AXA augments point-of-sale customer data with external sources for predictive underwriting
Louis DiModugno · Big Data in Insurance // Louis DiModugno, Chief Data Officer at AXA US [FirstMark's Data Driven] · May 23, 2016 · at 5:30
Louis DiModugno, Chief Data & Analytics Officer at AXA, outlines how AXA incorporates external health and wearable data into life insurance underwriting.
0:00 / 0:25exact quote · 25.9s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“Our actuary, ah, efforts and our predictive underwriting efforts have really started to take advantage of the information that we're collecting at the point of sale, augmenting it with additional information that's accessible maybe through prescription databases looking at down the road electronic health records maybe even using Fitbit type information to really give us a much better understanding of our customers going forward.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Louis DiModugno
AXA spends 90% of data project time on preparation, 10% on modeling
“One of the biggest challenges that we've had in this two years of existence is that of the biggest projects that we've been working on, it's been, you know, 90% data access and preparation, 10% model building.”
AXA leverages telematics in Europe for auto insurance pricing adjustments
“Ah, we also have a big, ah, initiative in Europe as well with telematics right now. So in the property and casualty space, specifically in auto, we get a much better, ah, capability to get better pricing based on how How people are using their automobiles and …”
DiModugno: Diabetes patients are a heavily underserved community in the insurance industry
“In the insurance space, it's a very underserved community.”
AXA aims to deploy algorithmic risk scoring across its entire book by 2017
“Our intent is that within the next, ah, year to 18 months, this scoring mechanism will be essentially across our entire book of business.”
AXA converts younger actuaries into internal data science roles
“What we found is that, quite frankly, we had some data scientists essentially within our midst, and so the actuaries that we had within our organizations, especially some of the younger ones, it was a very nice conversion for them to be able to pick up R codin…”
AXA US builds its analytics infrastructure on a Cloudera Hadoop stack
“So the one that we've got here in the U.S., it's primarily a Cloudera Hadoop stack that we've used a blueprint that was essentially blessed by our brethren over in French, in France and with that, we've got R and Python and Spark. We use some Dataiku along the…”