This Spotlight on Location Data blog series is brought to you by Esri, a leading provider of software that brings location intelligence to business decisions.
One of the most powerful uses of location data is its ability to inform far-reaching strategic decisions. One of the most forward-thinking issues businesses face is how to manage their risk. I interviewed Nigel Davis, executive director of product development at Willis Re, who was a speaker at this year’s Esri Business Summit, to learn more about how the insurance industry is putting location data and geographic information systems to work. The following is an edited transcript of our e-mail conversation.
How is Willis Re using GIS? What are some of the problems you’re trying to solve?
There are a wide range of GIS tools ranging from desktop analysis tools, server applications, mobile and Web browsers, 3D viewers and many more. At Willis, we use the whole spectrum of GIS capabilities because GIS is at the very heart of our work. Many of the problems that we are trying to solve in insurance are spatial ones, e.g., “Which Japanese properties fell within the recent tsunami inundation zone?”
At Wills and the Willis Research Network, our quest is to understand all there is to know about the world around us; where things are, what they are made of, how nature behaves, etc. To further our understanding and awareness of issues like these is as critical to insurance as it is to many other industries and parts of society. Indeed, this is the very reason we have created the Willis Research Network — the world’s largest collaboration between public science and the finance and insurance sectors. It is comprised of over 50 universities and public science organizations, steered by Willis towards furthering our knowledge for both insurance and other industries alike.
How has the use of GIS affected modeling and risk management?
GIS has always been an integral part of modeling risk. Desktop GIS allows us to do things like calculate a flood extent using water depth and a digital elevation model, and we have a team of model building GIS experts who do these types of things every day for a variety of perils as we build our Willis models. In terms of risk management, GIS allows us to produce thematic maps of risk for a given territory.
For example, we might produce a thematic map of counties in Florida, where each county is shaded according to the potential loss for an insurers portfolio. A map reveals spatial correlations and patterns that would be difficult to see in a spreadsheet. You would immediately see that the counties contributing to most of the losses might be those which are closer to the coast — of course this is a spatial relationship. In recent years, the improvements in Web GIS allows us to tell these stories via interactive Web pages rather than paper maps or static images and PDF reports. Web GIS is enabling collaboration and more interactive analysis of insurance portfolios through maps.
Are you using this technology together with social media or mobile tools? How so?
Our client-facing Web mapping site (eNCOMPASS Online) now includes some new functionality to mine geo-tagged data from social media sites such as Flickr, Twitter and You Tube. One of the core purposes of this site is for responding to natural catastrophe events. We take topical storm forecasts and allow an insurer to identify those insurance policies that may be affected by the current active storm. We also show earthquake shake maps immediately after an earthquake and, again, allow the insurer to identify affected polices.
Social media allows us to find pictures, videos and text during or after an event, and this might give an insurer a better understanding of the actual local conditions. If imagery from satellites and planes is called “remote sensing,” we might think of this as “near sensing” or perhaps “community sensing.” If we find social media to be a reliable source of information, we might find that this ground-truthing assists with claims validation, planning loss adjusting activities, contacting key policy holders and many other purposes. We don’t yet fully understand the benefits or the pitfalls but like many other businesses, we can see the potential.
Are there ways smaller insurers can put this technology to work too?
The challenge for a smaller insurer is whether to invest in GIS people, data and software to take more ownership of risk management. Quite often these duties are outsourced to the broker (Willis) given that the investment costs are not insignificant. This is indeed part of the reason that Willis offers certain value-added tools like access eNCOMPASS Online. However, Willis will also endeavor to help clients grow in to their own adoption of GIS and risk management and this has also been done when appropriate.
What advice do you have for insurers looking to build support for using GIS tools within their organization?
As always, adopting GIS software is less about the software and more about the people who use it and the operational business processes that are created. Software and GIS technology becomes the enabler, but it remains paramount that the users of the software are capable of drawing sensible conclusions from their work. Particularly for risk management, they need to be mindful of data quality, assumptions and uncertainties.
It is easy to buy off-the-shelf GIS software, but the overall adoption of course takes a little longer. That said, Esri makes it easy to build and deploy GIS applications within your organization and therefore the business benefits can be very quick once the adoption process has taken shape.