Skip to content
GetProfit
Log in

← All terms

RFM segmentation

RFM segmentation is a way of grouping a store's customers by how recently they last ordered, how often they order and how much they have spent.

How it works

RFM is one method of customer segmentation. Each customer gets three numbers from the order history: recency, the days since the last order; frequency, the number of orders; monetary value, the total spent. Each number becomes a score from 1 to 5, based on the customer’s rank among the store’s own customers. In Shopify, 5 marks the top 20% and 1 the bottom 20%.

Customers with similar scores form segments: Shopify names eleven, among them Champions, At risk and Dormant. The store chooses which period of orders to score. Each segment then gets its own action, such as win-back messages for lapsing customers or a Customer Match list of the best customers for ads.

A table built from an order export, one row per customer, is enough: see RFM analysis in a spreadsheet. Shopify’s guide warns that spreadsheets get error-prone with thousands of customers.

Ways to calculate

Example

Example store, not client data.

The tableware shop scores its 2,700 customers from the last 12 months; 2,100 of them ordered once. A fifth of 2,700 is 540 customers. One customer last ordered 12 days ago, placed 4 orders and spent 2,400; on each measure, they rank in the top 540. The code is 555; as a total, 5 + 5 + 5 = 15.

Not to be confused with

  • Cohort analysis — groups customers by when they first ordered. RFM groups them by how they buy.
  • Zombie products — products that spend ad budget without sales, labelled Dormant in the GetProfit portal. Shopify’s Dormant segment is a group of customers.

Right and wrong readings

  • Wrong: “Each frequency score holds exactly a fifth of the customers.” Right: 2,100 of the 2,700 customers (77.8%) have one order. They tie on frequency and cannot be split into fifths by order count, so the score depends on how ties are broken.

Sources