9 ways to maximize the value of analytics
A decade ago, first movers in analytics reaped market-changing advantages. Today, such disproportionate benefits are harder to find.
Insurers struggle to maximize the return on analytics investments largely due to outdated thinking. Senior leadership must shift their mindset about analytics to achieve the competitive edge they expect from the tools.
The nine “levers” below provide a framework for maximizing analytics ROI and creating the culture required for market leadership.
Senior leadership must let go of legacy decision-making models to explore the full potential of analytics. An analytics scan of a company’s most intractable challenges can prove effective in demonstrating the power of these tools.
Ideally, a Chief Analytics Officer should be added to the senior leadership team. Intense training and education (an analytics “boot camp”) might also be necessary.
Where does the analytics function fit best? Most insurers debate centralized vs. decentralized models, hybrid vs. federated approaches or vertical integration with actuarial or strategy groups.
Ultimately though, connection and communication with the C-suite trumps traditional design considerations and will strengthen the analytics function more quickly than anything else.
After approval by senior leadership, an analytics project enters the decidedly less-glamorous phase of sorting out local execution challenges. Failure to address implementation issues reduces analytics ROI more than any other factor.
Insurers who successfully navigate the last mile of implementation share several traits. They speak the same analytics language (which minimizes subjective debates) and thrive on the daily grind of collaborative experimentation and learning.
In complex environments, seemingly small details often have an outsized effect on performance and dilute the impact of large-scale investments.
Analytics investments are no exception, and day-to-day “firefighting” can lead to strategic drift that’s difficult to detect as it happens. To stay on course, insurance leaders must understand how the many facets of their environments play into overall system dynamics.
Local optimization issues can compromise the insurance value chain. Fortunately, addressing these issues offers the fastest ROI for advanced analytics.
By calibrating applications to local optimization issues and opportunities, insurers create efficiencies through automation. Also, optimization neutralizes legacy assumptions and models (fact-based insights cut through human bias).
In the early days of modeling, insurers focused on mastering the math and pursuing “lift curves”. Today, powerful models are necessary but not sufficient to drive strong analytics ROI.
Insurers have shifted focus to building applications that complement a company’s strategic assets and core value proposition. Without this alignment, an organization works against itself.
Models should also be recalibrated periodically to ensure assumptions remain valid and targets clear. Steeper lift curves are a good thing, but they are no substitute for alignment.
Today’s quants are industry-agnostic, highly mobile and motivated by the prospect of having an immediate impact (and reaping a commensurate reward). Insurers can and should be a destination of choice for top-tier analytics talent.
Insurers need professionals who view complexity as an opportunity, have the confidence to drive change across silos and can engage with the industry’s original quants—actuaries. As always, judgment and leadership are just as important as raw skills.
In the current economy, “sandboxes” and “labs” often get the short end of the budget. That’s unfortunate because a test-and-learn culture offers tremendous benefits.
Experimentation delivers insights about customers, processes and operations. It breaks down organizational silos and reduces project execution risk.
For analytics, experimentation can take many forms. Options include incorporating new data sources in the lab or using field pilots to test the reaction of channels and customers before new product launches.
The bottom line: experimentation should be explicitly budgeted and assigned to talented managers.
The rich insights made possible by analytics are remaking entire industries (retail and pharmaceuticals being notable examples). Yet chronic underinvestment has left the insurance industry overwhelmed by the notion of “big data” and how to manage it.
For insurers, collecting and preparing data remains the major burden of analytics, reinforcing a prevailing view that data is only a cost center. Strong leadership (backed by the new tools to manage and monetize data) is needed to change this perspective.