THE SHORT ANSWER
Segmentation compares users or customers with shared characteristics such as channel, geography, device or product. Cohort analysis groups them by a shared starting event or period—such as first purchase month—so behaviour can be compared at equivalent maturity.
Groups and starting points answer different questions
| Method | Grouping | Example question |
|---|---|---|
| Segmentation | A characteristic or behaviour | Do new and returning visitors convert differently? |
| Cohort analysis | A shared start event and time | Do customers acquired in March retain better by month three? |
Choose segments that can change action
- New versus returning
- Acquisition source or campaign
- Geography or service region
- Device and experience
- Product or category
- Customer type or qualification
- Offer or promotion
A segment is useful when it represents a meaningful difference in customer context, economics or controllable experience. Splitting by every available dimension increases the chance of finding unstable patterns.
Align cohorts by maturity
Customers acquired last week have had less time to repeat than customers acquired six months ago. Compare week one with week one, month three with month three, and define the return behaviour that matters to the product.
Revenue, retention and contribution per cohort member answer different questions. State cancellations, refunds, identity rules and the horizon beside the result.
Evidence & context: Google Analytics Help · Google Analytics Help
Differences are clues, not causes
A channel cohort may retain better because of customer mix, offer, season, product or measurement—not necessarily because the channel created loyalty. Use the pattern to form a hypothesis and seek evidence that distinguishes explanations.
Connect cohort economics to E-commerce Retention and LTV and use experimentation when a change can be tested credibly.
Sources & further reading
- Get started with Explorations
Google Analytics Help. Official guidance for funnel, cohort, path, segment and lifetime explorations. Technique availability does not establish that an observed pattern is causal.
- Understand user metrics
Google Analytics Help. Official definitions for total, active, new and returning users. Identity limits and configuration can affect interpretation.
- BigQuery Export user-data schema
Google Analytics Help. Documents observed lifetime revenue, purchases and sessions in one analytics system. Historical revenue is not the same as a forward-looking customer-profit model.
Examples and exercises are illustrative unless attributed to a source. No independent expert review is claimed.
A correction, a counterexample or an experience worth sharing?
Join the conversation ↗