endorsedCompleted

Insurance Claims Analytics & Policy Segmentation

Services Covered:

Big Data Analytics and Business Intelligence • Data Engineering • Custom Software Development

Industry

Financial Services

Duration

12 months

Budget

200K - 999K

Start Date

1 June 2018

Team Size

1-5 Employees

Engagement Model

Onshore

Project Capability Score (PCS)

75/100

Good

Project Capability Score (PCS) estimates how capable INSART is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

One Inc.

Folsom, United States of America

Project Summary

INSART implemented a big-data analytics engagement for an insurance client to identify which categories of policyholders were more likely to request compensation. They assembled a dedicated team to perform database querying, descriptive statistics in R, and in-database analytics to segment policies and identify outliers. The work helped define insurance fraud scenarios, improve underwriting accuracy, and increased operating margins; stakeholders reported satisfaction with the vendor's expertise.

Key Challenges

Identify which categories of policyholders were more likely to request compensation
Analyze policy pricing to determine overpriced and underpriced coverage types

Project Deliverables

Data querying and preprocessing scripts
Descriptive statistical analysis implemented in R
In-database analytics and policy segmentation (zero/one/two/three+ claims)
Outlier detection and comparative analytics across coverage types
Analysis comparing policy distribution over 6 or 12 months to identify overpriced/underpriced coverages

Project Solution

INSART assembled a dedicated engineering team to perform the engagement. In stage one they executed database querying and preprocessing to detect anomalies, applied descriptive statistics using R, and ran in-database analytics to bucket policies into groups by number of claims. They compared group properties and data metrics (claims counts, profit/loss ratios) against the full dataset to identify outliers. In stage two they analyzed policy distributions over 6- and 12-month windows across coverage types to determine which coverages were overpriced or underpriced and to identify groups with the highest claim values.

Project Outcome

The work helped define insurance fraud scenarios and decrease losses
Improved the accuracy of the underwriting process
Increased overall operating margins
Stakeholders were satisfied and noted INSART provided depth of expertise and skilled data engineers

Platforms

Cloud
ERP

Tech Stack

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Stakeholders are happy with the results. INSART offered the depth of expertise required to complete the difficult tasks. They went beyond technical tasks to offer valuable insight into how to reach internal goals.”
Dimitry Proshak
Project Manager
Date: 12/09/2019