SCORviews hero

The Big Idea: Transforming Business
June  2021

The past year or two have been all about “big” things. Big companies have been throwing big money at Big Data, hoping to see big results. The big COVID-19 pandemic event created big changes throughout the economy, requiring a big shift in how we work. This is certainly true in the insurance space as well. 

Huge problems can develop suddenly and rapidly. Our time is short, but big problems take a long time to solve – so what can we do? 

While nobody knows the solution to all the big problems, there are many little things that we can do that add up to a very productive working environment. Let me highlight a few techniques from SCOR’s Data Analytics (DA) group that produce a productive, efficient and even fun working environment. These strategies are even more important as the post-COVID work environment may become increasingly remote.

Collaboration – Agile Framework
The Agile Framework is certainly not unique to SCOR or to the DA team. For those who are not familiar with this method or have never worked in this way, this web page created by gives a nice overview of the Scrum process, our particular choice within the Agile framework. 

What does our US-DA team (and any effective Scrum team) do particularly well? True collaboration. Part of the magic of Scrum is that the developers truly determine the “how” of doing the work. 

In many groups, this tenet is only on paper. But in DA, I as a new product owner have relied on the team for the expertise needed. The team finds problems and solutions that I never would have imagined. This enables me as the product owner to focus on the “what” of the project and prioritize our work items in a way that most benefits stakeholders. The result is a collaborative work environment that avoids being overly prescriptive.

Collaboration – git
Many developers use the open-source version control software called git. SCOR’s DA team has set up a best-practices framework of completing Sprint goals on a git branch, reviewing and merging into a master branch while still preserving a sprint-based directory structure. What this achieves, beyond powerful version control and efficient updating, is the history of an entire project at the tip of your fingers, easily linked to documentations and Scrum project management practices. 

The team consistently follows these protocols, meaning anyone can pick up any part of the project at any time. This allows parallel development where all contributors can push/pull their work independently. Or, if desired, you can pull in someone’s work and take it in a direction of your own!
Even if you are not a developer, I still encourage you to check out git. You can version control any type of file (not just python scripts), which has many uses and benefits. For example, a pricing actuary can make sure that they are running analysis on the correct version of the software (or can roll it back to any prior version).
Package Development
Most of us are not working directly as package (software) developers. But the process of constantly asking ourselves “how else can we use this?” or “how can we adapt this tool so that it is reusable for a different set of problems?” like a developer makes sure that the valuable work done for a particular workflow can be easily used in future workflows. This reduces redundancy and builds a toolkit that can be used by anyone – not just those with advanced Predictive Modeling experience.

For non-developers, I encourage you to think of ways to generalize your own work. Even a re-usable spreadsheet is a good place to start.
Cross-Functional Knowledge
Data science work requires data science skills. But that is not enough – life insurance has a lot of quirks, an extremely long legacy and extensive data relationships that require solid business understanding (e.g., post-level term insurance mortality). A good team will include members with data science skills as well as those with insurance industry knowledge.
Our DA team members come from a variety of backgrounds: underwriting, actuarial, “pure” sciences (including data science) and others. My advice is to increase your knowledge constantly. While it’s hard to openly admit when you don’t know or understand something, you’ll be rewarded for your bravery by learning something new. 

Finally, consider expanding the knowledge breadth of your own team. Bringing a subject matter expert into your technical work, for example, can generate good insights and ideas quicker and more often. Or, conversely, bringing a data scientist into a non-data business problem may lead to solutions that have never been considered or imagined. 
To achieve greater productivity in our fast-paced and mobile environments, we need good tools like the Agile framework and git. But we need more than the tools. The values of collaboration, openness, honesty, respect and the drive to achieve success across the entire organization are the behaviors that set organizations apart. If you would like to learn more about SCOR’s Agile transformation, please contact me or Jennifer Nusbaum.