A luxury travel company wants to optimize its high-value customer retention strategy.

The target outcome is the total dollar amount that a customer is expected to spend, so the prediction variable is quantitative and continuous. Salesforce defines Regression for Numbers as the predictive-model type for numeric outcomes such as currency, counts, percentages, and other measures. Therefore, a regression model is appropriate for estimating future booking spend. Binary classification would be appropriate only for an outcome with two categorical states, such as “will purchase” versus “will not purchase.” Multiclass classification is used when predicting one of several categorical outcomes. A foundational model is not the native predictive-model type used for this structured numeric forecasting requirement. Consequently, B correctly aligns the model architecture with the numeric target variable specified by the business requirement.

================