By Montoux's Geoff Keast, originally shared in Forbes
Legacy actuarial technologies are responsible for significant yearly costs in productivity loss, opportunity cost and unnecessary "key person" and "black box" risks. Exceedingly long model run times, the need for highly specialized legacy software skills and the inability to integrate data sources with data science tooling severely limit actuarial productivity. These constraints also confine actuarial insights and outputs, limiting potential value to the rest of the business.
Many insurers understand the issues stemming from legacy technology and are aware that, with demand for actuaries increasing, cloud-based automation provides an option to solve longstanding issues.
Why automation?
Based on the factors mentioned above, here are a few ways that automation can improve insurers' actuarial processes.
- Increase in actuarial productivity. Automation can accelerate model run times from hours to minutes, or minutes to seconds, dramatically improving productivity for actuarial teams and reducing the cost of on-premise servers.
- Focus more on analysis, and less on data management. Actuaries are expensive — and rightly so. By automating data management, actuaries can spend more time providing insights into customers, competitors, pricing, product development, sales activity and more.
- Eliminate key person and black box risk. The "key person" and "black box" issue is one of the most costly, and potentially dangerous, issues insurers must contend with from an actuarial standpoint. Beyond it simply being extremely risky to have only one or two individuals capable of understanding a particular model, it also wastes time. Automation can remove this issue; all actuaries working on a team can gain a transparent view of how the model is working.
- Improve outputs and insights. Automation makes it possible for actuaries to analyze thousands of scenarios at once, delivering a deeper, richer set of outputs and providing the business with better data and clarity when making commercial decisions.
- Take actuarial data out of its silo. Actuarial teams spend more time dealing with insurance data than any other, yet traditional systems and processes leave a lot of the good stuff siloed away in those teams, limiting the commercial gains an insurer can make from that information. Automation delivers the data in an accessible, easy-to-read format which can be shared easily with decision makers.
With quicker runtimes, user-friendly models and clear outputs, actuarial teams can work more collaboratively and productively, putting their skills to work on projects which deliver strategic, high-value and impactful results for the company.
How do insurers go about gaining the benefits of actuarial automation?
Automation may provide significant opportunities, but it's not without its challenges. Many processes and technologies which can be automated have been in place for decades, and shifting away from them is not necessarily simple. An insurer looking to get started with actuarial automation should consider some of the following questions.
- Which processes and technologies are the most costly in terms of productivity, time and resource waste? Are there suitable automation alternatives?
- Which teams and individuals would be most significantly impacted by automating a particular process, and what resources do they need to facilitate that shift?
- Are there certain processes, like experience study, which are low-hanging fruit and can be easily automated without requiring significant effort?
Once an insurer has an idea of where they want to start automating, the next question is which vendor to work with. One place to start is within insurers' actuarial teams themselves — are there any vendors or technologies they've heard of or trust which would be suitable? There are a couple of qualities to look for, including being cloud-based and incorporating both actuarial and data science, which will make actuarial automation easier.
Once a potential partner or vendor is identified, here are a couple of starting questions to figure out of it's the right fit.
- Is the automation platform or technology compatible with existing data sources and can it incorporate new ones?
- Is the technology transparent? Will all actuaries on the team be able to understand the models and the outputs without difficulty?
- What are the benefits of automating a specific process with that technology, including speed, transparency, granularity, usability, etc.?
- How long will it take to automate a specific process with this technology?
Actuarial automation can dramatically reduce costs, improve actuarial productivity and reduce the dangerous risks associated with black-box technology and actuarial models. Life insurers can begin gaining these benefits by identifying appropriate processes to start and speaking with vendors on what the automation process looks like.
