The best indicator of future behaviour …. is past behaviour!
Did you do it before? You are likely to do it again
Did you do it often? You are more likely to do it again
Did you do it recently? You are likely to do it again soon!
Past Behaviour doesn’t tell us everything – but if you already have a database, it is probably your most accessible & cost-effective data source
Recency Frequency Monetary, or RFM, is a method of analyzing past customer behaviour, with a view to predict their future behavior.
Metrics: Recency Frequency Monetary (RFM)
The Question:
What simple ways are there to predict if customers will buy again?
Approach:
Analyze when they last purchased, how many times they have purchased, and how much they bought
Commentary:
Past behaviour is the best indicator of future behavior. In the direct mail days, RFM was more important (because of higher variable costs). It is less used now, but still useful for segmenting.
Recency = When was the last time you shopped
Frequency = how often have you shopped
Monetary = how much have you spent
Remember to eliminate the statistical outliers, who will skew the results.
Recency 5 (R5) = 3.49%
Recency 4 (R4) = 1.25%
Recency 3 (R3) = 1.08%
Recency 2 (R2) = 0.63%
Recency 1 (R1) = 0.25%
Frequency 5 (F5) = 1.99%
Frequency 4 (F4) = 1.56%
Frequency 3 (F3) = 1.31%
Frequency 2 (F2) = 0.92%
Frequency 1 (F1) = 0.93%
Monetary 5 (M5) = 1.62%
Monetary 4 (M4) = 1.45%
Monetary 3 (M3) = 1.20%
Monetary 2 (M2) = 1.21%
Monetary 1 (M1) = 1.23%