Novarica SVP Mitch Wein and I were looking forward to another fascinating Special Interest Group meeting on analytics this month. The group touched on many areas of this broad field, with a particular focus on insurers’ use cases for analytics, data governance, business intelligence, and data catalogs. We were pleased at the level of participation and idea sharing that occurred.
Analytics Use Cases in Insurance
Novarica research shows widespread industry interest in underwriting, claims, and customer use cases for analytics. A midsize Midwestern insurer shared its experience with overlaying historical CAT data with property exposures to identify areas for expanding underwriting or limiting growth. As often happens, the biggest challenges were related to human factors. The carrier reported that the biggest hurdles were preparing the sales force for changes in appetite and in the best means for communication. More recently, the team used the same analytics and data during the derecho that wreaked havoc across the Midwest in August. This enabled the insurer to predict losses with great accuracy.
Data Governance
No Novarica Analytics Special Interest Group meeting is complete without some discussion on data governance. Data governance was suggested as a solution to standardizing analytics and KPIs. The problems are with people and organizations creating their own version of “the truth.” Different sources, timing differences, and variation in the formulas that drive KPIs are the usual suspects. A data governance program that holds people accountable is key. This perspective was further endorsed by a participant who mentioned, “The lack of business/process ownership around KPIs is an inhibitor. Once you lock in business ownership and can define the formulas and data to use, the rest falls into place.”
Business Intelligence
One monoline workers’ comp carrier made a strategic investment in a modern BI solution that ended up being transformational. The initial goal was to extract themselves from “Excel Hell” as the big sheets, weak data governance, and collaboration challenges were negatively impacting the organization. A new solution was brought in after a thorough vendor selection. This led to essential organizational change as the role of “spreadsheet jockey” was eliminated. The enhanced collaboration drew additional participants into the data analysis. Data quality improved significantly and, more importantly, new insights were discovered.
Data Catalog
A participant from a large insurer shared their positive experience with a recently implemented cloud-based data catalog. In addition to capturing critical metadata, the organization’s data scientists are using it to document their analytic work and “loving it.” The same organization developed a creative solution in Python to automate data classification and look at hundreds of databases to identify potentially sensitive data, achieving a 95% success rate. This information is being used to help prioritize data security and satisfy data privacy regulations, such as the California Consumer Privacy Act (CCPA).
We look forward to continuing the highly engaged discussion with data and analytics professionals at our next Analytics Special Interest Group on December 3rd.
Add new comment