Growth Marketing Glossary

Cohort Analysis

co·hort a·nal·y·sis/ˈkoʊ.hɔːrt əˈnæl.ə.sɪs/noun

Grouping customers by when they started, then watching each group over time — the truth that blended averages hide.

periods since signup →
Schematic — cohort curves over time
Term
Cohort Analysis
Part of speech
Noun
Field
Measurement & Analytics
Also written
Cohort study

Forms & parts of speech

cohort analysis · noun
The technique of analyzing grouped customers over time.
"Cohort analysis showed our newest signups retained far worse than last year's."
cohort · noun
A group of customers sharing a start point.
"The January cohort still has 62% of its users active at day 90."

Definition in plain terms

Cohort analysis groups customers by a shared starting event — usually the period they signed up or first bought — and tracks how each group behaves over the weeks and months that follow. It answers questions a single blended number cannot: are the customers we acquire actually sticking, and is that getting better or worse over time?

The mechanics

You define the cohort (e.g. everyone who joined in January), pick a behavior to track (active use, repeat purchase, revenue), and measure it at each later period, producing a retention curve per cohort. Comparing cohorts before and after a product or onboarding change shows whether the change actually moved retention, rather than guessing from a noisy aggregate.

When it matters

Cohort analysis is how you see the truth a blended average conceals. A business can show flat overall retention while newer cohorts quietly retain worse — a problem the aggregate hides until growth stalls. Curves that flatten signal a sticky core and product-market fit; curves decaying to zero signal a leaky bucket no acquisition can fill.

Worked example. You launch a new onboarding flow in March. The February cohort retains 55% at day 60; the March cohort retains 68% at the same point. Because the only systematic difference is onboarding, the 13-point lift is strong evidence the change worked — a read no blended retention number could give you.
Failure modes to watch. Reading one blended retention figure instead of comparing cohorts; choosing cohorts too small to be statistically meaningful; and defining "active" so loosely that logins masquerade as genuine value-driven engagement.

Synonyms & antonyms

Synonyms

cohort studycohort trackingretention cohort analysis

Antonyms

blended analysisaggregate analysis

Origin & history

"Cohort" comes from the Latin cohors, a division of a Roman legion — a defined group moving together. Statistics borrowed the word for a set of people sharing a characteristic over time, and growth analytics adopted it for customers grouped by when they started.

Etymology: Online Etymology Dictionary.

Usage trends

Search interest for this term over the last five years:

View interest-over-time on Google Trends →

Common questions

What is a cohort in marketing?
A group of customers who share a starting point, such as the month they signed up, tracked together over time.
Why use cohort analysis instead of averages?
Because blended averages hide whether newer customers retain worse than older ones; cohorts reveal the real trend.
What does a flattening cohort curve mean?
A sticky core of customers found lasting value — a sign of product-market fit.

Related tools & calculators

Resources & people to follow

Curated, non-competitor resources verified per term.

Related training

Disciplines

Areas of marketing where cohort analysis is a core concern:

Sources

  1. trendsGoogle Trends — "cohort analysis"
  2. etymologyOnline Etymology Dictionary — "cohort"