Cortical.io > Case Studies > How a major commercial property insurer leveraged SemanticPro to globally standardize policies

How a major commercial property insurer leveraged SemanticPro to globally standardize policies

Customer Company Size

Large Corporate

Region

America

Country

United States

Product

SemanticPro Extract & Analyze

Tech Stack

Natural Language Processing
Machine Learning

Implementation Scale

Enterprise-wide Deployment

Impact Metrics

Productivity Improvements
Cost Savings

Technology Category

Analytics & Modeling - Natural Language Processing (NLP)
Analytics & Modeling - Machine Learning

Applicable Functions

Business Operation

Use Cases

Regulatory Compliance Monitoring
Fraud Detection

Services

Data Science Services

About The Customer

The customer is a major commercial property insurer based in the United States. The company has a global presence with offices worldwide and serves about 2,000 high-value commercial customers. Each of these customers has as many as 30 policies, which are created at the company's head office and then forwarded to regional offices for local adaptation. The company does not use industry-standard forms, which means that the binding copies of locally issued documents may differ from the original in both format and content. The company's team spends approximately one third of their time searching for differences between the source policy and the final version, a process that is both manual and time-consuming.

The Challenge

The commercial property insurer with offices worldwide has about 2,000 high value, commercial customers with as many as 30 policies each. The head office creates standard policies which are forwarded to the regional offices and adapted locally. As the company does not use industry standard forms, the binding copies of locally-issued documents may differ from the original in format and content. The process of reviewing the locally-issued policies is done manually and is time consuming. The team responsible for this task spends approximately one third of their time on searching for differences between source policy and final version. So far, this review process could not be automated because of the different file types and formats, and because no tool could understand semantic variations. However, the company sought for an automation solution, as about 70% of the documents still contained errors after the manual review.

The Solution

The company decided to leverage SemanticPro Extract & Analyze to automate the comparison of standard policies with locally-issued documents. To cover the variety of formats and language, SemanticPro Extract & Analyze was trained with 100 documents coming from the different regions based on annotations from the company’s subject matter experts. Once trained, the solution is able to compare policies on a word-by-word as well as clause-by-clause basis and understands different formulations of the same concept. For example, it recognizes a Force Majeure provision where the word “war” has been mistakenly replaced with “conflict”. The solution quickly and accurately reports the differences in all terms and conditions between original documents and locally-issued policies, enabling the company to perform timely corrections in accordance with tight deadlines.

Operational Impact

The company was able to automate the process of comparing standard policies with locally-issued documents.
The solution was able to understand different formulations of the same concept, thus increasing the accuracy of the comparison.
The company was able to perform timely corrections in accordance with tight deadlines thanks to the quick and accurate reports generated by the solution.
The company was able to reduce the risk of policy deviations to a minimum.

Quantitative Benefit

Reduced the time spent on searching for differences between source policy and final version by a significant amount.
Reduced the error rate in documents from 70% to a minimum.