RFM is commonly used in marketing, retail and professional services industries to assess customer value. The general idea behind the analysis can be summarized as
Recency : People who have purchased recently from you are much more likely to respond to a new offer than someone who you haven’t sold to in a long time.
Frequency : People who shop frequently at your store are more likely to respond to new offers than less frequent buyers.
Monetary : People who spend more money at your store are more likely to show interest in new offers.
in descending order of importance. There are a few different ways to calculate this metric but I will use the method outlined HERE with the sample Superstore database that comes with Tableau.
Quintiles needed for the analysis can be calculated using Tableau’s percentile rank function. For recency the formula would look like the following:
For frequency we can use Number of Records since each purchase is a record in the Superstore dataset or count OrderIDs.
And monetary is the simplest of all
Now let’s convert them to quintiles. I will just show the calculation for recency here but they’re the same for all of the above.
And finally combine the results into a single score to get the RFM metric :