RFM Segments

Prev Next

RFM stands for recency, frequency, and monetary value. This method analyzes a user's past transaction data, creates segment groups based on purchase history, and measures customer value by scoring customers based on how recently, how often, and how much they buy.

The available segments of RFM might vary depending on the product you are using.

Scoring model

  • Five levels of scores are available for Recency, Frequency, and Monetary values (scores 1–5).

  • Total possible combinations: 125 segments (5³)

Score interpretation

  • 5 = Highest ranking (e.g., most recent activity, highest frequency, highest spending)

  • 1 = Lowest ranking (e.g., long inactive, low frequency, low spending)

RFM group pages describe each score with a word. The following table shows which word matches each score:

Score

Word on RFM group pages

5

Highest

4

High

3

Medium

2

Low

1

Lowest

How users are distributed across RFM scores

RFM scores are relative rankings within the eligible user population. For each RFM dimension, eligible users are sorted by Recency, Frequency, or Monetary value and distributed across five score levels.

  • 5 represents the highest-ranking group.

  • 1 represents the lowest-ranking group.

  • Each score level contains approximately 20% of the eligible population.

Frequency is an exception to the approximately equal distribution. Frequency score 1 contains customers who purchased exactly once during the calculation period, so this group’s size depends on the share of one-time buyers. Customers who purchased more than once are distributed across Frequency scores 2–5.

RFM scores do not represent fixed spending, purchase count, or recency thresholds. The boundaries between score levels can change as the eligible population and its behavior change.

Definitions

  • Recency: Time since last purchase or activity with the brand.

  • Frequency: How often purchases or interactions occur with the brand during a particular period.

  • Monetary: How much a user spends during a defined period

Calculation basis

  • Ecommerce websites: RFM is calculated based on purchases.

  • Non-ecommerce websites: RFM is calculated based on sessions.

RFM segment groups

Insider One sorts users into 11 named RFM segments based on their Recency, Frequency, and Monetary scores. You can target any of these groups in a segment.

To choose the scores yourself, use a Custom RFM segment.

For more information, see RFM Segments: Custom.

The following table describes each group and its score level for each dimension:

Group

Who they are

Recency

Frequency

Monetary

Champions

Purchased very recently, shop frequently, and spend significant amounts

Highest

Highest

Highest

Loyal Customers

Purchased recently, buy frequently, and spend fair amounts

High

High

High

Potential Loyalists

Purchased recently, shop frequently, and spend average amounts

High

High

Medium

Promising

Purchased very recently, shop with average frequency, and spend fair amounts

Highest

Medium

High

New Customers

Recent first-time buyers who spent an average amount

High

Low

Medium

Need Attention

Purchased recently and spent fair amounts

High

Medium

High

About to Sleep

Have not purchased in a long time, shop occasionally, and spend low amounts

Low

Medium

Lowest

Cannot Lose Them

Have not purchased for a long time and spent significant amounts in the past

Low

Medium

High

At Risk

Last purchased a long time ago, shop frequently, and spend average amounts

Low

High

Medium

Hibernating

Have not purchased in a while, shop infrequently, and spend low amounts

Medium

Low

Low

Lost

Have not purchased for a long time, shop infrequently, and spend very little

Lowest

Low

Lowest

Score combinations in each group

Each group contains a set of Recency, Frequency, and Monetary score combinations. Each combination below is written as three digits in Recency, Frequency, and Monetary order. For example, 543 means Recency 5, Frequency 4, and Monetary 3.

Group

Score combinations

Champions

555, 554, 545, 544, 455, 454, 445

Loyal Customers

543, 444, 435, 355, 354, 345, 344, 335

Potential Loyalists

553, 552, 551, 542, 541, 533, 532, 531, 453, 452, 451, 442, 441, 433, 432, 431, 423, 353, 352, 351, 342, 341, 333, 323

Promising

525, 524, 523, 522, 521, 515, 514, 513, 425, 424, 415, 414, 413, 315, 314, 313

New Customers

512, 511, 422, 421, 412, 411, 311

Need Attention

535, 534, 443, 434, 343, 334, 325, 324

About to Sleep

331, 321, 312, 251, 241, 231, 221, 213

Cannot Lose Them

215, 214, 155, 154, 144, 115, 114, 113

At Risk

255, 254, 253, 252, 245, 244, 243, 242, 235, 234, 225, 224, 153, 152, 145, 143, 142, 135, 134, 133, 125, 124

Hibernating

332, 322, 251, 241, 233, 232, 231, 223, 222, 212, 211, 132, 123, 122

Lost

151, 141, 131, 121, 112, 111

Every user with an RFM score belongs to at least one group. Users with the score combinations 231, 241, and 251 belong to both About to Sleep and Hibernating, so they appear in both segments.

RFM segment requirements

  • Data used: Only purchase behavior is used for segmentation (app, web, CRM). Behavioral data, like session views or product interactions, is not considered.

  • Segment update frequency: Segments are updated daily. A customer’s segment can change daily based on their latest purchase activity.

  • Data span: Up to 12 months of purchase history is processed when calculating segments.

RFM segment sizes can change after the daily recalculation, even when you do not change the segment configuration. Users can move between RFM segments as their activity and relative scores change. Profile removal under the applicable User TTL can also affect segment counts.

Refer to Troubleshoot RFM Segment Changes for possible causes, investigation steps, and historical data limitations.

Use cases

RFM scores can help identify and engage a variety of customer segments:

  • High-spending new customers: Score high on recency and monetary, but low on frequency.

  • Frequent low spenders: Score high on frequency, but low on monetary value.

  • One-time big spenders: Purchased only once, but spent above average.

  • VIP customers: Score high across all three dimensions.

  • High-value churn risks: Score high on frequency and monetary value, but low on recency.

  • Low-value customers: Score low across recency, frequency, and monetary value.