Predictive Analytics in Life Insurance
Predictive analytics in Life insurance isn’t exactly new. In fact, actuaries have been using forms of predictive analytics for hundreds of years. And for hundreds of years, actuaries have used estimates of life expectancy in the form of mortality tables that reflect aggregate insured population mortality. This, coupled with underwriting techniques, is used to analyze individual risk as well. This process, while still widely used, is time-consuming and expensive. Individual policies can take months and hundreds of dollars to underwrite, resulting in higher premiums.
As carriers, both legacy and InsurTech, undergo the digital transformations they’ve made strides towards streamlining the underwriting and sales process. Improvements such as shortened, user-friendly applications and lower monthly premiums have led to more digital adoption.
But Life Insurance has a ways to go in its adoption of predictive analytics to keep pace with its insurance counterparts like Property & Casualty. Hope is not lost, however, as companies such as Breathe Life, Ladder Life, and Bestow have recently entered the market with an innovative accelerated underwriting, digital-first approach.
The challenge with any predictive analytics modeling is access to good data. Take an auto claim versus a life claim, for instance. 10% of drivers make a claim per year while life insurers can expect about one death in the first year per 1,000 policies issued. Given Life policies are typically very long (10-30+ years), using claims as the outcome to train your models can be challenging.
As a result, carriers are beginning to work with a growing number of next-generation datasets to assess risk and enhance predictive models more accurately. For instance, using behavioral intelligence to analyze how users interact with digital life insurance applications has proven to be effective at identifying outcomes such as medical or tobacco usage non-disclosure. Based on a user’s digital body language, carriers can determine if a user should be moved to accelerated underwriting or if the carrier needs to triage the application and add additional requirements such as a Telemed interview or fluid test before making a final offer.
How Predictive Analytics Benefit Life Insurance Companies
Willis Tower Watson ran a Life Predictive Analytics Survey in 2018 that broke down how carriers view predictive analytics. The respondents rated the following four factors as highly important:
- Competitive pressures in product development and pricing (78%)
- Customer relationship management (67%)
- Earnings and profitability pressures (64%)
- Technology innovation (60%)
The survey identified three areas in which predictive analytics has had the greatest impact on life insurers’ performance:
- Reduction in issue and underwriting expenses. 67% of companies report a reduction in expenses
- Significant increase in sales. 60% report an increase in sales
- Increase in profitability. 60% report an increase in profitability
The projections for the next few years far exceed these percentages. And, according to Willis Tower Watson, the #1 most important thing insurers are doing today is choosing and accessing the most valuable data. “We make no apology in repeating that the biggest, quickest wins will typically come from sourcing new (or better) experiences and/or customer data.”
The fastest-growing segment of new data adoption is clickstream data, which is up from 18% in 2018 to 45% in 2020.
Projected Growth of Predictive Analytics User Cases by Life Insurance Underwriters
The expanded use of predictive analytics by life insurers is expected to grow from 2018 to 2020 in four specific areas:
- Pricing and rate-setting use are projected to increase from 31% to 56% in two years for group life, and from 18% to 55% for individual life.
- Underwriting use is projected to increase from 52% to 92% in two years for individual life.
- Mortality and morbidity risk use is projected to increase from 19% to 56% in two years for group life, and from 23% to 75% for individual life.
- Claim management use is projected to increase from 37% to 87% in two years for group life, and from 10% to 40% for individual life.
Now that we’re well into 2021, our guess is those percentages are even higher today. Needless to say, the applications and adoption of predictive analytics in life insurance are growing.
Current and Future Investment Trends
Given the commoditized nature of insurance, competitive pressure and customer expectations are the key drivers of increased adoption. Below is the breakdown of estimated investment decisions for carriers.
Predictive Analytics + Behavioral Economics = Behavioral Intelligence
We write quite a bit about the benefits behavioral intelligence is having in the insurance industry. Combining predictive analytics and behavioral economics with a dash of machine learning and a pinch of proprietary behavioral data, ForMotiv has proven an ability to measure user behavior to predict intent.
And as ForMotiv is utilized more and more as a behavior-as-a-service solution, carriers have begun integrating behavioral data across the enterprise. Here are some example use cases that have resulted in 7 and 8 figure ROI for our customers.
- Marketing / Growth
- Quote to Bind Modeling
- Channel Preference Modeling
- Remarketing and Retargeting
- Premium Leakage
- Rate Pursuit
- Distribution Analytics
- Agent Behavior Reporting
- Loss-Ratio and Efficiency Modeling
- High-Risk Deterministic Signaling
- First Notice of Loss Modeling
- Claims Fraud
- Data Science
- LTV Modeling
- Policy Change Modeling
- New Proprietary Behavioral Dataset
- Cross-sell modeling
- Malicious Users
- Bot Detection
- Solving the Risk vs Customer Experience Challenge
- Behavioral Analytics to Predict New Account Opening Fraud
- Predictive Analytics to Solve Tobacco Usage Nondisclosure
- Top 6 Use Cases for Predictive Analytics in Insurance (2021)