Churn prediction model for account representatives at a large paint company
A large paint manufacturing company was facing difficulty knowing which high-value accounts were the most likely to churn. The company estimated that it cost them five times more to attract a new customer than it does to retain an existing one.
Without any visibility into those accounts’ churn probability, their account representatives had difficulties prioritizing and deciding the next best actions for their accounts.
However, to build a reliable churn prediction model and make its results readily available for the account representatives, the customer needed a solution that could seamlessly integrate with their existing SAP implementation.
Neal Analytics built a modern cloud architecture that enables data sharing across the various divisions of the customer’s parent company.
The Neal team helped them unlock and pull SAP data, integrate it with the other data sources, and process the data using Azure Databricks.
Neal also built a churn model using this new integrated data set to help the customer better predict churn likelihood for their accounts. This prediction model could then be used by account representatives to gain more advanced behavior insights about their customers.
The modern architecture developed by Neal Analytics allowed data sharing between business units that enabled both cross-sell opportunities and enriched customer experience. The unlocked SAP data and churn prediction models were operationalized by displaying a churn likelihood score directly on the ARs Salesforce account page.
Now, account representatives can more easily prioritize accounts at risk of churning and decide the next best action accordingly.