Melissa Butcher

How did you get into underwriting?

My underwriting career started in what we called the policy assembly unit at an insurance carrier. When I started in policy output, I had the chance to learn everything about life insurance policies—beneficiary changes, ownership updates, reinstatements, policy modifications, and more. Working closely with the underwriting team, it felt like a natural next step to transition into underwriting, even though I didn’t know much about it at the time.
‍
I’ve always loved doing research and solving puzzles, so underwriting made a lot of sense for me. When my company consolidated its regional life insurance offices, I moved to Colorado and was promoted into underwriting. That was over 20 years ago, and I’ve been at it ever since!

I’ve always loved doing research and solving puzzles, so underwriting made a lot of sense for me.

What does a typical day look like for you?

A couple of years ago, I transitioned from production underwriting to working on automated underwriting platforms. So now, my day is a mix of project work and quality reviews.
‍
A big part of my job is reviewing cases that go through our automated underwriting system to optimize decision rules. I look for trends, manage high-risk cases, and refine our system to make sure we're striking the right balance - not being too conservative or too lenient.
‍
I also work closely with data scientists and engineers to improve our platform. Our goal is to increase straight-through processing so underwriters don’t have to manually review every case, freeing them up for more complex risk assessments.

What’s the most rewarding part of your job, and what do you find most challenging?

The most rewarding part is when we make a system change based on trends, and then I see a case go straight through without manual intervention. I love sharing those wins with our data scientists and engineers—because when an applicant gets approved automatically, it means our system is working exactly as it should.

The biggest challenge? Getting the system to think like an underwriter. Risk assessment isn’t one-size-fits-all. A single piece of information—like alcohol consumption—can be assessed differently depending on the situation. How do we teach a system to make those nuanced decisions? That’s the hard part.

The biggest challenge? Getting the system to think like an underwriter.

Bias and applicant behavior also play a role. We’ve found that simply rewording a question for applicants can lead to completely different answers so we’re constantly refining how we ask questions and what we ask to strike the right balance—ensuring we collect the right information while minimizing anti-selection and misrepresentation. That’s the real challenge.

How has technology impacted life underwriting over the past 20 years?

I think the biggest change is that when I started, underwriting was completely paper-based. We had physical files lining the walls, and applications were mostly handwritten. Even before COVID, my previous company was still processing 80% of applications on paper. But once COVID hit, that flipped to 80% electronic almost overnight.

Beyond that, we now rely heavily on third-party data, which has drastically reduced underwriting requirements. We can access critical information faster and make more efficient decisions without always needing medical exams and lab work.

Beyond technology, what’s the biggest shift you’ve seen in underwriting?

Applications are being approved much faster with fewer requirements. Twenty years ago, almost every applicant had to undergo medical exams and lab work. Today, we gather risk insights through electronic health records and other data sources.

If we need additional details, we request specific information rather than automatically ordering full medical tests.

Have customer expectations around life insurance changed in recent years? What’s driving that change?

Definitely. People want things to be fast and easy. Customers expect little to no friction in the application process. Everything is instant now—grocery deliveries, online shopping—so they expect life insurance to be just as easy.

Everything is instant now—grocery deliveries, online shopping—so they expect life insurance to be just as easy.

Social media and influencers also shape expectations. If someone has a smooth experience with an insurer and shares it online, others expect the same. But when applicants don’t fully understand the underwriting process, it can lead to frustration. For example, some don’t realize why we need their driving history—it’s a risk factor, just like medical history.
‍
Without an agent or financial representative explaining the process, applicants can become frustrated by the information we require.

AI is a big topic in underwriting. Where do you think it will have the biggest impact?

I would say probably with speed and efficiency, which then will turn into improved risk selection. AI will be able to help summarize medical records faster, detect missing application data, and even flag inconsistencies. For example, if an income field looks inconsistent with the coverage amount applied for and stated occupation, what if AI would be able to automatically prompt an agent to verify it? This would save underwriters from having to verify income manually.

Right now, if you're a life underwriter, about 40% of your time probably goes to admin work—even if it doesn’t feel like it.

I think AI will cut down the repetitive admin work so underwriters can focus on the tough cases that require in depth human judgment. Right now, if you're a life underwriter, about 40% of your time probably goes to admin work—even if it doesn’t feel like it. Requesting more information from financial representatives, sorting through application data, and double-checking details — it all adds up. AI can take a big chunk of those tasks off your plate.

There’s a lot of talk about a talent gap in underwriting. Have you noticed that?

I don’t think it’s a lack of talent per se —I see it more as a knowledge transfer issue.
Remote work has made it harder for new underwriters to learn from experienced ones. When I started in an office, I could turn to a mentor, overhear conversations, and ask quick questions. That kind of organic learning is tough to replicate remotely. Without that daily exposure, new underwriters miss out on the practical, real-world insights that help them develop faster.

Fewer new perspectives enter the field, and industry knowledge isn’t being passed down as effectively.

Another challenge is that fewer companies are training underwriters from the ground up. Most hiring today is lateral—meaning underwriters move between companies rather than fresh talent being developed. As a result, fewer new perspectives enter the field, and industry knowledge isn’t being passed down as effectively.

Underwriting is a constant learning process. It takes at least six months for new underwriters to get up to speed, and they typically start with low approval limits while their work is closely reviewed.

Even after almost 10 years at my first company, switching to a new one made me feel like I was starting from scratch. Without structured mentorship and training, it’s easy for new talent to feel lost, making the learning curve even steeper.

Are there any industry conferences, podcasts, or events you attend regularly?

I’ve attended AHOU (Association of Home Office Underwriters) national conferences, which are fantastic for networking and staying up to date on what’s going on in the industry. 

I’ve also participated in NEHOUA (Northeast Home Office Underwriters Association) events. I even earned a scholarship to attend  an AHOU conference through NEHOUA. All eligible underwriters should check this out! 

I’m a big believer in lifelong learning. I recently went back to school for a data analytics certificate so I could better understand what our data scientists and engineers are talking about! It’s a challenging course, but I love being able to speak their language when discussing risk models and automation.

Do you have any book recommendations—either for someone starting in underwriting, learning about risk, or just a book you really like?

I don’t have any underwriting-specific book recommendations because I read everything—mystery, love stories, thrillers. But I do have a favorite quote:

"Don't wait until life isn't hard anymore before you decide to be happy."

I think it’s important to focus on the good, because there will always be challenges. And so far, you’ve made it through 100% of your challenges.

Looking ahead five to ten years, how do you think life underwriting will change?

I see a continued shift in how we underwrite. Requirements will no longer be strictly based on product, age, and coverage amount. Instead, we’ll move from traditional underwriting requirements to more of a waterfall approach—where data from third-party sources flows in as an application is submitted, helping determine what additional information is actually needed.

For example, instead of requiring full medical exams, we may only need electronic health records based on what the initial data tells us. With advancements in AI—while ensuring we’re not introducing bias—a lot of the repetitive tasks underwriters handle today will be automated.

With advancements in AI—while ensuring we’re not introducing bias—a lot of the repetitive tasks underwriters handle today will be automated.

We’ll still use the same key tools—MIB, MVRs, prescription history, medical diagnosis codes—but we’ll apply them more selectively to answer only the most relevant underwriting questions. As data science continues to evolve, we’ll collaborate more with data teams to develop models that combat misrepresentation and fraud. These models will be flexible, allowing adjustments based on a carrier’s specific needs, distribution channels, customer base, and products.

I am very positive towards the future of life underwriting

I am very positive towards the future of life underwriting! I think we’ll be able to manage workloads more efficiently, even with fewer underwriting professionals, while still maintaining mortality expectations.