What is Retention Curves?
A retention curve plots the percentage of a cohort still active over time. The curve typically declines steeply in early periods (early churn — bad fit, onboarding failure) and gradually flattens to a long-term retention rate. The shape matters: a curve that flattens at 60% means 60% of customers stay long-term, supporting strong CLV; a curve that asymptotes to 0% means every customer eventually churns, severely limiting CLV. Brian Balfour's viral framework distinguishes smile curves (retention initially declines but reverses up — uncommon, signals product-market fit), flat curves (retention stabilizes — strong PMF), and declining curves (retention always declining — weak or no PMF). The retention curve is one of the most diagnostic metrics in product-market fit assessment.
How Retention Curves actually works
The framework breaks down into the following moving parts. Knowing what each piece is — and what it is not — is what separates a B-grade answer from an A-grade answer in a written assignment.
- Plot % retained vs time since acquisition
- Steep early decline = early churn problem
- Flat curve = strong PMF
- Smile curve = exceptional PMF (rare)
- Declining curve = no PMF
- Long-term retention rate determines CLV
A worked example: Facebook's 2012 retention
Facebook in 2012 reportedly had a "smile curve" for new users — initial decline, but users who stayed past 90 days actually engaged more over time as their friend network grew (network effects). The smile shape was operational evidence of network-effect-driven product-market fit and supported continued growth investment. By contrast, many startup retention curves never flatten — they just decline to zero, indicating poor PMF regardless of growth-rate vanity metrics. Investors increasingly demand to see retention curves before funding growth — the shape predicts long-term value far better than monthly active users.
Don't lose marks for these
- Reporting only short-term retention (curves can decline later)
- Not segmenting by cohort (curves shift over time)
- Confusing growth with retention
How to use this on the exam
Score-maximizing moves
- Distinguish smile, flat, declining shapes
- Cite long-term retention as CLV input
- Use to diagnose PMF
When to use Retention Curves (and when not to)
Use Retention Curves when your assignment asks you to analyze, structure, or recommend — and when you have at least two data points to populate every cell of the framework. Skip it when the question is asking for a numerical answer or a single recommendation, since Retention Curves is a structuring tool, not a calculator.