What is the RFM Model?

RFM Model is a customer segmentation model that allows businesses to understand their customers’ behavior and value based on their historical transactional data. 

Following are the three variables taken into account by Predictable Media to conduct the RFM segmentation:

Predictable Media taps into this data to derive the following three behavioral variables: 

Our Predictable Media RFM Model automatically understands and clusters the customers into quarters. For each of the three behavioral variables, the platform associates a user quartile. And these variables together contribute towards building the “User Score.”

Recency Frequency Monetary Value
Q4 [Most Recent] Q4 [Most Frequent] Q4 [Highest Spend]
Q3 Q3 Q3
Q2 Q2 Q2
Q1 [Least Recent] Q1 [Lest Frequent] Q1 [Lowest Spend]

The user score is calculated using the formula:

Final Score = R * 100 + F * 10 + M * 1

This gives us 4 x 4 x 4 = 64 possible combinations that will enter our decision tree.

RFM model by Predictable Media, decision tree.
Predictable Media RFM model

Our Clusters 

These 64 clusters are categorized into 10 groups with common characteristics: 


  1. Champions: These are, undoubtedly, your best customers. Not only do they spend more, but they purchase from you often. Moreover, they have also purchased from you recently. Their final score stands at 4-4-4.
  2. Lost: These customers are the ones who used to purchase from you frequently and usually spent a lot; however, they haven't bought from you recently. You are probably missing out on these important customers. 
  3. Big Spenders: This group of people spends a lot of money on your business; yet don’t buy from you frequently and haven’t purchased from you recently. 
  4. Loyals: These customers purchase from you frequently; however they don’t spend a lot.
  5. Recents: This group of people have purchased from you recently, but don’t have a big spending and frequency. 
  6.  Compulsive: This group of customers have spent a big amount on your brand, yet aren’t your best customers. They are generally expected to be discount-sensitive.
  7. Occasional: These customers are the ones with a good purchase frequency.
  8. Sleeping: These customers haven't purchased from you in a while. 
  9. Lost Cheap: These are the ones with a low frequency, recency and spending. a 
  10.  No category: This group of people comprise customers who can’t be associated with any of the above clusters. 

RFM Analysis

Predictable Media’s advanced algorithm automatically processes your customer data and categorizes your customers and adds them to the above-mentioned groups. Now, let’s learn how you can smartly apply the marketing tactics for each of these groups.


Champions: You can target the customers belonging to this group in several ways like

Lost: You are missing out on these customers and need to reconnect with them. You can do this by

Big spenders: These customers hold the potential to become your “Champions.” You need to push them a bit. You can reach out to them in several ways

Loyals: For this group of people, your mission should be to increase their average spending. You can do this by

Recents: You can turn this group of people into your “Loyals” by

Compulsive: Your goal with these customers should be to turn them into “Big Spenders.” This goal can be achieved by: 

Occasional: This group of people is just one step away from becoming the “Loyals.” And you can lead the way by

Sleeping: You are soon about to lose these customers. You can avoid missing out on them by

Lost Cheap: Like we all know, converting a new customer is way more expensive than retaining an old one. You can target this set of people by

Incorporate these marketing tactics into your existing marketing strategy to get the best possible results.