What is cohort analysis?
Nicklas Segatz Mortensen · Growth Hacker · Fractional CMO · Meta Ads Nerd · 8 July 2026 · 4 min.
Definition
A cohort analysis divides customers into groups (cohorts) by a shared starting point — typically their first purchase month — and follows each group's behavior over time, e.g. retention, repeat purchases and accumulated CLTV.
Also called: Cohort analysis, Cohort Analysis, Cohort study
Sådan virker det
En kohorteanalyse grupperer kunder efter, hvornår de kom ind (fx købsmåned), og følger hver gruppe over tid. Det afslører, om nye kunder er mere eller mindre værd end tidligere — og om retention og CLTV forbedres eller forværres, hvilket et gennemsnit skjuler.
01Why averages lie
A single overall average blends new and old customers together and hides which way things are heading. A cohort analysis separates them: it shows whether customers you acquired in March are worth more or less than those from January — and whether your retention and CLTV are improving or deteriorating.
That's crucial when scaling. If you turn acquisition up and the new cohorts have worse retention, you're buying growth that looks fine on the top line but worsens the economics under the surface. Only the cohort view catches it in time.
02How to read a cohort table
A classic cohort table has cohorts in the rows (e.g. purchase month) and time since first purchase in the columns (month 0, 1, 2 …). Each cell shows a value — share of active customers, accumulated contribution margin or repeat purchase rate — so patterns emerge: how fast retention falls, when CAC is recouped, and whether newer cohorts are getting better.
Cohorts are also the foundation under an honest payback and CLTV calculation: instead of guessing the customer lifetime, you see the actual curve for when each group has earned back its acquisition.
Frequently asked questions
What is a cohort analysis used for?+
To see whether customer value, retention and repeat purchases are improving or deteriorating over time for new customers — something an overall average hides. It's central to honest CLTV and payback calculation and to judging whether scaling is degrading customer quality.
How are cohorts grouped?+
Most often by first purchase month, but they can also be formed by acquisition channel, campaign or product. The point is a shared starting point, so each group can be followed over time and compared.
Related terms
Glossary
What is retention rate?
Retention rate is the share of customers who are still active from one period to the next. A 68% retention means a good two-thirds of customers buy again.
Read the entry →Glossary
What is CLTV?
CLTV (Customer Lifetime Value) is the total gross profit an average customer contributes across their entire lifetime as a customer — from first purchase to last.
Read the entry →Glossary
What is payback period?
Payback period (CAC payback) is the time it takes before the profit a new customer generates has covered what it cost to acquire them. The shorter the payback, the faster capital can be reinvested in growth.
Read the entry →Glossary
What is RFM segmentation?
RFM segmentation scores each customer on three dimensions: Recency (how recently they bought), Frequency (how often) and Monetary (how much for). The scores divide the base into segments like VIPs, loyal, dormant and lost.
Read the entry →
Nicklas Segatz Mortensen
Growth Hacker · Fractional CMO · Meta Ads Nerd at Oaksmond
Growth hacker and fractional CMO with 10+ years' experience and hundreds of millions in managed ad spend behind him. Background from larger Danish and international scale-ups, and from the agency world.
Meet the team →