally, the dashboard also included different discount parameters, which enabled the digital team to check the impact of discounts on the package profitability in real time.

This project was a first for me in many ways; my first data analytics project, my first project within an insurance company, my first time working remotely on a project… As a result, this project challenged me in multiple ways and enabled me to strongly develop my soft- and technical skills.

Firstly, both my SAS skills and my Insurance knowledge significantly improved throughout the project. By the end of the project, I was able to create elaborate analyses using SAS and challenge different stakeholders on specific insurance topics. Furthermore, I learned which data analysis steps to follow in a data journey in order to get the correct insights out of the data at hand.

Finally, as I was the sole responsible of each analysis, I had to take ownership, manage different stakeholders and communicate the results. It was the first project on which I was able to put these soft skills into practice. This project really required me to get out of my comfort zone, but as I can confirm now, it is out of your comfort zone that you learn and develop the most!