- Klaviyo's RFM analysis report is part of Personalization (formerly Marketing Analytics) or Advanced KDP. It isn't in the standard marketing plan.
- It scores recency, frequency and monetary value from 1 to 3 and puts customers in six groups: Champions, Loyal, Recent, Needs attention, At risk and Inactive.
- You need at least 500 customers who've ordered, 180 days of order history and orders in the last 30 days before the report works.
- No add-on? Build RFM-style segments with Placed Order conditions and Historic CLV. It's more work, but it does the job.
- The point of RFM is smaller, sharper sends. A sports training tech brand's engaged send earned $0.19 per person versus $0.04 for a full-list blast.
This is for Shopify founders who keep sending every campaign to the whole list and want a simple way to decide who actually gets what.
What is RFM segmentation?
RFM is a simple way to rank customers. You look at three things about every customer:
- Recency: how long since their last order.
- Frequency: how many orders they've placed.
- Monetary value: how much they've spent.
Someone who bought last week, buys often and spends a lot is your best customer. Someone who bought once, two years ago, for a small amount is not. They shouldn't get the same email, the same offer or the same send frequency.
RFM gives you a consistent way to sort people instead of guessing.
Does Klaviyo have RFM analysis?
Yes. Klaviyo has an RFM analysis report. Per Klaviyo's help center, it shows how recently customers bought, how often, and how much they spent inside the model's time frame, then combines those to assign each profile to a customer group. You'll find it under Personalization, then Customer insights, then RFM analysis.
What plan you need. Klaviyo says RFM analysis requires Advanced KDP or Personalization, and that these "are not included in Klaviyo's standard marketing application." An active trial also gives access. Klaviyo's multiple-models article calls it "Personalization (formerly Marketing Analytics)," so if you've seen the old name, it's the same thing. Klaviyo's Personalization page says you need both Personalization and an email plan, and lists pricing starting at $100 a month for up to 13,500 active profiles.
What data you need. Klaviyo lists these requirements for the RFM report:
- At least 500 customers who have placed an order (customers, not total profiles).
- An ecommerce integration like Shopify, or placed orders sent through the API.
- At least 180 days of order history, with orders in the last 30 days.
- At least some customers with 3 or more orders.
Only Owners, Admins, Managers and Analysts can see the report.
How does Klaviyo score RFM?
Each of the three gets a score from 1 to 3. Per Klaviyo:
- Recency uses day limits. By default, a purchase in the last 180 days scores 3, 181 to 365 days scores 2, and more than 365 days scores 1.
- Frequency uses percentiles of your own customer data by default. At or above the 66th percentile scores 3, at or above the 33rd scores 2, below that scores 1.
- Monetary value works the same way as frequency, but on spend.
You can change these thresholds per model, by value or by percentile. Klaviyo only counts orders with a positive value, so free orders don't help anyone's score. A profile with no purchase in the time frame doesn't get an RFM group at all.
One thing I'd change for most consumable brands: the default recency window. If your product runs out in 30 days, someone who last bought 170 days ago still scores a 3 on recency by default. That's not a recent customer for you. Klaviyo lets you edit those day limits, so set them to match how fast your product actually gets used.
What are Klaviyo's RFM groups?
Klaviyo turns the 27 possible score combinations into six groups. Scores are listed in R, F, M order:
| Group | Score combinations | Klaviyo's description |
|---|---|---|
| Champions | 333, 332, 323 | Your best customers |
| Loyal | 321, 322, 331, 232, 233 | Valuable customers that are engaged |
| Recent | 312, 313, 311, 222, 223 | Recent customers |
| Needs attention | 213, 221, 123, 132, 133 | A valuable customer that has not purchased recently |
| At risk | 231, 212, 122, 131, 211 | Hasn't purchased recently, low or medium spend |
| Inactive | 111, 112, 113, 121 | A lapsed customer |
Klaviyo writes three properties onto each profile: Current RFM group, Previous RFM group and RFM group last changed. Those properties refresh every night, so the dashboard and your segment counts can be a little off from each other for a few hours.
How do you build a segment from RFM groups?
From the report, click Create segment in the RFM segments card. Or build it yourself:
- Go to Audience, then Lists & segments, then Create, then Segment.
- Choose Properties about someone.
- Pick Current RFM group, set it to equals (or does not equal), and choose the group. Leave the type set to Text.
- Combine groups with OR. Klaviyo's own example is Current RFM group equals Inactive OR equals At risk.
The useful trick is RFM group last changed. Add it with AND, set the type to Date, and choose "is in the last 30 days." Now you've got people who just moved groups, like a customer who just slipped from Loyal to Needs attention. That's a great trigger for a segment-triggered flow.
For the full difference between segments and lists, read my Klaviyo lists vs segments post.
How do I build RFM segments without the add-on?
You don't need the report to think in RFM. You can build each piece with standard segment conditions. Klaviyo's segment conditions reference lists "What someone has done (or not done)" with count options like "At least once," "Zero times" and "Is at least," and timeframes like "Over all time" and "In the last."
Recency. Placed Order, at least once, in the last 60 days. Or flip it: Placed Order zero times in the last 60 days, AND Placed Order at least once over all time. That second one is your lapsing customers.
Frequency. Placed Order, is at least 3, over all time. That's your repeat buyers. Placed Order equals 1 over all time is your one-time buyers.
Monetary value. Choose Predictive analytics about someone, then Historic Customer Lifetime Value, and set a comparison. The catch: CLV conditions need the same data minimums as predictive analytics, including 500 customers who've ordered. Below that, you can filter Placed Order by its Value property to find big single orders. My Klaviyo predictive analytics post covers what CLV can and can't tell you.
Here's a hypothetical example of how I'd stack them. A "best customers" segment: Placed Order is at least 3 over all time, AND Placed Order at least once in the last 60 days, AND Historic CLV is at least your top-spender number. That's a manual Champions group. Pair it with what to send in my VIP email segment post.
How should you use RFM groups in your emails?
Building the segment is easy. Using it is what makes money.
Shrink your campaign send lists. In the second 90 days of working with a brand, I condense campaign send lists: fewer people, more revenue, more qualified leads. RFM is one clean way to decide who stays in. The results are why. A sports training tech brand's engaged send to 7,145 people earned $0.19 per person. Its blast to 49,475 earned $0.04 per person. A flavored diffuser brand tightened its audiences and open rate went from 22% to 44%.
Use a recency window in busy months. For December, an engaged window of "clicked or bought in the last 60 to 90 days" is one I'd use. It keeps sends tight and still pulls back customers who've lapsed a bit.
Give lapsing customers something to read. For replenishable products, I email customers with no order in 60 days and bring them into an educational cycle. That's basically a manual At risk segment, and it doesn't need a model.
Match the discount to the group. Klaviyo's over-discounting guide says Champions and Loyal customers "don't always require discounts" and suggests lowering or removing them there, with bigger offers for At risk and Needs attention. Klaviyo's campaign guide also suggests asking Loyal and Champions customers for reviews and offering Recent customers a subscription to drive a repeat order.
Add your own tags. RFM doesn't know why someone bought. Before the holidays, I tag likely gift buyers by order value or quantity. More data is better. A big order in December from a one-time buyer might be a gift, not a new Champion.
Test it on revenue per recipient. When you test a segment against your normal list, judge it on revenue per recipient, not opens. More on that in revenue per recipient.
If you want the bigger picture on which segments to build first, start with email segmentation for ecommerce.
Is Klaviyo's RFM add-on worth paying for?
It depends on your size. If you're under 500 customers who've ordered, the report won't run anyway, so build the manual segments above. If you're well past that, the add-on saves you building and maintaining a pile of segments, gives you the group-change property for flows, and lets you run up to 4 models per account, per Klaviyo. One caution from Klaviyo's own docs: models can't be deleted or deactivated yet, so think before you create extras.
My take: the report is a nice shortcut, not a strategy. The win comes from actually sending different things to different groups. A brand with three manual segments that it uses every week beats a brand with RFM groups it never touches.
FAQ
What does RFM stand for in Klaviyo?
Recency, frequency and monetary value. How recently someone bought, how often, and how much they've spent.
Is RFM analysis included in every Klaviyo plan?
No. Klaviyo says it requires Personalization (formerly Marketing Analytics) or Advanced KDP, which aren't part of the standard marketing application.
What are the RFM groups in Klaviyo?
Champions, Loyal, Recent, Needs attention, At risk and Inactive. Customers with no purchase in the model's time frame don't get a group.
Why can't I see the RFM report?
Check your plan first, then the data minimums: 500 customers who've ordered, 180 days of order history, orders in the last 30 days, and some customers with 3 or more orders. Your user role also has to be Owner, Admin, Manager or Analyst.
How often do RFM groups update?
Klaviyo says RFM profile properties refresh nightly, and within about 2 hours after you change model settings.
Can I do RFM segmentation in Klaviyo without the add-on?
Yes. Use Placed Order count and timeframe conditions for recency and frequency, and Historic CLV (if you meet the data minimums) or Placed Order value for monetary value.
- Getting started with the recency, frequency, and monetary (RFM) analysis report, Klaviyo Help Center
- Understanding scoring and customer groups in the recency, frequency, and monetary analysis (RFM) report, Klaviyo Help Center
- How to build a segment using RFM properties, Klaviyo Help Center
- How to strategically use RFM properties in campaigns and flows, Klaviyo Help Center
- How to avoid over-discounting using RFM groups, Klaviyo Help Center
- How to create multiple RFM models for different customer cohorts, Klaviyo Help Center
- Getting started with Personalization, Klaviyo Help Center
- Segment conditions reference, Klaviyo Help Center
- How to segment by customer lifetime value (CLV), Klaviyo Help Center
- Getting started with segments, Klaviyo Help Center