Sales machine learning enhancements

• Enhanced Feature Engineering:
  Machine learning models now use quantile binning for numerical and date fields, enabling more accurate representation of skewed sales data.

• Richer Prediction Explanations:
  Explanations now include key factors from related and child records, as well as aggregate values (e.g., maximum or average from child records) that influence predictions.

• Improved Model Analysis:
  A new report provides summary statistics and trends for the last five prediction runs, including prediction counts and distributions.

• Increased Model Accuracy:
  More precise data preparation and feature selection drive better model performance and more reliable outcomes, even for complex or uneven datasets.

• Deeper Predictive Insights:
  Users gain clearer visibility into the factors affecting predictions, supporting informed decision-making at every sales stage.

• Actionable Model Transparency:
  Enhanced reporting and explanation tools help users understand, trust, and act on machine learning predictions

Steps to Enable and Configure

Build and deploy machine learning models for sales use cases

1. Navigate to Configure Sales Machine learning setup and maintenance task
2. Duplicate from available system use cases and configure machine learning model
3. Alternatively, use create button to create a custom use case from scratch
4. Follow guided step process to build and deploy the machine learning model