The THiNK Blog

grab a coffee and give yourself the freedom to read



coffee-cup-zoomed-in-hero

Big Data and Predictive Analytics in Life Insurance Underwriting - PART II

Tue, 28 Jun 2016 14:02:36 +0000 / by Mervyn Gillson

life-insurance-underwriting-bi-data-part-2-1

How does big data truly relate to life insurance? In Part I of this series, I defined Big Data, the difference between structured and unstructured data, data analytics vs predictive analytics and briefly shed light on the current buzz big data is having in the life insurance industry. Now it’s time to expand more on that last note, and really understand the role big data has in life insurance and how it affects the profession of life underwriting. So let’s dive in:

How is Big Data currently being used in life insurance?

Underwriters should be aware of some of the basic ideas of predictive models as they are used in both life insurance pricing and marketing and now in more and more aspects of underwriting. In Deloitte’s 2010 article “Predictive Modeling for Life Insurance” there is a section on Life Underwriting. Therefore, even 6 years ago, the life insurance industry was looking for ways to incorporate/improve the use of this science to find ways to make the underwriting process faster, more economical, more efficient, and more consistent. Since 2010, new types of data are being collected and there is no doubt the life insurance industry will continue to find ways to incorporate this data. Some of these new types of data include:

  • more electronic health records,
  • data from smartphone apps,
  • data from wearables, etc.
  • public available data (medical research, government published statistics)

While they bring new opportunities to collect information, they also bring new challenges on how this data can be verified and used.

How does Big Data affect life insurance underwriting?

As mentioned in Part I of this series, predictive analytics is used to leverage models that predict underwriting decisions based on historical and current data. This has already helped some insurers cut down the time is takes to issue policies. In these instances, the turnaround time has been shortened to just days (versus a much longer period) because of fewer underwriting requirements. So what does this mean for decision accuracy? Well in the Deloitte article mentioned above, it was noted that frequently, the underwriting decision recommended by predictive analytics matched the decision given by full underwriting. This will only continue to improve going forward with the appropriate mining of the data and the building of applicable predictive models.

How will it affect underwriters and the profession of life underwriting?

Gen Re did a survey of life insurance companies in the US in 2014 and found that there were two forms of predictive modeling in life insurance.

  • The first one examines conventional underwriting evidence in new ways to improve risk classification and mortality projection.
  • The other form uses unconventional underwriting evidence to classify applications quickly and/or cheaply while replicating the decisions reached with conventional evidence.

So what will the future bring? I would expect that more and more companies will start to use the form that uses unconventional underwriting evidence as the collection of health data. My colleague, Nick Joly, wrote a blog back in December 2014, “Emerging Trends in Life Underwriting”, and in it he mentioned “Wearable Technology”. Many of the applications for these technologies are capable of monitoring a breadth of information that is then connected to the person’s computer/device to monitor fitness levels.

Will there be a time when all of this data will reside in a central location? Probably. In the US, there are already databases of medications that have been prescribed. Will this become more prevalent in other regions? In my opinion, yes likely. Will everyone’s medical records be stored in a central, confidential database? This process has already begun.

So how will this affect underwriting? There will be vast amounts of analytical data that can be leveraged to improve classification of risk.   As a result, there are already conversations around replacing the types of underwriting requirements that already exist. So stay aware of these developments. Be prepared to get involved in your company’s evolution into the future of life underwriting.

How can you make sure you don’t miss the next big thing in Life Insurance News?

LOGiQ3 Blog Subscribe

Topics: Underwriting, Innovation

Mervyn Gillson

Written by Mervyn Gillson

Become an industry thought leader. Don’t forget to share.