By Neil Spector
Today’s personal lines property insurance underwriting departments are under increased pressure to trim operating budgets, yet at the same time they are expected to drive growth and improve risk assessment. Key underwriting decisions revolve around property inspection programs and the paradox they present to insurers. Inspections are a significant underwriting cost but also serve as an important — but difficult-to-measure — tool in making accurate pricing and risk selections.
Verisk estimates the property/casualty insurance industry spends more than $100 million annually on homeowner inspection programs. But many carriers admit their inspection selection process is often not much better than throwing darts at a board. Low-risk homes are sometimes inspected needlessly, while homes with “problems” are identified only after a claim. The situation is exacerbated because claim losses — especially from roof-related claims — have skyrocketed in recent years because of severe weather events, such as hailstorms and straight-line winds. This has many carriers reevaluating their inspection programs.
Insurers can realize significant savings and better risk assessment when they target costly physical inspections using data and science. The challenge is to find and isolate those risks that require an inspection or proper pricing, underwriting, or the identification of risk factors that the insured can address. Traditional guidelines that drive inspections, such as the age or value of the home alone, are not enough to optimize carriers’ inspection spend. So, how can carriers achieve more precision with their inspection programs?
Benchmarking portfolio replacement values to industry averages can help confirm an underwriting strategy (for example, writing high- or low-value homes) or help identify potential over- or underinsurance.
The first step is to determine what modifications can improve the decision criteria that drive the inspection ordering process. Simple rule changes that examine new internal or external data points can help make the process better. For example, detailed weather information can focus on properties within an area that experienced a recent hail event. With today’s data analytics and technology, insurers can use predictive analytics to optimize inspection spend by employing a model that systematically identifies the risks most likely to have hazard conditions or insurance-to-value deficiency issues. A model that feeds inspection results back into the system can “learn” and become more predictive over time as more data is evaluated.
Another important component to help guide a carrier’s proprietary inspection process is the use of aggregated industry data. Knowing how your company stacks up against industry benchmarks will provide valuable insights into risk-specific decisions. This solution takes more time and effort to develop and implement than modifying current inspection rules but will provide much greater return on investment.
The challenge in building a predictive model is obtaining the right data sets, because insurers will typically need to supplement their own policy and claims data with third-party information and aggregated industry profiles. The right mix of data points will allow them to build a model that recommends risks to inspect based on characteristics tied to predictors of high frequency and/or severity of loss or replacement cost issues.
An important component of an inspection program that is often overlooked by many carriers is the reinspection process. New data sets and emerging technologies support the idea of ongoing policy monitoring. Risks change over time. Targeting a smart reinspection program as part of the renewal and portfolio management process is as important as the new-business underwriting phase.
Since costs for physical inspections can range from about $25 to more than $150 depending on the type of inspection, a truly optimized inspection program will contain a model capable of identifying properties that require mitigating actions or those that need a change in insurance to value. The model should also identify the type of inspection most appropriate for each risk. A model using policy information coupled with third-party data could help determine whether to conduct an outside or inside inspection and what areas may be of concern, such as the roof, plumbing, or electrical wiring.
For some risks, an actual physical inspection by a professional is not necessary. An optimized inspection program can identify policies for which telephone or mailed surveys could be used in lieu of a traditional inspection to obtain important underwriting information. Some carriers today are also making use of remote imagery and other data points to detect hazards, verify property characteristics and conditions, or identify possible insurance-to-value issues. Another emerging technology that can become an important part of an insurer's inspection program is the smartphone. As the use of mobile devices proliferates, forward-thinking carries can leverage the inherent technology within those devices by allowing policyholders to verify and document details about their property to identify problems, mitigate potential hazards, and ensure sufficient insurance-to-value coverage.
In today’s competitive marketplace, underwriting departments must take crucial measures that enable them to develop a better risk management approach to their book of business. By improving and optimizing inspection programs, insurers take an important step toward streamlining risk assessment while cutting underwriting expenses and improving risk selection and pricing.
Neil Spector is president of Verisk Insurance Solutions – Underwriting, which develops and delivers underwriting information and services for the property/casualty insurance industry.