There are many ways that machine learning and cognitive computing can help insurance companies evolve into the types of service providers that customers are demanding in our connected and advanced world.

Some of these business benefits are connected to greater efficiency, while others help provide a unique customer experience (and quite a few, such as assessment automation, check both boxes).

Better Risk Detection and Accuracy During Underwriting

To help underwriters focus on the most valuable business, insurance companies can use machine learning to analyze massive amounts of existing data, including existing premiums, appraiser reports and more, to predict:

  • Types of insurance and coverage plans new customers will purchase
  • Premiums and policy updates
  • Coverage changes and forms of insurance (health, life, property, flood) that will most likely be dominant
  • Conversion
  • Losses for policies that brokers submit based on data available on the first day
  • Fraudulent insurance claim volumes

Detecting good risks early in the process enables insurers to make better use of underwriters’ time and delivers a huge competitive advantage.

Fraud Detection

Fraud is a growing problem in the insurance industry. For example, insurance companies in the U.S. lost more than $50 billion in insurance fraud in 2016. Data on fraud patterns has always existed. Now, machine learning can help insurance companies identify these fraud patterns in an automated way by unearthing exceptions and alerting the insurance companies to potential fraud before it takes place. Insurance companies can also use machine learning to study and identify suspicious claims that need to be further investigated. The potential for savings in this area is immense.

Assessment Automation

Machine learning’s ability to understand images and other unstructured data means that it can handle appraisal tasks formerly undertaken by human insurance appraisers. For example, following a storm, cognitive computing systems can quickly and accurately analyze images retrieved from drones flown over damaged properties. The systems can be taught to identify the damage by viewing thousands of images of similar damage and appraiser reports. They can then perform analyses almost instantly. This results in quick customer payouts, without requiring an appraiser visit. The result: lower costs for the insurance company and increased customer satisfaction.

For additional information, please check out my NEW white paper: Learning about Machine Learning. It examines the upcoming challenges and explains how insurers can maximize opportunities.

  • accuracy
  • assessment automation
  • Cognitive computing
  • drones
  • fraud
  • Machine learning
  • Risk detection
Gil Maletski

Gil Maletski Gil Maletski is the chief technology officer for the general insurance division at Sapiens. He possesses strong software architecture and design capabilities, with deep managerial, business and technical understanding.