The Recency, Frequency, Monetary (RFM) Aggregate node allows you to take customers' historical transactional data, strip away any unused data, and combine all of their remaining transaction data into a single row (using their unique customer ID as a key) that lists when they last dealt with you (recency), how many transactions they have made (frequency), and the total value of those transactions (monetary).

Before proceeding with any aggregation, you should take time to clean the data, concentrating especially on any missing values.

After you identify and transform the data using the RFM Aggregate node, you might use an RFM Analysis node to carry out further analysis.

Note that after the data file has been run through the RFM Aggregate node, it won't have any target values; therefore, before using the data file as input for further predictive analysis with any modeling nodes such as C5.0 or CHAID, you need to merge it with other customer data (for example, by matching the customer IDs).

The RFM Aggregate and RFM Analysis nodes use independent binning; that is, they rank and bin data on each measure of recency, frequency, and monetary value, without regard to their values or the other two measures.