Overview
RFM stands for Recency, Frequency, and Monetary — three key indicators of customer purchasing behavior. Each component is scored on a scale of 1 to 5, where 5 represents the most favorable value (for example, most recent purchase, most frequent purchases, highest spend).
- Recency (R): How recently a customer made a purchase
- Frequency (F): How often the customer makes purchases
- Monetary (M): How much the customer has spent
You can view a customer’s RFM scores and segment directly on their profile under the RFM scores section.
Blueshift uses these individual scores to assign a combined RFM score and automatically categorize customers into one of 11 predefined segments, such as Champions, Loyal, or At Risk. These segments help marketers quickly identify engagement levels and customer value to drive more effective campaigns.
How RFM segmentation can be used
RFM segmentation helps marketers understand customer value and tailor outreach strategies based on behavior. Grouping users into segments like Champions, Potential Loyalists, or Hibernating can help optimize messaging, timing, and offers.
Typical use cases include:
- Loyalty campaigns for high-value customers
- Re-engagement efforts for customers who are slipping away
- Win-back campaigns targeting inactive customers
- Upsell and cross-sell strategies based on spending patterns
Recommended data requirement
RFM analysis is most effective when there’s at least one year of purchase data, as this provides enough behavior history to generate meaningful insights.
RFM score definitions
Each RFM score is calculated on a scale of 1 to 5, with higher values indicating stronger performance in that dimension.
- Recency (R): Measures how recently a customer made a purchase. A score of 5 indicates a very recent purchase, while 1 means it has been a long time.
- Frequency (F): Measures how often the customer has made purchases. A score of 5 indicates frequent purchases; 1 means very few.
- Monetary (M): Measures how much the customer has spent. A score of 5 indicates high spenders; 1 indicates low spenders.
These three scores are combined into a three-digit number to form the overall RFM score. For example, if a customer has R = 4, F = 3, and M = 2, their RFM score would be 432
.
These individual scores are used to determine a customer’s overall RFM segment.
RFM segments
Based on the combined RFM score, customers are grouped into one of 11 predefined segments. These segments help marketers tailor messaging and campaigns based on customer value and engagement.
Segment | RFM scores | Engagement pattern | Suggested strategy |
---|---|---|---|
Champions | 555, 554, 544, 545, 454, 455, 445 | Highly engaged and valuable customers | Reward with early access, premium offers, or loyalty perks |
Loyal | 543, 444, 435, 355, 354, 345, 344, 335 | Repeat buyers who trust your brand | Keep engaged with new arrivals or exclusive content |
Potential loyalists | 553, 551, 552, 541, 542, 533, 532, 531, 452, 451, 442, 441, 431, 453, 433, 432, 423, 353, 352, 351, 342, 341, 333, 323 | Recently active with potential to become loyal or champion customers | Nurture with personalized offers or incentives |
New customers | 512, 511, 422, 421, 412, 411, 311 | New buyers with low frequency | Engage with a welcome series and personalized recommendations |
Promising | 525, 524, 523, 522, 521, 515, 514, 513, 425, 424, 413, 414, 415, 315, 314, 313 | High recency, likely to become more engaged | Encourage repeat actions with targeted nudges |
Need attention | 535, 534, 443, 434, 343, 334, 325, 324 | Engagement is dropping | Re-engage with targeted messaging or discounts |
About to sleep | 331, 321, 312, 221, 213, 231, 241, 251 | Below-average recency and frequency | Send reminders or time-sensitive offers |
At risk | 255, 254, 245, 244, 253, 252, 243, 242, 235, 234, 225, 224, 153, 152, 145, 143, 142, 135, 134, 133, 125, 124 | Past high-frequency or high-value users, now inactive | Use win-back or reactivation campaigns |
Cannot lose them | 155, 154, 144, 214, 215, 115, 114, 113 | Once highly active, now at risk | Prioritize with exclusive re-engagement offers |
Hibernating customers | 332, 322, 233, 232, 223, 222, 132, 123, 122, 212, 211 | Low engagement across the board | Try reactivation with tailored messaging or offers |
Lost customers | 111, 112, 121, 131, 141, 151 | No recent activity or spend | Consider excluding from campaigns or minimal outreach |
Note: Customers who do not have an RFM score will be classified under a system-defined segment called Not rated. This typically includes users without relevant purchase activity within the configured lookback period.
Data requirements and availability
RFM scores are available only for accounts that track purchase events and have an associated revenue attribute. This ensures meaningful segmentation based on actual purchase behavior.
- The account must have at least one year’s worth of purchase data to generate accurate RFM scores.
- RFM is currently enabled by default for e-commerce and retail accounts that meet these conditions.
Accounts without purchase
events
- If your account uses events like subscription or donation instead of
purchase
, RFM scoring can only be enabled if those events include the standard revenue attribute. - Custom monetary fields or alternative revenue-like attributes are not currently supported.
- To enable RFM in such cases, please get in touch with your Customer Support Manager (CSM).
RFM attributes
The following customer-level attributes are available for RFM segmentation. These attributes are automatically computed and updated based on a customer’s purchase activity. They can be used across audience insights reports, journey trigger filters, personalization in templates, and the segment builder. For details on where RFM attributes appear in Blueshift, refer to the Visibility in Blueshift section.
Attribute | Description |
---|---|
rfm_segment |
The current RFM segment assigned to the customer (e.g., Champions, At Risk). |
rfm_recency_bin |
Score from 1 to 5 indicating how recently the customer made a purchase. |
rfm_frequency_bin |
Score from 1 to 5 indicating how frequently the customer purchases. |
rfm_monetary_bin |
Score from 1 to 5 indicating how much the customer spends. |
prev_rfm_segment |
The customer’s previously assigned RFM segment (e.g., Champions, At Risk - useful for comparison or tracking movement across segments). |
Visibility in Blueshift
RFM scores and segments are available across various areas in Blueshift to help you analyze and engage customers more effectively.
You can:
- View RFM scores and segments on the customer profile
- Use RFM attributes in the segment builder to create targeted audiences
- Reference RFM segments in journeys to personalize message flows
- Insert dynamic content using
user.derived_attributes.rfm_segment
in the template studio.
For example:{% if user.derived_attributes.rfm_segment == "Champions" %}
Thanks for being a top customer! Enjoy early access to our newest collection.
{% elsif user.derived_attributes.rfm_segment == "At risk" %}
We miss you! Here’s 20% off to bring you back.
{% else %}
Check out what’s new in store — curated just for you.
{% endif %}
- Analyze customer distribution by RFM segment in audience insights
- Access two out-of-the-box reports in the Reports section:
- Customers by RFM segment – Overview of customer distribution across RFM segments
- Revenue by RFM segment – Weekly order trends for each segment
FAQs and edge cases
How often are RFM scores refreshed?
RFM scores are refreshed every Monday at 12:00 AM UTC, based on purchase activity in the past year.
What is the default lookback period?
By default, RFM scores are computed using a 1-year lookback window for both goal and non-goal purchase events.
What events and attributes are used to compute RFM?
- Recency: Timestamp of last purchase
- Frequency: Count of purchases over the past year
- Monetary: Sum of purchase revenue over the past year
What if my event isn’t named “purchase” or the revenue field is named differently?
You can define an equivalent event and monetary attribute during enablement. For example, donation or subscription events with a revenue field. Please reach out to your Customer Support Manager (CSM) to check the data requirements and enable RFM scoring.
Can I view previous RFM scores?
Yes. The prev_rfm_segment
attribute stores the customer’s previous segment and can be used in segmentation.
Can RFM scores be edited or customized?
Currently, we don't support custom logic for computing RFM scores. Please log a feature request with your Customer Success Manager (CSM) if you require this functionality.
What if a customer has no recent purchases?
The customer won’t be assigned a new RFM score. Their most recent score is retained for reference.
Are there any out-of-the-box reports that use RFM segments?
Yes. A Blueshift-created Audience Insights report showing customer counts in each RFM segment is available out of the box. You can also create additional reports using the RFM attributes.
How are RFM scores interpreted?
RFM scores are represented as a 3-digit number. For example, a customer with a Recency score of 5, Frequency of 2, and Monetary of 1 will receive an RFM score of 521
. This corresponds to the Promising segment—indicating high recent activity but lower purchase frequency and value.
How is the "Monetary" score calculated?
Blueshift calculates the total revenue per customer from the past year using purchase events. Customers are ranked by this value in ascending order and divided into five equal groups. Each customer is then assigned a Monetary score from 1 (lowest revenue) to 5 (highest revenue).
Need help?
If you’re unsure whether your account is eligible for RFM scoring or need help enabling it, please reach out to your CSM or support@blueshift.com.
Comments
0 comments