The insurance industry is no stranger to machine learning and predictive analytics. In fact, it's arguably the most sophisticated, having pioneered some of the most advanced statistics and modeling techniques more than a hundred years ago.
Consider how these insurers use machine learning today:
— Progressive Insurance is predicting claims from telematics and geospatial data;
— Zurich Insurance is supporting marketing, fraud detection, and claims management; and
— Transamerica is recommending products to customers.
Gary Reader, KPMG's Global Head of Insurance, summarizes the benefits of machine learning perfectly:
"For the insurance sector, we see machine learning as a fundamental game-changer since most insurance companies today focus on three main objectives: improving compliance, improving cost structures and improving competitiveness. Machine learning can form at least part of the answer to all three."
With mounting pressure from increased competition, a highly elastic marketplace, complex claims and fraud behavior, and a tighter regulatory environment, insurance carriers can no longer maintain a competitive edge without predictive modeling and machine learning.
For smaller companies that may not be investing as heavily in machine learning as some of the larger, well-known giants, it is important to realize how this technology can help improve organizational compliance and cost structures. Here are three practical ways the insurance industry is using machine learning today that can immediately improve business efficiency.
Insurers use machine learning to predict premiums, conversion, and losses for the policies that brokers submit based on the data available on the first day. This practice helps underwriters focus on the most valuable business. Detecting good risks early in the process enables insurers to make better use of underwriters' time and delivers a huge competitive advantage.
Insurance fraud is one of the biggest issues in the industry… and it's escalating. Conservatively, fraud steals $80 billion a year across all lines of insurance (Coalition Against Insurance Fraud). Fraudulent claims are costly, but it is too expensive and inefficient to investigate every claim. Even if possible, investigating innocent customers could prove to be a very poor experience for the insured, leading some to leave the business.
Machine learning enables insurers to sift through claims and identify those that warrant deeper investigation. Once fraudulent claims are prioritized, fraud units can then create a data-based queue — investigating only those incidents likely to require it. The resulting benefits are two-fold. First, your resources are deployed where you will see the greatest return on your investigative investment. Secondly, you optimize customer satisfaction by not challenging innocent claims.
In the time between an insurance claim’s initial filing and full payment, the amount of the claim can change drastically. The ability to predict the final claim amount has significant impact on financial statements, specifically the reserves and Incurred But Not Reported (IBNR) amounts reported in Quarterly Earnings statements. Additionally, loss cost modeling relies on Incurred Loss Amounts, which are undeveloped compared to fully developed loss amounts.
Building accurate predictive models leads to a better understanding of how much a claim will ultimately cost, and has the potential to save millions of dollars in claim costs through proactive management, fast settlement, targeted investigations, and case management. As a result, insurers can have confidence in how much to reserve for IBNR loss amounts. Using the predicted developed loss for each claim as the dependent variable, you will build more robust, accurate pricing models.
These are not the only use cases. Insurers currently use machine learning across a variety of different business functions, including optimal pricing, direct marketing, conversion, targeting inspections and audits, predicting litigation, claims forecasting, and customer retention. The value and business benefits of the technology are significant:
- Delivers more accurate predictions than traditional analysis or human judgment;
- Modern techniques make these predictions easy to understand and transparent.
- With better predictions, managers make smarter decisions; and
- Smarter decisions produce more revenue, lower costs, and a better bottom line.
While machine learning used to be the exclusive domain of data scientists — who are hard to find, hire and retain — it is now possible for business users of all skill levels to build data models and make better predictions faster. Your organization already has domain experts: actuaries, claims managers, product and underwriting managers, marketing managers, and underwriters. With the right tools and training, anyone can contribute to machine learning projects and reap these benefits.
As you consider and evaluate machine learning for your organization, keep the importance of automation in mind. Seek out platforms that automate the entire workflow, not just pieces of it. Look for built-in best practices that ensure business users can produce valid and high-quality insights that your organization can put into production. And most importantly, seek transparency. When your team delivers a predictive model, it should be clear and interpretable to executives, stakeholders, and regulators.