While many property owners understand the need for contents coverage, they often face challenges determining whether they have enough insurance. Homeowners and businesses often don’t have the time to inventory and price all of their furnishings and equipment. They also may lack the expertise to decide how much coverage they need.
To help policyholders address this challenge, insurers may provide guidance to help insureds estimate the coverage needed by using a percentage of each building’s value.
To help policyholders address this challenge, insurers may provide guidance to help insureds estimate the coverage needed by using a percentage of each building’s value.
However, these estimates can be imprecise. There are many other factors that can impact the value of contents in a property, including who is living or working there, how it’s being used, and its loss history.
The result can be insufficient coverage, leaving policyholders frustrated in the event of a loss and insurers missing out on premium dollars.
The benefits of artificial intelligence and robust analytics
Fortunately, artificial intelligence (AI) and advanced analytics can remove some of the guesswork from contents evaluations without adding more work for those involved.
Models based on machine learning can calculate the value of a home’s contents based on the number of inhabitants and rooms.
Predictive analytics can also be used to calculate the value of business contents. Data collected from policies — including the type and size of businesses — can be used to feed models and generate more precise limits for contents coverage.
And this doesn’t require manual calculations for each account. Insurers can often leverage automation to generate this information with just the address and business name or names and ages of the occupants.
Evaluating the quality and accessibility of the data
Not all models, though, are created equal. There may be significant differences in the quality of the models and their ease of use, and the experience of the analytics provider.
Data quality: Models that are verified with actual claims data can be more reliable. By looking at a large collection of claims data over time, insurers can see how close the modeled results come to actual insured losses and feel confident using the model during underwriting.
Model accessibility: Insurers may also want to consider how easy it is to integrate the modeled data into their existing workflows. If the model isn’t part of an open and flexible ecosystem, then it may be too challenging to use at point of sale and renewal.
Experience of data provider: Finally, it’s important to consider the experience and knowledge of the analytics provider. Does the modeler produce any other property insurance analytics, such as replacement cost estimates?
Contents and structural estimates in a single workflow
Estimating the costs to replace contents and repair a structure don’t need to be separate processes. Advanced analytics and automation can be used to generate a complete estimate for contents and building replacement.
With just an address, technology can gather localized information on replacement costs that takes into account itemized labor and materials. Those costs can be refreshed regularly with survey feedback from contractors and real-world claims data.
Insurers can run those contents and building estimates on all or part of their book of business to get a sense of how much premium leakage they may have accumulated. They can also get new estimates on individual accounts, at point of sale and renewal, or annually, to as a starting point to help policyholders confirm whether they have obtained enough coverage.
This article originally appeared in PC360 and is republished with permission.