In Einstein Prediction Builder, what is the main distinction between binary and numeric predictions?

In Einstein Prediction Builder, the main distinction between binary and numeric predictions lies in the nature of the outcomes they predict. Binary predictions classify outcomes into two distinct categories, such as "Yes/No," "True/False," or "Success/Failure." This allows businesses to focus on understanding two opposing scenarios, which can be particularly useful for decisions that involve a definitive choice or outcome.

For example, a binary prediction could determine whether a customer is likely to respond positively to a marketing campaign (Yes) or not (No). This type of model is designed to clarify the likelihood of one of two particular results occurring, making it straightforward for users to take action based on the prediction.

In contrast, numeric predictions deal with continuous values or quantities. They are utilized for forecasting numerical outcomes, such as predicting sales revenue or the number of products to be sold. This type of prediction provides an estimate within a range rather than classifying into two categories, leading to insights that are more about measurement than classification.

Therefore, the essence of binary predictions is in their ability to facilitate decision-making through a dichotomous framework, which is effectively captured in the correct choice.