Focus: How insurers can reduce operational costs and improve accuracy with AI driven workflows.

The intake layer in Group Benefits is still mostly manual. Census files arrive in dozens of formats. Column headers vary by employer. Data is incomplete, inconsistent, and sometimes contradictory. Before any underwriting model can run, before any eligibility check can happen, someone has to normalize all of it by hand. Underwriters spend 30 to 40% of their time on this kind of administrative work. Analysts can spend 40% just validating data before any real analysis begins.

This is not new. What is new is that the old fix, rule-based automation, has hit its ceiling. Standard RPA breaks when every employer’s file has a different layout. The variability is the problem, and bots built on fixed templates cannot handle it. Meanwhile, healthcare costs are rising 6 to 9%, employers want more flexible and personalized coverage, product lines are expanding, and Q4 enrollment season compresses the majority of annual volume into a few months. The team that cannot process cases fast enough loses business.

Carriers pulling ahead are applying AI-driven document intelligence directly to census intake: extracting, classifying, and normalizing employer data regardless of format, then feeding clean structured data into underwriting and eligibility workflows. This roundtable brings together Group Benefits leaders who are deploying these capabilities or evaluating whether to start, for an honest conversation about what is working and what is not.