Life insurance underwriting processes are evolving rapidly - data-driven models deliver speed and scale, while traditional methods, like paramedical exams and laboratory testing, provide direct measures of current health. While data accelerates decisions, traditional underwriting continues to play a critical role in ensuring accuracy, risk integrity, and long-term performance. So which approach is the "best" for underwriting moving forward?
The shift to data-driven underwriting
The way life insurance underwriting is processed has undergone a dramatic transformation. Advances in data, analytics, and AI have enabled insurers to evaluate risk faster than ever, eliminating the need for medical exams and fresh lab results in many situations.
Modern underwriting increasingly relies on predictive risk models leveraging third-party data sources such as prescription histories, medical databases, and financial indicators. These tools enable insurers to assess risk using information that already exists, often reducing underwriting requirements and allowing many applicants to move through the underwriting process more efficiently. For straightforward cases, this can improve the applicant experience while helping insurers evaluate risk at scale.
Balancing efficiency and risk assessment
While accelerated underwriting programs have delivered meaningful efficiency gains, industry research highlights the challenge of mortality slippage - where risk class decisions diverge from fully underwritten outcomes. Reported experience varies widely, but many studies place average slippage in the mid-teens, with common drivers including build, blood pressure, tobacco or substance use, and depression or anxiety -factors traditional evidence can validate.
One reason for this is that automated underwriting relies on existing and observable data. If information hasn't been captured, recorded, or surfaced, it simply doesn't exist within the model. Conditions that are undiagnosed, in their early stages, not yet treated with prescription medication or treated with non-scripted medications including compounds, or not disclosed on an application may not have created a data footprint and therefore may not be visible through available data sources alone.
Measuring the unknown
Traditional underwriting, including paramedical exams and lab testing, helps close this gap by directly measuring current health. By capturing vitals, build, and biomarkers (such as lipids, glucose indicators, liver and kidney function) plus signals for nicotine and other substances, this evidence provides an objective snapshot at the time of underwriting that can surface physiological signals before they appear as diagnoses, prescriptions, or coded medical encounters.
The path forward: A tiered, hybrid model
The future of underwriting is not a choice between data and traditional methods—it is a blending of both.
Accelerated, data-driven underwriting works best for verifiably healthy applicants, moderate face amounts, and straightforward histories. But as age, coverage, or medical complexity rises, so does uncertainty - and the value of independent evidence.
This highlights the need for a hybrid approach that pairs data-first triage with targeted evidence collection: accelerating cases where confidence is high, gathering additional evidence where uncertainty or value is greater, and monitoring outcomes through holdouts and post-issue audits. This kind of model can help balance customer experience with mortality and anti-selection management.
PSG perspective
At PSG, we see the strongest underwriting models as those that can adjust evidence requirements to fit each case - using automation and data to streamline decisions while applying configurable rules and workflows to request more evidence when it is likely to improve confidence. The objective is a repeatable, auditable process that strikes the right balance of speed, precision, and outcomes.
Each organization's implementation plan will vary based on portfolio composition and risk tolerance. Wherever you fall on the continuum between automation and traditional underwriting, PSG can support your approach. Our capabilities span the full spectrum - from digitizing application and requirements gathering to enabling fully automated underwriting platforms, which help you execute efficiently while staying aligned to your risk objectives.
References
[1] Swiss Re. ‘Accelerated Underwriting in Focus’ (Sep 12, 2024)
About Jeff McCauley
Jeff McCauley is the President of Paperless Solutions Group (PSG), an MIB business. He joined MIB as part of the acquisition of PSG in December of 2020. Prior to that, he served as PSG’s President / COO for 11 years. Jeff has extensive management, sales, marketing, and operational experience, having worked for vendors, service providers and carriers all focused on the life insurance market. These experiences allow Jeff to see the industry from many different perspectives. Jeff is very involved in several industry trade groups, all of which are focused on progressing automation. Jeff graduated from Virginia Computer College and has taken advanced courses from LOMA and others that focus on Management and Leadership.