BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe

Automated underwriting brings insight and efficiency to corporate insurers

Increased data sources, improved tech tools aid insight into risks

Automated underwriting brings insight and efficiency to corporate insurers

To get a sense of the future of insurance underwriting, a look at its current state is instructive.

Today, underwriters gather data to evaluate exposures. What is changing rapidly is the amount and variety of data available for analysis and the tools used to aggregate and analyze the data.

Cliff Hope, executive vice president, chief property underwriter for U.S. property for Aspen Insurance Holdings Ltd., said the technology available to underwriters has come “light years during the last 15 years.”

The improved technology enables them to consider a wider array of variables and ultimately achieve a more granular classification of risks, said Karlyn Carnahan, principal at New York-based insurance advisory firm Novarica, a unit of Novantas L.L.C. Risks once assigned to broad underwriting tiers such as preferred, standard and substandard now are grouped in finer categories more specific to individual risks.

“The shift to multivariate rating has really allowed micro-rating,” she said. “It's really a rating for one.”

This more exacting segregation of risk also has made the underwriting process more efficient, enabling fewer underwriters to do more work. In certain lines of business, such as personal lines automobile coverage, underwriting is largely an automated process where the vast majority of submissions are handled without human interaction and with few exceptions that fall outside of business rules tagged for additional review.

While fully automated underwriting is more established in volume-heavy, homogenous lines of business such as personal lines auto, it is beginning to make inroads on the commercial side, Ms. Carnahan said, adding that the increasing availability of modern underwriting systems paired with models have catalyzed this transformation.

“What we are now starting to see is fully automated underwriting for more complex lines such as workers compensation and business owners policies,” she said. “Even on very complex accounts, we are seeing analytics and scoring providing valuable guidance to the underwriter.”


The burgeoning use of predictive models in underwriting has coincided with a broadening array of external data streams to feed the models, said John Belizaire, founder and CEO of FirstBest Systems Inc., a Bedford, Mass.-based provider of insurance underwriting workstations and portals.

Whereas underwriters once largely depended on application and historical data to assess a given risk, modelers now can blend internal data with massive amounts of third-party data culled from sources ranging from public records to embedded sensors to satellites (see related story).

“The type of data used in the underwriting process is changing over time,” Mr. Belizaire said.

Yet the move to fuse data-intensive technology into the underwriting process will take longer, said Jamie Bisker, research relationship manager of insurance at IBM Research in Columbus, Ohio. This is especially so of big data, the data sets so large that they strain traditional database and network architectures. Mr. Bisker said big data presents great potential for insurers, but they first need to optimize the use of data within their existing underwriting mechanisms. “It's not about having a lot more data,” he said. “It's about having the right data.”

Looking ahead, competitive pressures may force insurers to adopt big data sooner rather than later, said Scott Busse, Chicago-based director of U.S. insurance advisory services for PricewaterhouseCoopers L.L.P.

“The tools and techniques for the big data are already out there,” he said. “So the ability to look at large data sets and draw insights from them will soon be a differentiator for insurers.”


Ultimately, the technology wielded by underwriters will need to leverage and reflect the broader technology ecosystem in which the insurance industry operates.

Separately and in confluence, emerging technologies such as cloud computing, mobile technologies and big data are fundamentally altering many industries and, in turn, the risks they pose to the companies that insure them.

“This is the first time in the history of the technology business that you have multiple elements of deep transformational change happening at the same time,” Mr. Belizaire said.

Read Next