Before you train a predictive model, specify a preconfigured template in Model Builder instead of building from scratch. Templates are available only for specific use cases. When you select an available template, Model Builder prepopulates the model details, variables, filter conditions, and a recommended algorithm to help you train an accurate model faster. You can review and customize these settings in the guided setup flow before training the model.
Where: This change applies to Lightning Experience in Enterprise, Performance, and Unlimited editions with Data 360.
When: This functionality is available starting in August 2026.
Why: Creating a model from scratch requires you to define every input, filter, and algorithm on your own. When you create a model from a template, you have a working starting point that reflects specific use cases.
-
Start with a set of prediction inputs and filters, and then customize them to fit your data.
-
Review your selections before training, and edit steps in the guided setup flow.
How: In Data Cloud, from the AI Models tab, click Add Predictive Model. Select Create from Template, choose a template, and then follow the steps in the guided setup flow.