Cohort & RFM Tracking

Cohort analysis groups customers by a shared trait, usually the month they first bought, and tracks how each group behaves over time. RFM analysis scores each customer on three things: how recently they bought (Recency), how often they buy (Frequency), and how much they spend (Monetary). Together they show which customers are healthy, which are slipping, and where marketing spend will do the most.

How it works

Averages hide trends. A steady overall churn rate can hide one monthly cohort that retains well and another that drops off fast, which points to a product change or a weaker acquisition source in that period. RFM turns the current customer base into groups you can act on:

• Champions (high R, F, and M): reward them.
• At Risk (used to buy often and spend well, but not recently): win them back.
• New Customers (recent, low F and M): onboard them.

Cohorts give you the view over time. RFM gives you a snapshot today.

How to do it

1. Group customers by acquisition week or month. For each cohort, track retention rate, cumulative revenue, and order count in each later period. Show it as a retention heat map.
2. Score every customer on days since last purchase (R), number of purchases (F), and total or average spend (M). Rank each on a 1 to 5 scale by quintile.
3. Define segments from the scores: Champions, Loyal, Potential Loyalists, At Risk, Hibernating, and Lost.
4. Set up an automated campaign for each segment: rewards for Champions, win-back offers for At Risk, onboarding for New Customers, reactivation for Hibernating.
5. Break cohorts down by acquisition channel. If one channel’s customers keep landing in low RFM segments, that channel brings in weaker customers, whatever its volume.

What to measure

• Cohort retention curve: retention rate by period for each cohort, compared across cohorts.
• RFM segment distribution: share of customers in each segment, tracked monthly.
• Lifetime value by cohort and segment: expected revenue per group, used to set budget.

Best practices

• Refresh cohort and RFM reports daily or weekly. Stale segments mean you contact At Risk customers after they have already left.
• Use segments to change messages, offers, and channels. They are not just report labels.
• Track the first 90 days of each cohort weekly to catch retention problems early.

Common pitfalls

• Running RFM on a customer base too small or too new to give segments of useful size.
• Treating segments as fixed. Customers move between them, so trigger campaigns on those moves.
• Leaving acquisition channel out of cohort analysis. Without it you cannot tell a product problem from an acquisition problem.