Analysts have declared a "big data" revolution. Large and smallbusinesses across multiple industries are unlocking the potentialof ever-increasing volumes of data with powerful analytics tools.When used properly, big data helps businesses to grow market share,increase profits and reduce risk and uncertainty.   

The insurance industry—often regarded as a laggard intechnology-driven innovation—is moving quickly to respond to theopportunities and challenges of big data.

Large carriers are driving adoption and have achievedsignificant success in personal lines.

These carriers are using powerful predictive models to mine hugepolicy datasets in real time to assess risk and optimize pricing atan individual submission level. They are also using advancedalgorithms to find patterns among millions of claims to identifyfraud with great accuracy. They also are leveraging telematicsdevices to collect detailed data on motorists' driving practices,which enable them to offer usage-based insurance through whichcustomers can "pay as they drive." In these cases, big data istransformative, enabling insurers to conduct business indramatically new ways. 

Big Data for Commercial Lines

Does such potential exist outside the world of personallines?  What does big data mean for the corporate riskmanager, the global broker, and the carriers that serve them?

To answer these questions, we have to look at how the dynamicsof the commercial lines market challenge the analytic techniquestypically employed by big data applications. Carriers have done farless here than those in the personal lines space.  Samplesize is a significant issue, as even the largest carriers servingthe commercial lines marketplace write a relatively small number ofpolicies. Claim volumes also are typically low, which limit theeffectiveness of the statistical models used to predict either thefrequency or severity of loss and to price riskaccordingly. 

Variability within the data also presents challenges. Large commercial entities defy easy categorization; each has uniqueaspects to its organization, business activities and operationsthat make it, in many respects, a "sample of one." Similar variability applies to the insurance products that largecompanies buy. Complex program structures, tailored coverages, andmanuscripted endorsements make it difficult to structure policydata sets to enable true "apples-to-apples" comparisons. Theseissues have limited the take-up of big data initiatives amongcarriers focused on large commercial accounts.

An exception to this general rule has been in the area of publiccompany directors & officers (D&O) insurance.  Thewidespread availability of class action lawsuit informationdocumenting incurred losses and its correlation with publicallyavailable financial data has encouraged several leading carriers todevelop sophisticated predictive models. These models score andprice individual risks based on estimates of the likely frequencyand probable severity of a loss based on hundreds of data elementsrelating to the historical financial performance of a company andits stock.  

Global Brokers Are Taking Advantage of BigData 

The leading global brokers have taken greater advantage of thebig data opportunity.  As intermediaries, Marsh, Aon andWillis have access to more policy data than any singlecarrier.  In recent years, they have begun to warehousetheir clients' submission and program data in large consolidateddatabases. These allow new analytic offerings built on thefoundation of these big data assets. Enhanced price and limitedbenchmarking, more finely tuned retention analysis, and competitiveinsights into the carrier marketplace promise better programstructures and optimized placements. Risk managers can and shouldbenefit from this scenario-based modeling of the total cost of risk(TCOR) provided to them by their brokers.  

Claims data is on the frontier of big data initiatives. Brokershave begun to aggregate claims information across theirclients.  This work will take time, but the potential ofsuch datasets is enormous. Statistically meaningful samples ofhigh-severity claims offer the possibility of building predictivemodels for lower-frequency casualty lines. These models promisebetter estimates of both the probability and probable maximum loss(PML) for a variety of large-loss scenarios not captured throughtraditional catastrophe modeling techniques. With such models, riskmanagers can take more strategic approaches to risk transfer andenterprise risk management (ERM).

The Potential for Risk Managers

Risk managers do not need to depend on brokers to enjoy thebenefits of big data.   Here are three areas where riskmanagers can take the initiative and deliver value to theirenterprises today:

• Telematics: The inexpensive devices whichpower usage-based insurance in the personal lines marketplace arealso widely used in commercial fleet-management applications. Thedata produced by these devices, properly mined and analyzed, offersgreat potential to drive risk management initiatives directed atimproving the safety of driving practices throughout theenterprise. Telematics provides the foundation for proactive motorvehicle risk management and can contribute to substantialreductions in TCOR.
• Workers' compensation: Many largeorganizations have a significant number of attritional workers'comp claims that fall within their retention.  Data aboutthese claims, whether managed through an internal risk managementinformation system (RMIS), offshore captive or TPA, is an asset.Predictive models based on this data can identify factors drivinglost activity and suggest risk mitigation approaches to lower claimfrequency and reduce cost.
• Supply chain risk: The recent Fukushimaearthquake highlighted the vulnerability of many enterprises tonatural disasters affecting their suppliers. Companies can leveragethe data stored in supply chain management systems to quantify andmitigate such risks.  This data can be mined to generategeospatial maps of principal suppliers. Property catastrophemodeling software can leverage these maps to calculate the PML fora series of potential supply chain disruptions. Risk managers canrespond proactively, chartering supply chain diversificationinitiatives or purchasing contingent business interruptioncoverage.

The complexity of commercial lines business makes the adoptionof big data approaches more of a marathon than a sprint. Thedeliberate pace of adoption should not be confused with a lack ofmomentum. Although it may not be a revolution, big data offerslarge commercial enterprises significant opportunities to improveboth risk management practices and the effectiveness and efficiencyof their insurance purchases.