Two properties can sit only a few kilometres apart, carry similar replacement values and appear almost identical on an insurer’s underwriting system. Yet their actual exposure to loss can be significantly different: One property may sit in a low-lying area exposed to flash flooding, while another is positioned on higher ground but faces different security or fire risks. One building may have recently upgraded plumbing, solar power, battery backup and leak-detection technology, while another has an ageing roof and older infrastructure.
For insurers, the problem is that traditional rating territories can flatten these differences into broad geographic averages. Properties that fall within the same suburb or rating area may consequently be treated as carrying similar risk, even when conditions at individual locations tell a different story.
According to Laurette Coetzee, Senior Account Manager at Esri South Africa, this is creating a strong case for insurers to make greater use of hyperlocal data when assessing property risk.
“When insurers begin looking at what is happening around a specific property, as well as the characteristics of the property itself, they can build a much more complete picture of the risk they are taking onto their books,” she says.
The problem with averages
Insurance underwriting has historically relied on data to understand and price risk. The difference today is the level of detail that insurers are able to bring into those decisions.
Rather than evaluating a property primarily according to a broad geographic area, hyperlocal underwriting allows insurers to analyse risk at or around the precise location of the insured asset. This can bring together geographic, environmental, infrastructure and property information to identify risk factors that may not be apparent from an address or suburb alone.
For commercial property insurers, the differences can be equally important. Two buildings located only a few streets apart may experience very different levels of exposure depending on the condition of surrounding municipal infrastructure, drainage, utilities or other local factors.
This matters because inaccurate risk segmentation can create problems at both ends of the portfolio. Lower-risk customers may effectively subsidise higher-risk customers, while insurers can unknowingly take on exposures that are not adequately reflected in premiums.
“Insurance has always been about distinguishing one risk from another,” says Coetzee. “The opportunity now is to do that with much greater precision. Instead of assuming that properties in the same area face broadly the same conditions, insurers can start identifying the factors that make each location different.”
Portfolio risk is also a location problem
The value of location intelligence extends beyond individual policy pricing. One of the more significant applications for insurers is identifying concentrations of risk across an entire portfolio.
An insurer may, for example, hold a number of policies in an area susceptible to flooding. Each individual property may fall within acceptable underwriting parameters. Viewed collectively, however, those policies could represent a substantial accumulation of exposure should a severe weather event affect the area.
Granular location analysis gives insurers greater visibility of these concentrations and can help them understand where a single event could generate claims across multiple insured assets simultaneously. This can support portfolio rebalancing, risk selection and capital management before an accumulation becomes a costly problem.
For an industry built around understanding the probability and financial impact of future events, being able to see where risks are clustered is increasingly important.
From underwriting to claims
More detailed location intelligence also has applications beyond the point of sale. When an insurer has an accurate view of where an insured asset is situated and the conditions surrounding it, the same information can support claims assessment and help validate whether a loss is consistent with an event affecting that location.
Coetzee says granular geographic information can also assist insurers in identifying anomalies that could warrant closer examination, including potential fraud, while more accurate asset information can contribute to quicker processing of legitimate claims.
Artificial intelligence is making it increasingly practical to analyse large volumes of this information and identify relationships between datasets that would previously have been difficult to assess at scale.
But Coetzee cautions that the value is not simply in having more data. “More information does not automatically lead to better underwriting. The real value comes from being able to connect different datasets to a location and turn them into information that an underwriter, risk manager or claims team can actually use to make a decision.”
A more precise view of risk
For insurers, greater underwriting precision ultimately creates an opportunity to align price more closely with actual exposure. It can help distinguish between customers who may previously have been grouped together, identify areas of portfolio concentration and give underwriting teams a more detailed understanding of the assets they insure. It can also change how insurers think about geography. A suburb, municipality or postal area may still provide useful context, but it should not necessarily define the risk. Increasingly, insurers have the data and technology to see those differences. The competitive advantage will lie in what they do with that information.