Insurance Claims and insurance data in general can be really fascinating data. When you think about how rates are set based on location, age, student grades, driving distance and even the color of the car, the actual analytics and analysis behind the data can be mind-boggling!
The 3rd dataset from Real World Fake Data was a dataset mocked up around Insurance Claims and I wanted to point out a couple of entries that I saw that were really well done and completely different approaches to the data.
This first is from Brigid where she approached it as a Pareto Analysis. By combining segments she could identify common demographics and see how the affect the rate and cost of claims. Really nice use of color and the interactivity with the ability to adjust the percentage of claims and the segment being examined give the user a nice interface to explore the data. The additional details below add those additional details around how many customers fall in to each segment.
The second viz is from Zak Geis, and his Claims Analysis. Here he is giving the user the ability to choose their own demographics on the left, and then highlight the distribution on those claims on the right. Really nice summary BANs to the left of the scatter chart, and the ability to change the X and Y axis on the scatter chart leads to multiple insight possibilities. Again, really effective use of color for highlighting the chosen demographics.
Thanks to everyone for their contributions!
The next dataset is now live on data.world based on Call Center activity, be sure and go check it out!