Einstein Prediction Builder: Understanding Your Scorecard Metrics and Prediction Quality

So, you’ve built a prediction with Einstein Prediction Builder . You’ve taken the steps to understand whether or not that prediction is indeed working correctly. You’ve even researched how to think through those predictions while keeping bias in mind in order to get the most accurate set of predictions moving forward. And now, you’re ready to take a deeper look at your scorecard metrics and the quality of your prediction. Congratulations, #AwesomeAdmin—you’re building incredible efficiency into your org by utilizing all the valuable components of AI and machine learning! Now, let’s dive into how you can better understand your scorecard metrics.

In this post, we’ll walk through an example scorecard to learn more about your predictions in Einstein Prediction Builder. The scorecard is the go-to place to determine prediction performance and to identify possible modeling issues. In this use case, we’re using example customer data to predict whether or not