As I discussed last week, risk scoring is a powerful way to evaluate risk by including disparate and complex datasets, business rules, and experience — all prioritized and weighted — in an algorithm. While it’s not a new idea, it is underutilized, and therefore I think it's worth looking at a few examples of how it's done. (Since it’s a vast subject, this post will concentrate on flood risk scoring.)
The most common type of flood risk scoring is the combination of flood likelihood (return period) and depth. This combination of frequency and loss-driver information is possible for all perils, and offers greater insight for underwriting than just a zone assessment. The score is determined with a matrix of values, such as this:
It is possible to incorporate more variables into a flood risk score, however. For example: RiskMeter Online™ (from Core Logic) has a Flood Risk Score that is based on six variables: flood zone, terrain elevation, flood water elevation (static and dynamic), hydrology, and levees/dams; and it calculates a score from 1 – 100. Once a score is returned for a location, the user can apply business rules based on the score. The actual algorithm is proprietary, so users don't see how it is calculated.
Since risk scoring is a way to incorporate experience and organizational priorities, it's very useful for an insurer or a broker to be able to devise the scoring algorithm themselves — including deciding what datasets to use, how to weight them, and how to assign scores to the output. Not only does this let users of a risk score leverage their own experience (and claims data), it ensures the factors that are particularly important to the types of properties they underwrite (or broke) are used fully, while less important data can be made less influential on the final score.
InsitePro™ (from Intermap Technologies®), allows users to define the score that best fits their business needs, including choosing which factors to evaluate and how to evaluate them. Examples of variables that can be evaluated by InsitePro’s risk scoring include: local elevations (around a specified location), natural barrier detection between a location and water, distance to water, (more importantly) distance above water, and any other dataset a user needs to reference.
Once risk scores are determined, they can be applied to subsequent datasets to create exposure, damage estimates, or pricing/rating output.
Of course, having a risk score doesn’t predict the future of a location — we'll leave that to the fortune tellers and shamans of the world. But risk scoring does build efficiency and repeatability into complex evaluation processes. Communication about risks is simplified by distilling all the necessary information into a number, letter, or code that can immediately convey characteristics of the risk.