RFM Analysis Node Settings

Recency. Using the Field Chooser (button to the right of the text box), select the recency field. This may be a date, timestamp, or simple number. Note that when a date or timestamp represents the date of the most recent transaction, the highest value is considered the most recent; where a number is specified, it represents the time elapsed since the most recent transaction and the lowest value is considered as the most recent.

Note: If the RFM Analysis node is preceded in the stream by an RFM Aggregate node, the Recency, Frequency, and Monetary fields generated by the RFM Aggregate node should be selected as inputs in the RFM Analysis node.

Frequency. Using the Field Chooser, select the frequency field to be used.

Monetary. Using the Field Chooser, select the monetary field to be used.

Number of bins. For each of the three output types, select the number of bins to be created. The default is 5.

Note: The minimum number of bins is 2, and the maximum is 9.

Weight. By default, the highest importance when calculating scores is given to the recency data, followed by frequency, and then monetary. If required, you can amend the weighting affecting one or several of these to change which is given the highest importance.

The RFM score is calculated as follows: (Recency score x Recency weight) + (Frequency score x Frequency weight) + (Monetary score x Monetary weight).

Ties. Specify how identical (tied) scores are to be binned. The options are:

• Add to next. Select to move the tie values up to the next bin.
• Keep in current. Keeps tie values in the current (lower) bin. This method may result in fewer total bins being created. (This is the default value.)

Bin thresholds. Specify whether the RFM scores and bin allocations are always recomputed when the node is executed or that they are computed only as necessary (for example, when new data has been added). If you select Read from Bin Values tab if available you can edit the upper and lower cut points for the different bins on the Bin Values tab.

When executed, the RFM Analysis node bins the raw recency, frequency, and monetary fields and adds the following new fields to the dataset:

• Recency score. A rank (bin value) for Recency
• Frequency score. A rank (bin value) for Frequency
• Monetary score. A rank (bin value) for Monetary
• RFM score. The weighted sum of the recency, frequency, and monetary scores.

Add outliers to end bins. If you select this check box, records that lie below the lower bin are added to the lower bin, and records above the highest bin are added to the highest bin—otherwise, they are given a null value. This box is available only if you select Read from Bin Values tab if available.