Rapid Predictive Modeling for Better Underwriting & Risk Management
Case Studies
Case Studies

Rapid Predictive Modeling for Better Underwriting & Risk Management

Cover Whale reduced data engineering overhead, realized faster insights, and more accurately underwrote policies.

About the Client

Cover Whale is an insurtech company focused on technology, underwriting, and data to provide innovative insurance programs for the commercial auto industry.

Use Case

Reduce Insurance Underwriting Risk

Industry

Financial Services & Insurance

Before
  • Delayed insight & complex risk driver identification process
  • Expensive data engineering overhead
  • High cost for risk misalignment
After
  • Faster time to insight & underwriting
  • Lower underwriting risk & loss runs
  • Reduced data engineering overhead
  • Better customer cost for risk alignment

The Objective

Cover Whale sought to further improve driver safety and reduce underwriting risk by better understanding prospective and active customer behaviors without building extensive ETL pipelines.

The Challenge

Cover Whale is the leader in fleet and commercial trucking insurance. They must analyze risk when determining premiums and whether to approve or renew policy coverage. To do so, it needs to assess risk indicators from several data sources, including a plethora of driving telematics, internal claims data, and third-party data (e.g., weather information, national highway data, etc.).

Traditionally, Cover Whale must duplicate, move, align, and/or combine these disparate data sources to glean insights. This process was laborious and costly. In addition, the time-consuming nature of data movement inherently delays and restricts the number of third-party sources it can utilize, limiting its ability to experiment with new data sources and optimize risk predictions.

Data Challenges
3+ Disparate Data Sources
3+ Disparate Data Sources
Heterogeneous Schema
Heterogeneous Schema
Sensitive Data (PII)
Sensitive Data (PII)
Data Description
Tabular Format
Tabular Format
Gigabytes of Data
Gigabytes of Data
Internal Claims Data
Internal Claims Data
Third-Party Vendor Data
Third-Party Vendor Data

Devron Solution

Rapid Predictive Modeling for Better Insurance Underwriting & Risk Management

In a matter of hours, Cover Whale deployed Devron satellites across multiple disparate data sources (internal and external)—completely bypassing the lengthy data movement process. By unlocking access to more data, Cover Whale was able to use Devron's federated machine learning and data science platform to develop and train AI models rapidly incorporating novel and alternative data sources to better predict risk, to provide coaching to policyholders, and to better align premiums to risk.

In addition, Devron's privacy-enhancing technologies allowed Cover Whale to quickly analyze sensitive datasets without exposing any personally identifiable information (PII) contained with their claims database or otherwise.

Client Results

Devron empowered Cover Whale to continue to innovate services by prioritizing agility, data access, and the ability to gain data-driven insights over constructing data movement infrastructures.

As a result, they are able to reduce data engineering overhead, realize faster insights, and more accurately underwrite policies—ultimately improving risk management and safety.

Download case study
 ↗

Devron’s platform enables us to prioritize data science over data movement. It allows us to leverage predictive modeling across more types of data to price more accurately and respond to changing driving behaviors with enhanced policyholder privacy.

We anticipate increasing safety gains as Devron’s platform accelerates our integrations across many telematics and camera device vendors.

Tony Beirne
Chief Data Officer & Chief Actuary Cover Whale

Devron Benefits

Devron enables data science teams to realize enhanced outcomes, improved data science agility, and reduced overhead—all without moving or exposing data.

Better Pricing & Lower Risk

Cover Whale identified new risk patterns and indicators, enabling them to price their policies better and more accurately than their competitors—ultimately lowering the company's underwriting risk and loss runs.

Customer Guidance & Improved Safety

Cover Whale highlighted higher-risk policyholders, allowing them to remediate potentially risky driving behavior before safety incidents occur.

Foundation for Data-driven Insights

Cover Whale gained a future-proof system for data-driven insights enabling them to rapidly add additional data sources, formats, and types down the road for deeper insights without needing to go through the data movement process.