In short
A retention curve shows how many users are still around days, weeks and months after they start. Its shape tells you, at a glance, whether your product has a future.
A chart of the percentage of users still active N days after signup — a flattening curve ('smile curve') signals product-market fit; a curve heading to zero signals a leaky bucket.
Think of it like this
The leaky bucket
Fill a bucket with holes and water pours out fast at first. If the holes are only near the top, the level drops, then stops at some point. If they're at the bottom, it empties. You want a bucket that keeps some water.
Watch it happen
Two shapes, two futures
Curve to zero
- Drops week after week
- Almost nobody left by month 3
- No product-market fit yet
Curve that flattens
- Drops early, then levels off
- A loyal group keeps using it
- A base you can grow on
Every curve drops at first. What matters is whether it flattens, and how high the flat part sits.
The big ideas
What you need to know
How to read it
Day 0 is 100%. Each point shows the share of that group still active on that day or week.
Look for the plateau
A flat tail means some users found lasting value. That's your core audience.
Raise the floor
Better onboarding often lifts the whole curve. The higher the plateau, the healthier the product.
Pick the right rhythm
Daily for chat apps, weekly for food delivery, monthly for tax filing. Match how often users need you.
Save yourself the pain
Common mistakes
Oops
Judging retention by day-1 numbers only.
Do this instead
Day 1 always looks scary. Watch whether the curve flattens over weeks.
Oops
Using a daily curve for a product people need once a month.
Do this instead
Measure in the natural usage rhythm, or a healthy product will look broken.
Your mission
Draw two curves
Quick check
Your retention curve drops to 30% by week 2, then stays at about 28% through week 12. What does this suggest?
Go deeper
- 1Sketch what a healthy vs unhealthy retention curve looks like.
- 2Find a public retention benchmark for an app category you're interested in.
- 3Learn why acquiring more users doesn't fix a bad retention curve — it just hides it temporarily.
More topics at this stop
Product Metrics Fundamentals
AARRR / Pirate Metrics
North Star Metric Design
Leading vs Lagging Indicators
Funnel Analysis
Cohort Analysis
SQL Basics for PMs
Analytics Tools (Amplitude, Mixpanel, GA4)
Building Dashboards
Statistical Significance Basics
Defining Success Metrics for a Feature
Stuck on this topic? Ask an Insider
Get 1:1 guidance from people who've walked this exact path — on the InsideEdge app.
