How do you segment customers with RFM analysis?
RFM analysis scores customers on three numbers: days since their last order (recency), how many orders they've placed (frequency) and how much they've spent (monetary). Each dimension usually gets a score from 1 to 5, and score combinations become segments such as champions, loyal, at risk and dormant. Each segment then gets its own message, offer and channel.
The appeal of RFM is how simple it is. It needs nothing beyond order data and can run in a spreadsheet. Its weakness is the same simplicity: it doesn't see product category, margin or why a customer bought. Treat RFM as a starting point, not a strategy on its own.
What do the three letters in RFM measure?
- Recency: How many days since the customer's last order? Recent buyers are the most likely to buy again.
- Frequency: How many orders in a given period? Frequent buyers have made your brand a habit.
- Monetary: How much have they spent in total? High spend doesn't prove loyalty, but it does show value.
Shopify's RFM guide suggests scoring each dimension from 1 to 5 and adjusting the thresholds to your business. Klaviyo's glossary uses a simpler 1-3 scale with three-digit profiles such as 333 and 111. The right scale depends on your customer count. With a few thousand customers, 1-3 is enough; with tens of thousands, 1-5 gives finer separation.
How do you score RFM step by step?
- Pull the data. Export the last 12 or 24 months of orders with customer ID, order date and order value. Remove returns and cancellations.
- Summarise per customer. For each customer, calculate last order date, order count and total net spend.
- Set thresholds. Split customers into five groups on each dimension. The easiest way is to sort and cut into five equal-sized bands.
- Assign scores. For recency, the most recent date gets a 5; for frequency and monetary, the highest value gets a 5.
- Define segments. Group score combinations into meaningful segments like the ones below.
- Push segments into your email and SMS tool. Some tools, Klaviyo among them, build RFM analysis into the platform; if yours doesn't, upload the segments as lists.
- Refresh monthly. Customers move between segments. The most useful insight is who is moving up and who is slipping.
Which RFM segments work for ecommerce?
Segment names vary by tool. The table below shows a lean structure that works for most stores, with a suggested approach for each.
| Segment | Typical RFM score | Who they are | What to do |
|---|---|---|---|
| Champions | R 5, F 4-5, M 4-5 | Bought recently, buy often, spend the most | Early access, new products, perks without discounts |
| Loyal customers | R 3-5, F 4-5 | Buy regularly but not most recently | Loyalty program, cross-sell |
| New customers | R 5, F 1 | Just placed a first order | Welcome flow, nudge toward a second order |
| Potential loyalists | R 4-5, F 2-3 | Bought two or three times, habit forming | Complementary products, personal recommendations |
| At risk | R 1-2, F 4-5, M 4-5 | Used to be valuable, gone quiet | Personal win-back message, ask for feedback |
| Dormant | R 1, F 1-2 | Bought once or twice and vanished | Low-cost channel, one last attempt |
The segment where stores lose the most money is usually at risk. These customers already know and trust the brand, so winning them back is often cheaper than finding new ones.
What should you send each RFM segment?
A mistake we see often is sending the same discount code to the whole list. That means giving champions a discount they didn't need and burning margin. Champions buy without discounts; give them a sense of privilege instead. Save discounts for the segments you're about to lose.
An example setup: in a store with an average order of 500 TL, the at-risk segment gets a personal email followed by an SMS, while the dormant segment gets email only. SMS costs money and needs consent, so it goes to the most valuable segment. Our post on essential ecommerce email automations covers the core flows.
How do you use RFM in advertising?
RFM segments are useful well beyond email:
- Lookalike audiences: Upload your champions list to Meta and Google as a customer list and use it as a lookalike seed. Starting from your most valuable segment, rather than every customer, gives a stronger signal.
- Exclusions: Remove customers who bought in the last 30 days and have a longer repurchase cycle from prospecting campaigns.
- Win-back ads: Retargeting the at-risk segment with a small budget reaches customers who stopped opening your emails a while ago.
Before uploading customer lists, make sure your privacy notices and consent cover that use under the data protection rules that apply to you, such as KVKK in Turkey.
What are the limits of RFM analysis?
RFM doesn't show everything. Know its limits:
- It ignores purchase cycles. A sofa bought once a year and coffee bought monthly can't share one recency scale. Set thresholds per category.
- It ignores margin. Monetary measures revenue, not profit. Use net contribution where you can.
- It doesn't separate deal hunters. A customer who only buys during sales can still score high on frequency. Track the share of discounted orders separately.
- It doesn't predict. RFM looks backward; forecasting customer lifetime value needs additional models.
None of this makes RFM useless. It just means you should use it as a first filter, not as the final answer. For other ways to drive repeat orders, see our guide on how to increase repeat purchases.
Which tools can you use for RFM?
A spreadsheet is enough to start. As your customer base grows, built-in RFM features in email and CRM tools, or reports built on a data warehouse, become more practical. Shopify's guide makes the same point: spreadsheets get unwieldy at scale, so move to a tool with RFM scoring built in. Whatever you choose, the real gain comes when segments refresh automatically and plug straight into your automations.
Our smart automation service page explains how we connect RFM segments to email, SMS and ad automations. If you'd like to see which segments your own customer data produces, you can ask Performetic for a free growth audit.
Key takeaways
- RFM analysis scores customers on recency, frequency and spend, then groups them into segments.
- Choose a 1-3 or 1-5 scale and set thresholds to fit your customer count and purchase cycle.
- Give champions perks, not discounts; save discounts for the at-risk segment.
- Your champions segment makes a strong lookalike seed for ads.
- RFM doesn't see margin or purchase cycles, so use it as a first filter.
- Refresh segments monthly and watch how customers move between them.