Help

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

Commercial insurers embrace big data technology

Familiarity with predictive modeling encourages trend

Reprints
Commercial insurers embrace big data technology

Sophisticated data storage and processing frameworks, collectively known as big data technology, are gaining ground among major commercial insurers as they build complex models to gauge the risks they underwrite.

Sam Medina, New York-based vice president and head of strategy and transformation for insurance and health care at Tata Consultancy Services Ltd., said big data projects are in the pipeline at the insurers with which he consults.

“The progress has been pretty steady,” Mr. Medina said. “There is not a single commercial lines carrier that we deal with that does not have big data on their agenda.”

He said the insurance industry's longtime use of predictive and catastrophe models give it a head start on big data relative to other industries.

“The big insurers that invested in data all along and have proper data warehouses and already make use of analytics see big data as a way to gather more information to turn into insight,” Mr. Medina said. “So they are not phased by it.”

Christina Colby, New York-based vice president of insurance of Capgemini Financial Services, agreed that while the volume and variety of data sets that big data can handle is novel, the underlying process should be familiar to insurers.

%%BREAK%%

“A lot the buzz about big data has made it more confusing that it needs be,” Ms. Colby said. “It's about using more advanced technologies than were available previously, but you are applying them to concepts than have been there for a long time.''

Chubb Corp. is one insurer building its big data capabilities.

“We have actually made some significant investments,” said Jon Bidwell, New York-based chief innovation officer at Chubb. “We have our first chief data scientist. There is a tremendous amount of potential given the volume of information we have.”

Likewise, David N. Fields, Boston-based executive vice president and chief underwriting officer at Berkshire Hathaway Specialty Insurance, said big data tools have a variety of uses in insurance, with claims and underwriting being the most obvious uses.

“It's something that's immediate and a very useful tool to underwrite business,” but will vary according to the specifics of a particular insurance company and line of business, he said.

“The more homogenous the business, the larger the business, the more deeply ingrained it will be,” Mr. Fields said. “At the other end of the spectrum, you have unique, one-off transactions where big data is not a relevant factor. But when it comes to ordinary commercial insurance, the use of big data can accelerate your understanding and ability to differentiate risks from another to come up with the most appropriate price.”

%%BREAK%%

Big data is particularly well-suited to help insurers price risks, as well as individual underwriting decisions, Ms. Colby said.

“You can use a greater volume and variety of factors for analysis when you get into big data technologies,” she said. “So you can run more what-if scenarios against an entire portfolio and, perhaps, shift the risk appetite of a carrier.”

Nonetheless, Jim Guszcza, Santa Monica, Calif.-based predictive analytics lead at Deloitte Consulting L.L.P., said a crucial consideration for commercial insurers adopting big data is determining the right strategic application of the data and realizing that more data are not necessarily better.

“Particularly for domains like commercial lines insurance where the data can be noisy, I think it's most helpful to shift the focus away from big data to the right data,” Mr. Guszcza said. “Organizations that undergo expensive data gathering operations before clear-use cases have been articulated are often in for a disappointment.”

Mr. Bidwell agreed that companies approaching big data solely as a technology issue without considering broader operational and staffing issues are likely to fail to derive maximum value from the endeavor.

“You can have all your data on (open source distributed data platform) Hadoop, but you won't be successful unless you can organize your firm at a granular level to act on it,” Mr. Bidwell said.