From Risk Model to Life and Health Insurance Underwriting

Qumata collects digital data such as steps, exercise and heart rate. Our algorithm uses this information to identify the associated risk as well as extra mortality and morbidity value for each individual.

We can also generate an overall Extra Mortality and Extra Morbidity rate for insurers to use in their pricing.


Our performance has been validated by leading academic institutions and multiple insurance clients in Europe, Asia and the United States.

This includes studies on the correct predictions for risks of diagnosis and detailed comparisons to traditional underwriting. On average, we achieve an 87% accuracy in correctly predicting the onset of a condition (AUC 0.87 - area under the curve - a level of 87%).

Read the white paper below for more details on these studies and Qumata's results.

We provide results when it comes to differentiating prices for health and life insurance products. Our output delivers better risk classification and pricing differentiation.

Proof Points

Case Study: Jane’s life insurance policy

Jane’s Profile

Jane is 32 years old, a mother of two children and is looking to take out life insurance to protect her family from the unexpected.

Qumata Underwriting Result

Qumata's data-based, digital approach compares Jane's data to millions of similar people to accurately predict her medical risk profile and uses it to quantify her insurance premium.

Qumata includes hundreds of variables for Jane in these calculations, her health and lifestyle can significantly affect the final price for coverage.