High RFM Customers Are Not Always High-Value Customers

Why High RFM Scores Can Look Convincing

A high RFM score can look very convincing.

1. The customer purchased recently.
2. They often buy.
3. They spend more than others.

In many retail organizations, the customer quickly becomes a priority. They get pulled into campaigns. They receive richer offers. They may be treated as part of the “best customer” group.  Sometimes that is exactly right.  But not always.

The Limit of Behavior-Only Segmentation

The issue is not that RFM is wrong. The issue is that RFM is incomplete when the business is trying to understand value.  RFM measures visible customer behavior. It does not fully explain customer economics.

1. A customer can buy often and still be expensive.
2. A customer can spend heavily and still be margin-dilutive.
3. A customer can look loyal and still be loyal mainly to discounts.
4. A customer can show up repeatedly in the dashboard while quietly weakening contribution.

That is why I am careful with any customer strategy that stops at RFM.  Retailers need to know who is active. But they also need to know whether that activity is good for the business.  Those are not the same questions.

A Simple Retail Example

A simple example makes the issue clear. Customer A buys often, spends heavily, and responds quickly to promotions. In an RFM model, this customer may look excellent.  Once the business looks deeper, the picture changes.

Customer A buys mostly during markdowns, returns a meaningful share of purchases, and rarely expands into categories with healthier economics.

Customer B buys less often and spends less today. But the customer buys at healthier margins, returns less, and shows early signs of category movement.

A narrow RFM view may over-prioritize Customer A.  A stronger CLV view may tell the business to manage Customer A carefully and develop Customer B more intentionally.  That is the difference between behavior and value quality.

Using CLV to Test Customer Quality

RFM helps answer the question: What did the customer do?  CLV helps answer: what is that relationship likely worth?

The distinction matters because customer investment has a cost. Every incentive, loyalty treatment, retention effort, and personalized offer carries some economic trade-off. Sometimes the cost is visible. Sometimes it sits inside the margin. Sometimes it shows up later as trained discount dependency.

That is where high RFM customers can become tricky.  The customer looks important because they are active. The team wants to keep them active. The business keeps offering them more. But if that activity depends on constant subsidy, the retailer may be protecting volume without improving value.  That is not loyalty. That is expensive, repetitive behavior.  

This is where CLV earns its place.  A useful CLV model should account for margin, discount dependence, return behavior, retention quality, and the likely future direction of the customer relationship. It should help the business separate strong customer activity from weak customer economics.

That is not an academic distinction. It changes decisions. It changes which customers deserve more investment. It changes which offers should be used. It changes how loyalty should be measured. It changes whether a segment should be grown, protected, corrected, or allowed to cool.

The practical rule is simple: use RFM to identify behavior. Use CLV to test whether that behavior creates durable value.  Retailers need both because high activity is not always high value. When the business confuses the two, it can end up funding the wrong customer behaviors.

Source note: Gartner, 2025 CMO Spend Survey; referenced in LinkedIn post only.