Creating a churn prediction using an ML model

This demo shows you how to create a new Prediction in Prediction Studio.

U+ Bank uses Pega Customer Decision Hub™ to personalize the credit card offered to customers on their website.

For customers who are likely to leave the bank soon, like Troy, the bank wants to make a proactive retention offer instead of a credit card offer. To predict the likelihood that a customer may leave the bank soon, you use a predictive model to drive a new Prediction. Predictions are managed in Prediction Studio.

You use Customer Decision Hub Predictions to improve customer engagement. Predictions for case automation and text analytics are also available.

To create a Prediction that aims to calculate the likelihood that a customer might churn soon, set the outcome to Churn and the subject of the Prediction to Customer.

Notice that initially, a placeholder scorecard is generated and used to drive the new Prediction. This placeholder is useful in case you do not have a predictive model yet, as it allows the Next-Best-Action Designer to continue work while a predictive model is built.

As you already have a predictive churn model, the next step is to replace the placeholder scorecard with your model. When the replacement is ready for review, approve the candidate model.

When you submit the new Prediction for deployment, a change request is automatically created in the current revision.

The changes come into effect after the Revision Manager deploys the revision. Changes from the Business Operations environment are pushed to the Development environment, and from there to other environments through the Business Change pipeline.

You have reached the end of this demo. What did it show you?

• How to create a new Prediction.
• How to replace the generated scorecard with a predictive model.
• How the new Prediction is deployed.