Not all customers are equal. A customer who bought yesterday, buys every month, and spends $200 per order is fundamentally different from someone who made a single $15 purchase nine months ago. Treating them the same -- same emails, same offers, same level of attention -- is one of the most common and costly mistakes in ecommerce.
RFM analysis is the fix. It's a segmentation framework that scores every customer on three dimensions: how recently they purchased, how often they purchase, and how much they spend. The result is a clear map of your customer base that tells you who to invest in, who to win back, and who to let go.
What RFM stands for
RFM breaks customer behavior into three measurable components:
Recency (R) -- How many days since the customer's last purchase. A customer who bought 3 days ago is more likely to buy again than one who bought 180 days ago. Recency is the strongest single predictor of future purchases.
Frequency (F) -- How many orders the customer has placed in a given period (typically the last 12 months). A customer with 8 orders is far more engaged than a customer with 1. Frequency indicates habit formation -- the behavior you want to reinforce.
Monetary (M) -- How much total revenue the customer has generated over that same period. This distinguishes your high-value buyers from your bargain-hunting one-timers.
Each dimension captures something different. A customer can be high-recency but low-frequency (they just made their first purchase). Another can be high-frequency but low-monetary (they buy often but only small items). RFM captures these nuances in a way that simple "top customers by revenue" lists cannot.
How to score customers
The standard approach is to divide each dimension into 5 equal segments (quintiles), scored 1 through 5:
| Score | Recency | Frequency | Monetary |
|---|---|---|---|
| 5 | Purchased in last 7 days | 10+ orders | $800+ total |
| 4 | 8-30 days | 6-9 orders | $400-799 |
| 3 | 31-90 days | 3-5 orders | $150-399 |
| 2 | 91-180 days | 2 orders | $50-149 |
| 1 | 180+ days | 1 order | Under $50 |
The exact thresholds depend on your business. A subscription coffee brand might set "high frequency" at 12+ orders per year, while a furniture store might consider 2 orders in 3 years as frequent. Calibrate the ranges to your purchase cycle.
Each customer gets a three-digit score. A customer scored 5-5-5 is your best: they bought recently, buy often, and spend a lot. A customer scored 1-1-1 is the opposite.
You don't need to track all 125 possible score combinations. Instead, map them into actionable segments.
The practical segments
Here are the segments that matter for most ecommerce merchants, along with the RFM score ranges and what to do about each one.
Champions (R: 5, F: 4-5, M: 4-5)
These are your best customers. They buy frequently, spend generously, and bought recently. In a typical store, Champions make up 5-10% of customers but generate 25-40% of revenue.
What to do: Protect them. Give them early access to new products. Offer a loyalty tier. Ask for reviews and referrals. Never send them a generic "we miss you" email -- they don't need convincing. The goal is to make them feel recognized, not marketed to.
Example: You have 340 Champions. Their average order value is $112 and they've placed 7.2 orders each on average. They generated $274,000 in revenue over the last 12 months. These customers are worth more attention than your entire bottom 50% combined.
Loyal Customers (R: 3-5, F: 3-5, M: 3-5)
Solid repeat buyers. They may not spend as much per order as Champions, but they come back consistently. This segment often represents the core of your recurring revenue.
What to do: Move them toward Champion status. Cross-sell higher-margin products. Offer bundle deals that increase AOV. If they tend to buy one category, introduce them to complementary products. A loyalty program with spend-based tiers works well here.
Recent Customers (R: 4-5, F: 1, M: 1-3)
New buyers. They just made their first (or second) purchase. You don't yet know if they'll come back. This is the most critical segment for retention because the drop-off between first and second purchase is where most customer relationships die.
What to do: Nail the post-purchase experience. Send a thank-you email. Follow up with a product recommendation based on what they bought. If they bought a consumable, time your follow-up to when they'd logically need a refill. The goal is to convert them from one-time buyers into repeat customers. A small incentive for the second purchase (free shipping, 10% off) can move the needle significantly.
At-Risk (R: 1-2, F: 3-5, M: 3-5)
These were loyal customers who have gone quiet. High frequency and monetary scores mean they used to buy a lot. Low recency means they've stopped. This is where you lose revenue if you don't act.
What to do: Win them back. Send a personalized reactivation email that acknowledges their history: "You've been one of our best customers -- we noticed you haven't ordered in a while." Pair it with an exclusive offer. If you've launched new products since their last purchase, highlight those. But act quickly -- the longer they stay dormant, the harder (and more expensive) reactivation becomes.
Example: You have 180 At-Risk customers with a combined historical spend of $156,000. If your reactivation campaign converts even 15% of them, that's 27 customers generating an estimated $12,000-18,000 in recovered revenue.
Lost (R: 1, F: 1-2, M: 1-2)
Low scores across the board. They bought once or twice a long time ago and haven't come back. This is typically the largest segment -- often 30-50% of your database.
What to do: Don't waste resources chasing them aggressively. Send one reactivation attempt. If there's no response, suppress them from regular campaigns. They inflate your email list size and drag down your open rates if you keep mailing them. Focus your budget on segments with higher return potential.
Can't Lose Them (R: 1-2, F: 4-5, M: 4-5)
The most alarming segment. These customers used to be Champions or Loyal, with high frequency and high spend -- but they haven't purchased recently. Something caused them to leave, and whatever it was, the revenue impact is significant.
What to do: This requires direct attention. Consider personal outreach -- a phone call or a direct email from a real person, not an automated flow. Find out what went wrong. Was it a bad experience? A competitor? A life change? The information is as valuable as the potential reactivation.
What actions map to which segments
| Segment | Size (typical) | Revenue share | Action |
|---|---|---|---|
| Champions | 5-10% | 25-40% | Reward, refer, early access |
| Loyal | 10-15% | 20-30% | Cross-sell, increase AOV |
| Recent | 15-25% | 5-10% | Onboard, second purchase push |
| At-Risk | 5-10% | 10-15% (historical) | Reactivation campaign |
| Lost | 30-50% | 2-5% | One attempt, then suppress |
| Can't Lose | 2-5% | 5-10% (historical) | Personal outreach |
How to build this without a data team
If you're running RFM manually, it's a spreadsheet exercise: export your order data, calculate the three values per customer, assign scores, and sort into segments. It works, but it's tedious and goes stale quickly.
Spark by MishiPay automates this. Connect your Shopify, WooCommerce, or Square store, and Spark calculates RFM scores across your customer base automatically. You can ask questions like "How many customers are At-Risk right now?" or "What's the revenue at stake in my Can't Lose segment?" and get answers with specific customer counts, revenue figures, and recommended actions.
The real value is in tracking segments over time. If your At-Risk segment is growing month over month, something is wrong with your retention -- even if top-line revenue looks healthy. If your Recent Customers segment is converting to Loyal at a higher rate than last quarter, whatever you changed in your post-purchase flow is working.
Common mistakes to avoid
Using revenue alone for segmentation. A customer who spent $500 in one order six months ago is not the same as one who spent $500 across 10 orders over the last 3 months. Single-dimension segmentation misses this entirely.
Setting static thresholds. Your RFM scoring boundaries should shift as your business grows. What counts as "high frequency" when you have 500 customers will be different when you have 50,000. Recalibrate quarterly.
Treating all segments with equal investment. The math is clear: a dollar spent retaining a Champion generates more return than a dollar spent reactivating a Lost customer. Allocate accordingly.
Running the analysis once. RFM is not a one-time project. Customers move between segments constantly. Run the analysis monthly at minimum, and build your marketing automation around segment transitions -- not just segment membership.
The bottom line
RFM analysis turns a flat customer list into a strategic asset. Instead of sending the same campaign to everyone, you send the right message to the right segment at the right time. Champions get rewarded. New customers get onboarded. At-Risk customers get reactivated before they become Lost.
The merchants who segment their customers outperform those who don't -- not because they have more customers, but because they know which customers matter most and act accordingly.
See your customer segments instantly
Spark calculates RFM scores automatically across your Shopify, WooCommerce, or Square store. Know who your Champions and At-Risk customers are today.