Other parts of this series:
In the first two parts of this series, we looked at how high performing carriers in the future will be much more proactive and forward-looking in how they manage their portfolios through the utilization of new data, new analytics and visualization tools, and fast and integrated processes.
The most important change that can be made, however, is to decentralize the insights generated by portfolio management. As portfolio management becomes more capable in different analysis techniques, in the use of external data, and in visualizations, the insights can be integrated into the underwriter’s risk analysis and the evaluation of the actions that should be taken. Improve the analysis with these considerations:
- How the risk characteristics of a particular account compared with those of accounts of a similar size, industry and location, using both internal and external data sets?
- The average pricing variability to the technical price compared to similar accounts for that specific risk?
- The performance and pricing of the agent for that type of risk?
- The pricing, profitability, and claims trend for this risk, across the carrier’s peer group?
The advent of new risks, increased automation, and predictive models has made underwriting an increasingly complex function. Portfolio management needs to evolve to provide management with the insights it needs to manage this complexity. This evolution is not about having better data and analytics. It is about improving the nervous system of the underwriting organization to be able to sense, assess, and respond to the market faster to produce superior underwriting results.
The questions insurers need to ask themselves are:
- Does my portfolio management function adequately evaluate and manage how I underwrite?
- Is my portfolio management function able to provide fresh insights rather than just base performance information?
- Am I set up to act rapidly on the insights and findings that my portfolio management is producing?
To learn more, register to download: