(Image credit: Adobe Stock.)

Data is an asset. That is the mindset companies in the insurance industry must quickly embrace if they are to remain competitive in a landscape that is being challenged from seemingly all angles. From product commoditization, shrinking margins and disruptive startups to aging technology and processes, changing customer expectations and regulatory uncertainty, insurance companies face a critical imperative to harness data assets existing within their enterprise today.

But hurdles remain, including significant concerns about data quality and accuracy, analysis paralysis and segmentation issues. There’s also no clear answer when it comes to ownership of advanced analytics within organizations and a lack of clarity on where it fits into insurance companies’ budgets.

Where Do You Go from Here?

We’ve seen remarkable strides in recent years when it comes to the insurance industry utilizing the mountains of available data. But there’s a mindset that still persists: Big data can help control costs, but using it for revenue generation or elsewhere is confusing. Where should organizations struggling to take advantage of advanced analytics begin?

Here is our advice.

Step 1: Start with a Specific Business Problem First. Focus on a specific business problem—with the goal of demonstrating advanced analytics’ value in ways that can be broadcast across the company. One example would be to use consumer information (perhaps based on engagements to be married, home sales, etc.) to identify suitable products to recommend to those individuals at their time of need or life event. The information, in the hands of producers, can increase sales. Measuring the impact can help demonstrate the value of advanced analytics to executives and entire organizations, paving the way for broader efforts and investments.

The key here is showing return on investment—which insurance executives identify as a major challenge. In addition to recommending the next product to purchase, or rolling out new products based on actionable and predictive modeling, organizations can use advanced analytics to assess top-performing products and find ways their success can be used as a model for others.

Step 2: Make Sure Data Is Clean and Accurate. It’s obviously difficult, if not impossible, for organizations to derive much value from bad data—and there’s the matter of convincing individuals throughout organizations that the data is accurate and will stay accurate.

Companies should measure data quality issues through an integrated exception reporting process—which documents abnormalities that demand attention—ideally managed by a data governance committee. Without knowing where the problems lie, steps cannot be taken to address and permanently correct them.

Put another way: Without confidence in the quality of your data, planning for next steps is extremely difficult. The business owners of the datasets—e.g., client data, financial information, etc.—must be responsible for monitoring data quality. The problems must be identified and addressed before they go downstream for analysis.

Step 3: Attack Data Segmentation Problems. These issues typically stem from difficulty in unifying data from an acquired company or dealing with disparate legacy systems within an organization.

As far as acquisitions go, it is imperative that organizations have a formalized process to integrate data from an acquired company on the first day post-acquisition. When it comes to integrating information from disparate legacy systems, options include using lean analytics or data warehouses to aggregate data for analysis.

Extracting data from large mainframe systems is a problem that bedevils older industries, including insurance. It should be noted that many of the longtime employees who understand the data through their years working with it are retiring. The next five to seven years will be a critical window for organizations to use that institutional knowledge to modernize core systems and combine it with new technologies and experts.

Step 4: Focus on the Possible and Results. The biggest mistake companies make is trying to move forward without any sort of analytics strategy. To be effective, the analytics strategy must be aligned to organizational and market objectives, and include components that are measurable against business processes.

When creating an analytics strategy, gather input from business leaders within the organization to find opportunities to inject data into a business process or customer journey—staying focused on the goal of providing an edge for the business. Going back to the first step, this is about identifying an actionable hypothesis based on a  problem that data can help solve, then solving it and, finally, demonstrating the results. The proof, in other words, needs to be in the pudding.

Obviously, there’s a lot of talk about what advanced analytics can do for the insurance industry. Too often, organizations get excited about what the future holds and immediately turn to their technology experts. While that kind of input is important—and technology is a key partner in helping implement solutions—creating value from data via advanced analytics must be driven by business strategy leaders and the C-suite.

The last five years have seen changes in technology, such as the rise of cloud computing, that have made solution creation and the ability to drive value happen much more quickly. Further developments in areas like artificial intelligence and cognitive computing are becoming more accessible to improve everyday decision-making. Harnessing data assets and advanced analytics at this stage is not only important in and of itself. It’s the first step to prepare for these coming megatrends.