- Key insight: Customer demands drive push for faster operations
- Forward look: Underwriters freed from analysis work will deal with risk more effectively
Modernizing how insurers source and use data is proving key to insurers getting the most out of AI and technology. Getting a handle on data helps improve operational speed and scale, according to John Kain, financial services business development lead at AWS. Advancing technology through better data management changes underwriting and can improve
This article is excerpted from a longer interview and edited for clarity.
How does AWS provide services to insurers?
Most of our customers have a very complex data estate across the organization. Most of them having siloed either by business line or in different technology systems. The ability to unlock that across the organization and combine it with third-party data sources is so key to driving the analytics that drive customer experience.
As you drive insights about customers and improve the overall customer experience, you then have to actually get those into production. Legacy systems have been a little bit of a detriment for how fast we can get applications into production. So being able to drive modernization across the whole stack is key to driving that success across our customers.
What will AWS focus on at ITC Vegas, in presentations and with customers?
It's fundamentally where are we from an industry perspective in insurance, from a Gen AI adoption perspective. The quick story there is I don't think we're really struggling with what the use cases in insurance are that are particularly open from a generative AI perspective. There's been enough repeatability across the industry and enough customer examples to know where to invest.
Customers understand what they have to do from a governance, compliance, and security perspective to move individual applications to production. What the industry's really focused on is more about how to do that at scale, at a much higher velocity. That means investing in platform capabilities on the data side, but also from the generative AI side to centralize governance and compliance. Then distribute the ability to innovate across the organization.
Insurers now have to do everything faster. What challenges does that create?
Customers expect an experience that's much more like they're used to from retail. Not only does the experience have to be more fluid, it needs to be within the cycle of where you're already engaging from an adoption perspective. A lot of the velocity is driven by the fact that many insurance processes are still quite paper-oriented, deal with unstructured data, and require a fair amount of human analysis of large data sets that are complex to actually make decisions.
We're seeing that the most in the underwriting process. There are examples where in order to kick off the process, the first thing you're doing is finding a subject matter expert to read a 200-page engineering report and come up with an assessment before you can even begin the underwriting process.
How is technology changing the underwriting process?
Traditionally, the underwriting process made sure to meet risks within a certain tolerance. Now with the ability to process that quickly, but also look at the overall portfolio more holistically, you can include the overall portfolio risk in the underwriting decisions, to make sure you're getting the right mix of products and exposure. The industry is improving underwriting processes for speed and consistency, and managing underwriting portfolio risk in real time.
How does technology free underwriters to focus more on risk?
Instead of having underwriters turning into analysts of unstructured data, they can focus much more on the overall risk of the underwriting. They're able to spend more of their time on the risk management side, like how do they think about the risk of the individual product that they're offering. It also frees them to think about how the risk fits within the insurer's overall existing risk pool, in real time. Underwriters can use judgment much more effectively from a risk management perspective than by thinking about the individual policy that they're underwriting.
