• 25 March 2021
  • 1 reply

Badge +5

For those who don't know what the RFM matrix is ​​and how to use it ->


The RFM matrix is ​​used to create custom audiences based on customer behavior with respect to the site (very useful for ecommerce).


The segments are created on 3 elements, Recency (R) the last time he made a purchase, Frequency (F) how many times he reiterated a purchase in the reference period, Monetary (M) the total value of all transactions of the customer.


These 3 elements are then numbered from 1 to 5, 5 being the highest / best grade.


In this matrix you will therefore find segmented customers of the group






Let's try to analyze 3 segments to understand who they are and how we can use them


445-> Customer who has purchased relatively recently, who has made many orders in the reference period and who is among the most profitable in the business.


We can reward them with exclusive offers or do activities to maintain and increase "brand loyalty" by thanking them personally.


253-> Customers who have not bought for some time but who have been very profitable in the past, somehow have not made purchases anymore. You have to do some activities to lead to a new purchase and potentially make him a "loyal" customer who will use the brand as a reference point for his needs in the future.


412-> He is a new customer who has recently purchased, he has yet to know the brand better and still has to be "used" to buying from our e-commerce.


At this moment we may not care so much about the margin on the single transaction but we are more interested in making him repeat various purchases to "educate" him.


My question is this: How can we segment users on Klaviyo on the basis of a scoring and assign them tags that will update from time to time?


Is there any type of guide that allows us to segment users based on this matrix?


To better understand and deepen:



1 reply

Userlevel 5
Badge +34

Hi @Vincenzo Cassese,


You have the ability within Klaviyo’s segment builder to identify these audiences based on their purchasing behavior. I’ve listed a few examples below:

  • Create 1 segment of users who have placed an order within the last x days (R)
  • Create 1 segment of users that have purchased item x at least x times (F) 
  • Create 1 segment of users that have revenue (spent) greater than x (M)
  • Create different segments that combines all these different definitions for varying RMF parameters.
  • To take things a step further, in theory, you would be able to set up segment triggered flows whose sole purpose would be to update a profile property for a user if they are added to a given segment. As an example, if a user places an order in the last 7 days, has purchased item x at least 3 times over all time, and has spent greater than $500 they will be added to a segment and once that happens tag their profile with a score of 3 or whatever scoring convention you would like to use. 

Here are a few help center articles that are somewhat related to what you were asking about:

I hope this helps!