How to Get Insurance Employees to Think Data-First

Big data is creating an unparalleled shift in how the insurance industry operates, and organizations are rushing to beat their competitors to collect, process and find value in the enormous amounts of information at their disposal. They’re devoting considerable resources to carving out those competitive advantages, too. Recent data from Forrester Research revealed that the big data market is projected to grow three times faster than the IT sector.

Insurance organizations are ahead of most other industries when it comes to finding business value in big data. Seventy-four percent of insurers say using information like big data and analytics gives them an edge over competitors, according to IBM. Among all industries, only 63 percent of companies see big data as a source of competitive advantage. Yet despite the significant investment insurers have committed to data analytics, many organizations fail to leverage what could provide the biggest leg up on the competition: their employees.

Insurers are devoting as much as 95 percent of their big data budgets to developing new analytics tools and modeling capabilities, McKinsey reports. Although this investment is well spent, companies are lacking in their training of frontline employees, who will increasingly need a deep understanding of data analytics as it permeates every aspect of an organization. Without aligning these new tools with the needs and expertise of frontline managers in claims, underwriting and other operations, data scientists are left to operate in a vacuum that limits the efficiency and impact of those tools.

Insurers have made strides in applying big data tactics to traditionally data-based applications such as risk selection and pricing, but that’s really just the tip of the iceberg. Insurers believe the use of big data and predictive modeling will double or triple in the next year in areas like fraud potential, claims triage, litigation potential, premium audits and even marketing and advertising, according to a Willis Towers Watson survey. These applications touch virtually every aspect of the insurance industry, yet most frontline managers in these areas lack the big data skills and mindset needed to keep up.

The demand for big data know-how

This situation creates a perilous disconnect for the industry. As big data spreads to more core insurance operations, data scientists will increasingly rely on leaders in traditional insurance roles to identify areas where data analytics and expensive modeling tools can have the greatest impact on business outcomes. Without an understanding of key concepts and applications of data science, frontline managers and department leaders will not only be unable to provide that insight, they’ll be left out of the conversation entirely.

By appropriating a portion of organizational resources for training frontline employees in data science, insurers can establish a data mindset in all employees. This shift in understanding also begins to infuse the entire organization with a focus on analytics-driven decision making. Just as data scientists must have a basic understanding of traditional insurance concepts, frontline managers should have a grasp of big data fundamentals, including some of the challenges that organizations face in implementing the new analytical techniques.

Where to focus educational efforts

One current challenge facing virtually all organizations grappling with big data is the overwhelming amount of information; enterprises don’t know where to start. Insurers now process only about 15 percent of the information they collect, and most of that data is structured—meaning that it’s easily categorized and analyzed. Unstructured data is difficult to interpret without first organizing and structuring it. It’s raw data unfit for human consumption, as data scientists like to say.

Adjusters’ notes are a good example of unstructured data from the traditional insurance world. To make the most of this raw information, data scientists and claims managers must effectively collaborate. Data scientists can provide the analytical tools, and claims pros should identify opportunities to improve the overall claims process with this newfound source of information.

Frontline managers should also familiarize themselves with data modeling techniques and be able to differentiate between the varying levels of analytical techniques, from descriptive and diagnostic business intelligence to predictive and prescriptive capabilities.

Big data is only going to get bigger—insurers will harness more and more information and apply efficient modeling techniques to more and more business operations. Enterprises that are looking to stay ahead of this curve must dedicate resources to making employees fluent in data analytics in addition to developing the tools that will process the data. Frontline managers and employees must think in terms of data to stay relevant, and insurance companies must support this education in pursuit of a truly data-driven organization.