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

Data analytics aid decision-making, cut costs: Bermuda panel

Reprints
Data analytics aid decision-making, cut costs: Bermuda panel

SOUTHAMPTON, Bermuda—Data analytics and predictive modeling can help companies make better risk financing decisions and deal with claims more efficiently and cost effectively, speakers said during the Bermuda Captive Conference.

Speaking Tuesday on a panel discussing data analytics and predictive modeling, David E. Heppen, managing director of Marsh Global Analytics at Marsh USA Inc. in New York, said the use of risk analytics should embrace three themes: a portfolio view of risks, a look at all the potential outcomes and putting a value on volatility.

“One way to think about this is a process,” Mr. Heppen said. “How do you get from start to finish?”

He described a five-step process beginning with devising a risk map, building loss distributions outlining the “potential distribution of outcomes,” quantifying the total cost of risk, designing and evaluating alternative risk financing structures, and viewing those structures in the context of the organization's overall risk.

“Ultimately, make a decision on the overall basis, not just line by line of business,” Mr. Heppen said.

Another panelist, Edward S. Koral, specialist leader at Deloitte Consulting L.L.P. in New York, discussed the applicability of data analytics to claims management and administration.

“Really the question is what kind of data are we collecting now and what data could we be collecting?” Mr. Koral said.

Organizations already collect most of the data that would be useful in such an analysis of claims at first notice of loss. “Data gets more plentiful every day,” Mr. Koral said, while the computer analytical tools available grow increasingly more powerful and less expensive.

%%BREAK%%

Ultimately, the use of predictive analytics can transform the way claims resources are deployed and enable more efficient and effective claims management, enhanced fraud detection and shorter claims cycles, Mr. Koral said.

“The purpose of doing that sort of exercise is now that I know which claims have the potential to be bad (and) which claims have the potential to be not so bad, how do I devote the right resources against them?” he said.

Mr. Koral said insurance companies have more experience in using predictive analytics in underwriting, but are increasingly using the approach in evaluating claims and seeing cost savings as a result. The approach is now being considered by captives and large self-insureds, he said, which could apply such analytics by employing cloud computing or subscription modeling services rather than having to incur the cost of building customized modeling systems.

“We definitely have seen some real-life examples of looking for these optimal structures,” said Mr. Heppen.

Following the panel, Liz Cunningham, senior manager of actuarial at Deloitte & Touche Ltd. in Hamilton, Bermuda, who moderated the session, said some large captives are in a position to employ such predictive analytics.

“I think what people forget is there are many really large captives out there with a lot of their own data,” Ms. Cunningham said.

In addition, captives can supplement their data with other applicable industry data, she said. “Being able to access industry data is a great help and is just starting to become available now through these subscription services,” Ms. Cunningham said.

%%BREAK%%

In Tuesday morning's keynote presentation, Peter Zaffino, chairman and CEO of Marsh Inc., also emphasized the importance of data and analytics in the insurance industry.

“There needs to be an investment in data and analytics,” Mr. Zaffino said. “A company has to commit to it.”

It's important that companies find ways to use available data on a more real-time basis, Mr. Zaffino said, adding that the integrity of data and the quality of analytics will distinguish insurance industry companies in the future.