How RFM scores customer behavior from 1 to 5

A customer was asking about how the RFM scoring in Repeat Customer Insights works:

How are your coming up with your recency, frequency, and monetary scores of 1-5?

How RFM scores the three components

All three of the RFM scores are based on five groups of 20% of your customer base (called quintiles).

• Recency is based on when they last ordered.
• Frequency is based on how many orders a customer has placed.
• Monetary is from how much they've spent over their lifetime.

Recency, Frequency, Monetary. R-F-M. RFM.

There are a few steps to the RFM analysis but the basics are:

1. Customers are sorted by each component (R, F, or M)
2. They are grouped into one of five groups
3. Their group determines their score for that component
4. Once all three components have been scored, the scores are combined into a single RFM score

Which group they are assigned to is the algorithm part. Basically the scores work like this:

• 5 - Top 20%
• 4 - Top 21-40%
• 3 - Middle 20%
• 2 - Bottom 21-40%
• 1 - Bottom 20%

So a score of Monetary 5 would mean those customers have spent the top 20% of your customers, while Monetary 1 would be the bottom 20%.

Sometimes the algorithm will use a modified RFM like what Repeat Customer Insights uses. This changes the behavior of the scoring but can be beneficial in certain cases e.g. ecommerce stores.

Reading a RFM score

A score of 515 would mean the customer was in:

• Group 5 for Recency
• Group 1 for Frequency
• Group 5 for Monetary

Similarly, a score of 331 would be:

• Group 3 for Recency
• Group 3 for Frequency
• Group 1 for Monetary

Each component is independently scored

Each of the components are scored individually so you can have a customer with a Recency of 5, Monetary of 5, but yet a Frequency of 1. They probably just ordered (Recency), spent a lot on their first order (Monetary), and that's their only order (Frequency).

Each component is independently evaluated too

Each of the components are also judged differently depending on what you're doing. A score of 511 isn't absolutely any better than a score of 151.

It's only better if you're evaluating the Recency (the first component). If you're looking at how often someone orders (Frequency) the 511 performs much worse than the 151.

Both of them are equally poor at how much they've spent (the last component).

In most cases, the Recency and Frequency components hold more power when it comes to customer loyalty and long-term retention. The Monetary component is more important with one-time customers.

(Interestingly, the Monetary component has some relationships with Average Order Value)

RFM scores tell a story of each customer

Interesting how RFM can create a story so easily for a customer right?

That's what makes it such an easy way to segment and market.

This RFM segmenting (and others too) are included in Repeat Customer Insights. It'll run all the algorithms and math for you automatically.

It comes with a 14-day free trial so you can see how it works before paying.

Eric Davis