Cohort Analysis
An analytical technique that groups users or customers into cohorts based on shared characteristics or experiences within a defined time period, then tracks their behaviour over time.
In-Depth Explanation
Cohort analysis divides customers or users into groups (cohorts) that share a common characteristic - most commonly, the time they first became a customer - and then tracks how each cohort's behaviour evolves over time. This reveals patterns that aggregate metrics would hide.
Types of cohorts:
- Time-based (acquisition) cohorts: Grouped by when they became a customer (month, quarter)
- Behavioural cohorts: Grouped by actions taken (first purchase category, feature used, channel)
- Demographic cohorts: Grouped by attributes (industry, size, location)
- Value cohorts: Grouped by initial purchase value or engagement level
What cohort analysis reveals:
- Retention trends: Are newer cohorts retaining better or worse than older ones?
- Revenue patterns: How does customer spending evolve over time?
- Engagement decay: When do customers typically disengage?
- Product changes impact: Did a product change affect subsequent cohort behaviour?
- Seasonal patterns: Do cohorts acquired at certain times perform differently?
How to create a cohort analysis:
- Define the cohort: Choose the grouping criteria (e.g., signup month)
- Define the metric: Choose what to measure (retention, revenue, usage)
- Define the time period: Set the observation window (weeks, months, quarters)
- Build the cohort table: Create a matrix showing each cohort's metric over time
- Visualise: Use heatmaps or line charts to identify patterns
- Analyse: Identify trends, anomalies, and actionable insights
Common cohort analysis pitfalls:
- Cohorts too small for meaningful analysis
- Not accounting for seasonality
- Confusing cohort effects with general trends
- Ignoring the composition of each cohort
Business Context
Cohort analysis reveals whether a business is genuinely improving customer outcomes over time, providing insights that aggregate metrics like overall retention rate can mask.
How Clever Ops Uses This
Clever Ops implements cohort analysis capabilities for Australian businesses, building automated cohort reports that track customer behaviour over time. We help clients understand how product changes, marketing strategies, and operational improvements are affecting customer outcomes for each new group of customers.
Example Use Case
"A subscription business uses cohort analysis to discover that customers acquired through referrals retain at 2x the rate of those acquired through paid advertising, leading to increased investment in the referral program."