What is RFM segmentation?
Used in a sentence: Before the sale, let's pull the RFM segments and give the lapsed big spenders a different offer from the regulars.
How it works
RFM starts from transaction events, not profile data. For each user, the analytics or engagement platform calculates three numbers over a chosen window: days since the last purchase, count of purchases, and total spend.
Each number is turned into a score, usually by ranking all users and splitting them into equal bands. A user who bought yesterday, buys often and spends a lot scores high on all three. Combinations of scores map to named segments such as champions, loyal regulars, new buyers, at risk and lapsed. Segments refresh on a schedule, so users move between them as their behaviour changes, and campaigns follow them automatically.
Example
A quick commerce app sees order volume dip among users who used to order every few days. The growth team builds RFM segments and finds a group with high frequency and spend but low recency. Instead of a blanket discount, these users see a bottom sheet on their next app open with a free delivery offer on their usual basket. Champions see a referral prompt instead, and new buyers get a nudge towards their second order.
RFM vs behavioural segmentation
| RFM segmentation | Behavioural segmentation | |
|---|---|---|
| Based on | Transactions only | Any tracked event or action |
| Answers | Who is valuable and who is slipping | What users do and where they get stuck |
| Best for | Retention, win-back and loyalty offers | Onboarding, feature adoption and journeys |
RFM is a focused subset of behavioural segmentation. Many teams use both: RFM for value, events for context.
Best practices
- Match the window to your purchase cycle. Grocery and gaming need short windows, travel and insurance need long ones.
- Name segments by the action you will take, not just the scores.
- Exclude users mid-payment or with an open support ticket from promotional campaigns.
- Track how many users move up a segment after a campaign, not only the click rate.
How to do this with Digia Engage
Segments usually live in the customer engagement platform. Digia reuses the segments and cohorts already defined in CleverTap, MoEngage or WebEngage, so a team can target an RFM segment without rebuilding it. They design a nudge or widget per segment in the no-code builder, set the trigger event and frequency limits, and publish without an app release. See how segments flow in on the CleverTap integration page.
Frequently asked questions
- How do you build RFM segments for a mobile app?
- Pick a transaction event, such as order placed or deposit made, and a lookback window. For each user, compute days since the last transaction, the number of transactions and total value. Score each dimension, for example into quintiles, then combine the scores and name the groups you will actually act on.
- Which in-app campaigns suit each RFM segment?
- High scorers on all three suit loyalty rewards, early access and referral asks. Frequent but low spenders suit bundles and upsells. Recent first-time buyers need onboarding to a second purchase. Lapsed high spenders deserve a win-back offer, while long-lapsed low spenders are often best left out to protect budget.
Related terms
Read more
Part of the In-app engagement glossary.