ValueSpring’s enterpriseMind is cognitive computing for Insurance Underwriting. enterpriseMind automatically interprets and synthesizes external and enterprise data to calculate more accurate policy risks.

The end result? Faster underwriting and premium quoting; improved combined ratios, risk segmentation, loss run analysis and pricing; and real-time risk and underwriting management.


“We need to better understand risk sources to find stronger risk predictors for faster and more precise underwriting decisions and pricing with less policy application data.”


Types of data used as input to underwriting analysis

  • Structured Prospect and Asset Data
  • Third-party market data
  • PDF files
  • Email
  • Social media

Result of enterpriseMind calculations

  • Unified view of customers and risks
  • Optimized underwriting models for more rapid classification of policy prospects
  • Hadoop Predictive Analytics Data Tables for Market Research

Need for understanding critical risk areas

Whether for commercial or personal lines of business, the Chief Underwriter needs to find more powerful predictors of asset and policy risk. Risk data can always be improved, but the real bottleneck lies in human cognitive limitations of being able to rapidly integrate and find the most critical risk information from information sources already available. The challenges are to 1) create a fused view of the data at hand and 2) highlight key points of potential risk and concern.

Need for understanding critical risk areas

Analytics capability is key

In order to better understand the critical risk areas, two types of analytics capabilities are required: 1) text analytics, to decompose documents and social media into analyzable data structures; and 2) cognitive analytics, to infer causal links between targeted customer characteristic and plausible predictors.



Automatically analyzing all data

enterpriseMind’s cognitive analytics, combined with its advanced grammars and machine learning, seamlessly fuse structured data and unstructured social media and documents into propositions representing more than just what is literally stated, providing the basis for more rapid and insightful underwriting.

Intelligent data for improved underwriting


enterpriseMind then uses real world knowledge from its Transformer Knowledge Libraries (TKLs) to perform Cognitive Analysis on the material, looking for any customer characteristics or other circumstances that might have a causal relationship to being a high value customer. In this way, underwriters and risk managers are able to focus on the risk and pricing factors that really matter, and insurance companies can make better use of their underwriting talent.