RFM Analysis
RFM, which stands for recency, frequency, and monetary value, is a long-standing analytical method that helps analyze and segment customer behavior based on the recency of their last purchase, the frequency of their purchases, and their monetary value, i.e., how much they spend with the firm.
By helping firms understand the purchasing behavior of acquired customers, RFM analysis can help increase retention and purchase per customer, identify which customers are not so great, better, and best, whether we are experiencing issues with a specific persona in terms of repurchase behavior, and so on.
To conduct an RFM analysis, a firm starts with its customer database. The first step is to assign value to customers associated with their recency, frequency, and monetary value. Since RFM analyses can be done by operationalizing these variables differently, let’s assume here that recency refers to the recency of the last purchase in days, frequency to the frequency of purchases over three months (or a quarter), and monetary value to the total amount spent during this period.
Firms will often start by indicating the exact number associated with each variable and move to create categories for each. For example:
| Recency | Frequency | Monetary value | |
| 1 | Very recent | Very frequent | High value |
| 2 | Recent | Frequent | Medium value |
| 3 | Not recent | Infrequent | Low value |
They will perform their analysis with these categories.
We can then create segments by combining these together. The following table shows examples of such segments, where “x” stands for any number (i.e., its value is not important for defining that segment).
| Segment | Recency | Frequency | Monetary value |
| Best customers | 1 | 1 | 1 |
| Loyal customers | x | 1 | x |
| Big spenders | x | x | 1 |
| Lost or almost lost customers | 3 | 1 | 1 |
| Thrifters | 3 | 3 | 1 |
Then, each customer will be coded based on the categories created above, as shown in the following table.
| Customer | Recency | Frequency | Monetary value |
| Jack | 3 | 3 | 1 |
| Jill | 1 | 1 | 1 |
| Bill | 3 | 1 | 2 |
| Sean | 3 | 1 | 3 |
| Raymond | 2 | 2 | 2 |
| Tom | 1 | 1 | 1 |
| Tina | 3 | 3 | 1 |
| Mariah | 2 | 3 | 2 |
| Sanjit | 1 | 3 | 3 |
| Todd | 1 | 2 | 3 |
| Becky | 1 | 1 | 1 |
| Seth | 2 | 3 | 2 |
| Caroline | 3 | 2 | 1 |
This allows firms to categorize customers into the categories just created (e.g., best customers, loyal customers, etc.). These categories of customers can help decide which segments to concentrate on and what kind of strategy to use to engage customers. Examples could include performing retention campaigns with big spenders, recuperating almost lost customers, or moving loyal customers to increase their monetary value over time. A firm could also target its best customer segment: send an appreciation letter, analyze their personal preferences for more personalized offers, or generally develop strategies to keep this segment highly satisfied.
Although simple, RFM analysis is a useful tool to foster engagement. A more thorough analysis could combine RFM with personas and evaluate whether personas also share commonalities or differences in their purchasing behaviors, leading to the creation of even more personalized campaigns.