Elizabeth M. Riczko

Past Awards

Innovative Applications in Analytics Award: Second Place

Insurers typically sell their products through three different distribution channel types: direct-to consumer, captive agents, or independent agents. For an insurance company that sells through the independent agency channel, they compete for market share, not just within a geographic area but within its agencies as well. Unable to exploit connections between past, present, and future events, the insurer makes decisions based on gut feel – legacy relationships and anecdotal evidence. In addition, the insurer has an incomplete understanding of the agency’s market and the “resources” at the agency’s disposal. This lack of understanding leads to inconsistent decision making for appointing, servicing and culling agencies. What insurance companies desire is a fact-based, decision-making process for managing their agency distribution channel such that it is productive and profitable. This paper discusses how Data Envelopment Analysis, Predictive Analytics, Collaborative Filtering, and Reporting were combined into a single, end-to-end process to meet these challenges. Results are discussed as well as areas for future research and on-going improvements.