With the rise of big data and the Internet of Things, intelligent analytics is a necessity to compete in today’s insurance market. Most carriers are already using analytics in one fashion or another. An interesting blog from Property Casualty 360° titled, Predictive Analytics: Who, What, Where, When, Why and How, cited some industry research that revealed 82% of P&C carriers are using analytics across the enterprise, but many face key challenges such as inefficient data sets for creating useful, predictive models.
Of all industries, the insurance sector is a data haven. The issue is not a lack of data. The issue is 1) the data is constantly changing, and 2) data paralysis halts analytic initiatives due to the massive data sets. In addition, most carriers are only applying one side of the data analytics equation – “predictive models”. There is another component that is crucial: adaptive analytics. Together, predictive and adaptive analytics function like a state-of-the-art GPS system.
“The issue is not a lack of data. The issue is the data is constantly changing”
To expand on this analogy, old school automobile GPS systems (circa 2004) were more like predictive models. They suggested the quickest route from Point A to Point B based on pre-loaded/historical map data. Today’s GPS systems apply predictive and adaptive intelligence. When I start my guidance system in my BMW to get from Boston to Marblehead it initially gives me the quickest route based on historical map information while I am sitting in my driveway. More importantly, it provides adaptive analytics based on real-time data throughout the journey such as road construction and bumper-to-bumper traffic. Based on this just-in-time data the guidance system dynamically adapts and re-routs me to avoid a traffic jam. What once was not the most direct route now dynamically became the best choice, due to conditions at the time. In this analogy you see how a GPS leverages historical data (predictive) with real-time data (adaptive) to guide me to make better decisions. Earlier generation GPS systems would provide sub-par results due to insufficient data, much like most carriers are using predictive analytics without adaptive capabilities.
One area where insurers can leverage both predictive and adaptive analytics is servicing policyholders at the point of interaction. The combination of this technology helps assure every service representative provides a high-value, intelligent conversation to improve the overall customer experience. Intelligent conversation management treats each interaction as part of an interconnected conversation, leveraging understanding of the customer, past interactions and real-time information to predict future interactions and manage each conversation within the overall context of the relationship. With intelligent conversation management, you can successfully navigate through each moment of truth to deliver experiences that enhance the value of your company to your customer – and the value of your customer to your company. Many interactions once considered simply service calls, such as a change to a policy, can now be leveraged to increase retention and wallet share. For example:
- Analytics can predict which additional products or policy upgrades a specific policyholder is most likely to buy and at what limits based on similar buying characteristics and interactions. This prediction can be combined with real-time information to adapt a typical service call into a highly relevant cross-sell or up-sell offer based on the customer’s buying propensity or information gathered as the conversation unfolds.
- Real-time awareness that the customer is unhappy during a conversation can immediately replace a planned up-sell offer with a retention incentive or, if analytics determine this policyholder is not worth retaining, no offer at all.
- Advanced analytics can also assess the risk for an individual expected loss at the portfolio level. That includes everything from the likelihood of a claim and the expected claim value, which can help present quotes to reduce underwriting costs when acquiring a new customer, or attempting to retain an existing one.
I’d like to hear your perspective on this. While predictive analytics is being used across the insurance service chain, how often are adaptive models being used in conjunction with predictive models to drive accuracy and usefulness to changing and incoming data? Predictive and adaptive analytics are an absolute necessity to function in today’s data-centric insurance sector. Analytics are the GPS system used to guide carriers on intelligent decision making. If your enterprises are anything like the construction-clogged cow paths in Cambridge, Massachusetts, you’ll need as much turn-by-turn guidance as you can get.