Oakland Athletics' general manager Billy Beane rocked Major League Baseball by applying a sabermetrics-based analytical technique that identified success factors that other teams failed to appreciate. It wasn't that competitors didn't use analytical techniques or lacked the relevant data; it was that they started from inadequate assumptions and asked the wrong questions. Insurers have a similar opportunity today to leap ahead of competitors by applying prescriptive analytics, a technique that optimizes pricing through sophisticated analysis of customer demand. But like Beane, champions of prescriptive analytics must overcome internal cultural resistance.
In the 2011 film "Moneyball," Beane has a fateful encounter with Peter Brand, a Yale economics graduate student. The Brand character insists that scouts and coaches are applying the wrong criteria of success, focusing on prospects with signs of potential greatness. His insight was that cumulative team criteria such as the aggregated propensity for getting on base - known as the on-base percentage (OBP) - were more important than traditional gauges of excellence in individual players. The conventional thinking of team managers was to think in terms of buying players, Brand notes. "Your goal shouldn't be to buy players, your goal should be to buy wins," he counsels Beane. "And in order to buy wins, you need to buy runs."
When attempting to act on Brand's advice, Beane runs into determined resistance from A's scouts and leadership. Champions of prescriptive analytics at insurance companies may initially face similar resistance from senior management, and for similar reasons: like successful baseball teams, senior insurance executives can point to a record of success based on a combination of intuitive judgment, rigorous analysis and raw experience. As in the case of overcoming the conventional wisdom of baseball scouts, the embrace of prescriptive analytics requires accepting a counter-intuitive proposition - that what has been proven to work over the years will not sustain future success.
What senior management must be shown is that their assumptions have not been wrong, but that prescriptive analysis offers a superior alternative that will set the pace for the industry. Historically, loss costs have been the primary basis for insurance pricing. Prescriptive analytics complements loss cost analysis through the addition of systematic quantification and modeling of pricing according to customer reactions. The process then involves the exploration of all possible scenarios based upon growth or profit objectives, resulting in decision recommendations or "analytical prescription."
Typically, insurers apply experience-driven "rules of thumb" about customer demand and price elasticity to determine effective pricing. For example, a carrier might apply the assumption that a 10-percent increase in premium will result in a 2-percent relative drop in retention. While the enforcers of such rules can provide statistical support for them, the rules are nevertheless insufficiently granular. They fail to acknowledge change in consumer price elasticity over time and, more significantly, they assume all customers react in the same way. In reality, customers in different segments react differently. Internet shoppers are more price-sensitive than customers who buy through an agent, multi-policy customers are more likely than single-policy customers to renew after an increase, and prospects are more sensitive to price than existing customers.
The first question that arises when considering a more sophisticated analytical technique is "What data will be needed?" The answer in most cases is that the insurer, even a small insurer, already possesses the right data in its internal systems. The data required is about which of the insurer's customers and prospects buy or renew at a given price point, and which don't. Important variables include absolute premium, change in premium, multi-policyholder status, driver age, credit score and distribution channel.
While insurers already have the data they need to perform prescriptive analysis, they need to acquire the tools, the methodology and, more fundamentally, the buy-in of their executive leadership. However, that should be easier when considering that prescriptive analytics can improve an insurer's combined ratio by one to three points, constituting a tremendous ROI for a relatively small capital expense.
The sabermetrics-based approach applied by Beane went on to become standard for Major League Baseball. Prescriptive analytics for customer demand modeling and price optimization currently provides a competitive edge for early adopters. Eventually, however, it will be table stakes. As Red Sox owner John Henry said to Billy Beane toward the end of Moneyball, "Anybody who's not building a team right and rebuilding it using your model, they're dinosaurs."
Aviv Cohen is VP Marketing at Earnix, a provider of pricing and customer analytics software for the financial services industry.
Register or login for access to this item and much more
All Digital Insurance content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access