Skip to main content
Client StoryData

Prioritising leads effectively at nib​

INDUSTRIES

Healthcare
Insurance

SERVICES WE PROVIDED

Data Science
Machine Learning

Business Challenge

nib is a health and medical insurance provider in Australia with approximately 1,500 employees and over 1.6 million members. The nib website provides an online quote service that enables potential members to browse insurance products.

nib wanted to determine the likelihood that a person using the quote service would purchase an insurance product, and provide this information to the sales team for follow-up.

nib’s data science team were seeking support to deploy the model, integrate with downstream systems, and ensure performance was optimised for production, read their own meter and provide the estimate, however getting an accurate reading is challenging without guidance.

Success Criteria

  • Optimise the speed and accuracy of the existing model to ensure it was ready for production use.
  • Engineer an end-to-end pipeline to deploy the model to production and enable future model enhancements to be released quickly and without risk.
  • Integrate the output of the model with Tealium (nib’s member data hub) to enable the sales team to focus on a refined subset of potential members who are most likely to buy.

Solutions

  • Mantel Group’s data scientists were able to optimise the model for performance in production without compromising accuracy.
  • Amazon SageMaker was selected for this project as it enabled integration with all other AWS services and external platforms (Tealium, Snowflake and Databricks).

Mantel Group’s machine-learning engineers containerised and deployed the model in nib’s AWS environment through the use of SageMaker. This provided nib with an automated, end-to-end pipeline making future model enhancements easy and low-risk to deploy.

Scoring customer leads

Outcomes

  • In 4 weeks Mantel Group optimised and deployed nib’s customer leads model using SageMaker.
  • Mantel Group worked with the nib data science team to upskill them on the model deployment process and tooling enabling them to run and optimise the service.
  • Provided the nib data science team with training and documentation so they could maintain and administer the service after hand-over.
Close Menu