Can Underwriting Be Made More Efficient?
Underwriters spend less than half their time actually underwriting. How machine learning and AI can remove the laborious verification that slows insurance down.
Underwriting is the process where individuals or institutions assume financial risk in exchange for compensation. This typically applies to loans, insurance or investments. The term’s origins trace back to practitioners writing their names under the risk amount they would accept for a given premium.
Current challenges
The underwriting process demands thorough examination of customer resources and extensive verification of applicant attributes. In health insurance, for instance, underwriters must validate BMI, lifestyle habits, medical history and related factors — often requiring periodic reassessment to ensure ongoing compliance.
Underwriters spend less than half of their time in underwriting, since their procedures are extremely laborious and involve a lot of verification. Key pain points include:
- Attending Physician Statements (APS) summaries — administrative burdens that disrupt workflows
- Trial applications / preliminary inquiries — hours spent reviewing extensive information for cases that may never materialize
- MIB follow-ups — time-intensive coordination across insurance companies
- Contestable claims — lengthy report reviews combined with original application analysis
Path forward
Developments in machine learning, artificial intelligence and deep learning offer solutions to these obstacles. At Sparshik Technologies we are making steady progress toward automating these inefficient processes.