Dramatic churn reduction in insurance policies
Detect and prevent churn from unsatisfied customers
Detect and prevent churn from unsatisfied customers
Reducing churn among yearly renewing homeowner policies is very challenging given the assault on traditional insurers by insurgents. Yet, acquiring a new customer costs five times as much as retaining one.
An abnormally high churn rate (>10%) is synonymous with a problem in customer relationship management, specifically customer loyalty. In this case, an insurance company want to prevent non-renewal in home-owners policies.
Octopeek ingested all appropriate data into the Data Lake, then cleaned and standardized it. After Data Preparation, the data is enriched. The resulting new dataset included 10% internal customer data and 90% new data. More than 100 new features were added to the dataset.
The Data Lake was enriched with Open Data, social networking, and alternative other data. The cleaned data was then enabled to be treated by our supervised ML model.
The model is based on the prediction of phenomena from past observations.
The output was scored in order to classify the individuals according to two criteria: likelihood to stay or leave.
This then allows the marketing department to stay one step ahead by proposing deeply personalized offers to improve customer loyalty.
The insurance industry needs to move from ad-hoc experimentation with AI to enable everybody within the organization to use it to improve both operational efficiency and deliver business value.
Octopeek aims to become the platform that democratizes AI in the enterprise: it makes it possible for dynamic creation and provisioning of full-fledged enterprise AI applications, customizable by business analysts.
Thanks to this new model, Octopeek innovates on traditional segmentation and targeting methods to better conquer, retain, and satisfy customers.