How product recommendation systems are helping insurers.
Brought to you by:
Over the past two decades our lives have been transformed by the information-rich Internet. At the hearts of digital giants like Google, Facebook, Amazon, Airbnb and Netflix we often find some ranking and filtering algorithms that use customer attributes to improve and customize predictions.
The insurance industry traditionally relied on human agents to analyse data and recommend products. The purchase of even a basic protection policy is often a slow and complicated process. Innovative pioneers, sometimes starting with simple price comparison websites, have developed sophisticated decision-support tools which simplified the process of insurance purchases, cutting costs and increasing customer satisfaction. Some of these digital companies have gone on to acquire government licenses to sell their own insurance policies. Such trends of technological innovation, collectively called Insurtech, have been accelerated in recent years by billions of investment dollars frequently seen in the news headlines. In this research, partnered with a traditional insurance company, we studied product recommendation systems and demonstrated how contemporary analytic and prediction techniques can be applied to a kind of customer data different in frequency and certain details from those of digital retailers. We surveyed existing paradigms and algorithms of recommender services both within and outside the insurance industry. We looked at the various problems they face, how they are adapting their algorithms to solve them, and drew common themes and new lessons for the insurance business. Along the way we also studied behavior-based classification of customers, customer life-time values to the company, and cross-selling and bundle-sales.
A free copy of the report is available here
To discuss the report or have a conversation about life insurance sales recommender systems email firstname.lastname@example.org
Disclaimer: This article presents information of a general nature and is intended solely for educational purposes. It is not intended as advice for any specific situation and may not be relied upon for any purpose. You should always consult qualified professionals familiar with your circumstances before adopting any strategy or taking any action. Milliman does not guarantee the veracity, reliability, or completeness of this e-Alert and has no responsibility for damages alleged to have been caused by it. This e-Alert is not directed at residents in any jurisdiction where it would contravene local law.