Predictive Analytics: Added Insurance for Insurers in a Volatile Economy

By Jessica Dolezal, Sr. Data Scientist, Prevedere

Despite the promise of big data to provide insurance companies with revolutionary insights that will drive business forward, the benefits of such data have proven elusive. In fact, less than 20% of insurers had leveraged or were planning to leverage big data as of 2015. But there is good news—new technologies that employ predictive analytics and cloud computing to make sense of big data are putting greater insights into the hands of insurers, allowing them to make smarter decisions and improve profits.

Insurers now have the power to analyze the entire world’s economic data and discover hidden leading economic indicators that impact future performance of the industry—all within minutes. Doing so often yields surprises about the true driving factors of performance, most notably how predictions for the industry and individual product sales can be projected just as they are for consumer-packaged goods. In looking at the insurance industry specifically, the following three economic indicators play key roles in the performance of insurance companies:

  • Consumer sentiment: Released monthly, consumer sentiment measures how buyers view their current financial situation, the economy in the short- and long-term growth prospects. While this metric will affect different product lines differently, buyer views on their own economic health and that of the economy can tell insurers a lot about future demand. For example, demand for annuities may spike when consumer sentiment is in decline as a measure of protection, while demand for auto insurance decreases as people wait to buy new cars and trim their policies to minimize costs.
  • Housing permits: Certainly a driver of future demand for homeowners’ insurance, economic data regarding housing permits are also a measurement indicating how people are feeling about the economy overall. As housing permits increase, insurers can expect improvements across multiple product lines, as homeowner’s insurance purchases are often linked with personal article policies, as well as life insurance. Housing permit data is available by geography, allowing insurers to anticipate demand on a granular, local level.
  • Population health: Understanding health trends in a population, segmented by age group or geography, can give signals to whether, when, and which products will be purchased. For example, if the aging population’s health largely varies between markets or certain regions, this affects the propensity to buy safeguard products like life insurance and risk assurance policies differently within each market.

In addition to these three driving factors, other indicators are key to the performance of the insurance industry. Insurance products behave similarly to that of retail environments. While this correlation may be obvious with large purchases—when auto sales increase, so does the need for auto insurance—the relationship actually holds for retail sales at all levels. Income level is also a driving influence, with higher incomes equating to higher insurance sales.

What Can Insurance Companies Expect in 2017

Based on all of these factors, current indicators are pointing to a weak first quarter of 2017, with demand for insurance products accelerating beginning in the second quarter. However, the outlook and driving factors can be different for each business, product, and locale. Using predictive analytics and cloud computing technology, insurers can hone in on the micro-economic factors affecting their individual business, product lines and geographic locations. Armed with this information, they can more accurately forecast demand by product, leading to smarter decisions—particularly in the marketing department. For example, advertising dollars can be allocated by region, product and target audience, and leveraged at the time when consumers are most likely to buy—increasing effectiveness and improving return on investment. When Nationwide incorporated predictive analytics into its marketing planning, the company increased demand by 15% per year while reducing the budget and increasing the productivity of its marketing team.

“Before we made the investment in marketing analytics, we were already spending hundreds of millions of dollars on promotional activities. We had to decide, how much goes to television? How much goes to digital? How much goes to sports marketing or sponsorships? We had to decide how many ad spots to run per year and whether to have a spokesperson or not,” former Nationwide CMO Matthew Jauchius told McKinsey & Company. “Many times, these decisions were more art than science, based on instinct and experience. Marketing analytics allows us to make every single decision I just mentioned better with data.”

While insurance executives may have initially been turned off by the challenges presented by big data, it’s time to take another look at the opportunities available. This information is now more easily accessible, more easily analyzed and more easily matched to business performance, helping insurers make smarter decisions and improve profits.