"machine learning", "research", "AI"

Where the research led: Better and faster

Do you know anyone who struggles with perfectionism?

It’s a curious condition. In a sense, it is a positive. Who among us does not wish our work to be perfect? But perfectionism is usually thought of as a negative for one very specific reason: time.

Getting things perfect often take time. A great deal of it. And time is a valuable commodity. This is a big concern when you do research in actuarial science, as we do here at LifeScore Labs. Our mathematicians and statisticians are dedicated to perfecting underwriting through the use of machine learning and artificial intelligence.

In the first post in this series, we wrote about how the LifeScore Labs team applied its new machine learning-based algorithm to years’ worth of historical policies from MassMutual… and showed that the new approach outperformed traditional underwriting by 6 percent on the basis of claims.

That whopping improvement is about as close to a perfect result as you can get in this industry.

Of course, LifeScore Labs’ executives, sales personnel, marketers, and such are worriers by nature. They’re not as interested in the perfectionism of science as they are in the performance metrics of the company. And as anyone who sells life insurance can tell you, speed is the most important metric. How long does it take from inquiry to sale? How much time passes from when a consumer asks about life insurance until they buy it?

So when the AI and machine learning experts created a model that beats the traditional approach by 6 percent, the cynics among us had one question: Is it fast?

Fortunately, and remarkably, the answer was yes. As MassMutual deployed the new LifeScore algorithm, the company came to realize the new approach led to quicker sales.

In the first two years of operation, the LifeScore system reduced time to approve by 25 percent, leading to a 30 percent improvement in acceptance rate.

In simpler terms, MassMutual saw a 30 percent boost in policy acceptance by consumers in ultra-preferred policies and shaved a week off of approval time. 

That, as you can imagine, saved the company millions of dollars in operational efficiency.

There’s a word for results like that: perfection.

If you’d like to see how the LifeScore model works compared with your existing underwriting methods, just let us know. We’re committed to transparency in our research and are eager to help you gain efficiencies in your operation. 

This article is the second in a two-part series on the early research results at LifeScore Labs. Click here to see the first article.

CRN202112-257276

LifeScore Labs

LifeScore Labs

At LifeScore Labs, we partner with life insurers to deliver a data-driven risk-scoring model that makes the underwriting process more efficient, and the results easier to interpret for both insurers and the consumers they serve.