The importance of analytics in the insurance industry—an industry driven by data—cannot be disputed, but with the size and number of insurance carriers in the U.S. market, where they are in the process ranges widely.

Unfortunately, many carriers have yet to take the first step as they grapple with other issues such as legacy systems and incomprehensible data. So we asked several insurance technology analysts the following question: What is one step insurance carriers can take to improve their ability to use analytics within the enterprise?

Click through the slide show to see the responses from: Matt Josefowicz, of Novarica; John Lucker, of Deloitte Consulting; Mark Breading, of Strategy Meets Action; Karen Pauli, of CEB TowerGroup; Ellen Carney, of Forrester; and Chuck Johnston, of Celent.

Matt Josefowicz, managing partner, Novarica

Unfortunately, most insurers that aren't already using analytics effectively have more than one step to take to get there. Investments in data quality, building and testing analytical models, and incorporation of analytics results into business processes are all required in order for insurers to improve their results through better use of analytics.

One step that insurers can take, though, is for business leaders to insist on data and reject business cases based on lore or gut feel. Only when business leaders demand to know in detail why underwriting results are trending a particular way or customer retention is trending another way, why service costs are going up or claims productivity is going down, can analytics start to deliver value within the company.

Then the right technology, process, and training improvements can be made. But until senior business leaders establish a culture where analytics are the critical foundation for taking action, IT investments in analytics will not yield positive results.

John Lucker, principal, Deloitte Consulting

Naming one step as most important to use analytics is challenging because it's important to think about six steps to end-to-end analytics execution—analytic strategy, analytics development, operational implementation, technology integration, change management, and performance management.

However the first step in the process—strategy—if not done well, can doom all downstream activities. Too often insurers fail to develop their analytic strategy choosing instead to focus on tactical tasks; projects that they want to do or feel they need to do. At times, models and solutions are built and then companies ask "now what?"

Analytics needs to be part of a holistic strategy of moving toward a more metrics-oriented, fact-based business process culture. An articulate strategy should have many components, for example" (a) areas of opportunity; (b) expected benefit to be realized and KPI's impacted by strategic streams; (c) how benefit will emanate from analytic driven business process and technology integration; (d) how strategy will be socialized and managed organizationally; (e) how will results be recognized and tuned, etc.

Without a full line of sight to a strategic vision it is hard to imagine how significant and lasting analytic driven business value can be realized.

Mark Breading, partner, Strategy Meets Action

There's no question about it. Insurers understand the power of analytics to improve every part of their business. SMA research indicates that the vast majority of insurers are increasing their spending on analytics projects; many of them are making significant increases.

Successfully creating and leveraging analytics capabilities takes a concerted effort. It involves business strategy, organizational changes, training and skills, enterprise data management, and new technology platforms and tools. All of these are important and require planning, resources, and coordinated implementation. But one area rises above the rest in importance—enterprise data management.

Data management may be less exciting—and more elusive—than strategy, organizational changes, technology, or training. But an organization's ability to capitalize on analytics ultimately boils down to a few simple but critical questions, all of which point to data management: How complete and accurate is the data? Is the data consistent across the enterprise? Are users able to easily access the required data? Is the data organized to make consolidation and aggregation straightforward?

Insurers that place a high value on data management will be able to answer these questions positively, positioning them to gain maximum leverage from analytics.

Karen Pauli, research director, insurance, CEB Tower Group

The most immediate response to this question is "get your data in order." However, from CEB TowerGroup's experience with customers adopting analytics, the response is "assure that your organizational structure is set up for an analytics driven business environment." Organizations that view analytics as the sole domain of actuaries or, following a trendier new job classification—data scientists—will not get the value they are hoping for.

It is critical for insurers to reorganize around blended teams for full analytics value. Representatives from field operations (e.g. claims, underwriters, loss control, sales, etc.), business analysts, statisticians, analytics experts, modeling experts, and business strategy should all be part of blended analytics teams. Certainly, the exact structure will depend on the product mix of a carrier. However, the point is that full value of analytics comes from multi-discipline collaboration and decisioning.

Actuaries and analytics will always be a cornerstone of insurance companies. However, the actuarial goal of creating perfect pricing through analytics is in conflict with driving innovative business outcomes through analytics which inherently has elements of creativity, imagination, and scientifically created risk. Blended analytics teams can deliver on a goal of innovation for competitive advantage and creating these teams is critical to get the most out of analytics investment.

Ellen Carney, principal analyst, insurance e-business and channel strategy, Forrester

Anything to do with data and analytics is getting big attention and even bigger investment increases. That means that insurers have to develop analytics strategies that demonstrate the data science team's (or whatever term your firm uses—Big Data, analytics, or metrics) impact on the insurer's financial performance, both directly through analytics that measure performance against the carrier's goals as well as the analytics that capture an indirect association to business performance like brand awareness or perception.

That means that insurers have to base their analytics decisions based on what the business objectives are. Analytics that measure business objectives, such as improving lead conversion, better risk profiling or even the basic stuff like using analytics to reduce service and sales costs help quantify the impact of data and analytics strategies and help to justify further investments.

Chuck Johnston. Director, Americas life/annuity and group practice, Celent

Insurers need to get past perfect and focus on good enough. Many insurers are still very focused on data quality to the point where they are missing out on near term opportunity.

Insurers have been doing data management and cleansing projects for decades and the state of industry data has gotten better but is far from cleansed. While there are immediate customer services issues in using ambiguous data in transaction processing systems, analytical systems can deal with ambiguity in information, especially if there is sufficient volume of data.

Remember, the more data you are processing the greater potential for cross reference and use of statistical tools to actually increase overall accuracy. The single most important step insurers can take to improve their ability to use analytics within the enterprise is to get started.

Special Reports /

Insurance Analytics: The First Step for Success

Related Stories

Recommended For You