At a glance
Retention Curve (Nday) is the share of a cohort that returns and performs the retention event on each specific day after first use, plotted as a curve over time in Amplitude. Its shape tells you how durable engagement is: a curve that flattens means a sticky habit has formed, while one that keeps falling means users churn out. The card pairs with sibling metrics in the Analytics category to build a complete diagnostic picture; cross-reference the related cards listed below for context.
| What it counts | The percentage of a cohort returning to perform the retention event on each day N after first activity, using Amplitude’s N-day retention method, shown as a curve. |
| Sample type | Backend API data from Amplitude, refreshed on the standard data refresh. |
| Why it matters | The curve reveals whether engagement is durable. A flattening tail is the signature of a sticky experience; a curve that decays to near zero warns of weak repeat behaviour. |
| Reading the value | Read the shape, not just one point. Compare where the curve flattens and how high the tail sits against prior cohorts to judge whether retention is improving. |
| Currency | percent |
| Time window | 90D |
| Alert trigger | - |
| Sentiment key | amp_retention_curve |
| Roles | owner, marketing |
Calculation
Calculated automatically from your Amplitude data. Amplitude’s N-day retention groups users into cohorts by their first event, then for each subsequent day measures the percentage who returned and performed the retention event on exactly that day. Plotting those percentages across days produces the retention curve. This card surfaces that curve over a 90-day span so the long-run tail is visible. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.Worked example
A representative reading of Retention Curve (Nday) for a typical merchant on Amplitude. A healthy curve might start at 100 percent on day zero, fall to 32 percent on day 1, 14 percent on day 7, and settle around 8 percent by day 30 where it flattens. The flattening tail is the encouraging part: it means a core of users keeps coming back rather than the whole cohort decaying away. If a newer cohort’s tail sits at 5 percent instead of 8 percent, durable engagement is weakening. Use Vortex Mind to trace which acquisition source or onboarding change shifted the curve; for natural-language exploration, ask Ask Viq to compare the curve for new versus returning buyers.Sibling cards merchants should reference together
| Card | Why merchants reach for it |
|---|---|
amp_retention_d1 | The first point on the curve, the strongest early-onboarding signal. |
amp_retention_d7 | The canonical mid-curve stickiness gauge for product health. |
amp_retention_d30 | The long-run tail point that shows whether a durable habit formed. |
amp_stickiness_dau_mau | A complementary ratio of how often active users return within a month. |
amp_returning_users | The absolute count of repeat users behind the retention percentages. |
Reconciling against Amplitude
Where to look in Amplitude’s own dashboard: Open the Retention chart, set the method to N-day, and choose the start and return events that match the Vortex IQ profile. Confirm the cohort date range and any segment filters align. Event Segmentation can verify the underlying return-event volume, and the Dashboards section may pin a saved retention chart for reference. Why the Vortex IQ value may legitimately differ:| Reason | Direction | What to do |
|---|---|---|
| Period boundary. Vortex IQ spans a 90-day window by default; an Amplitude retention chart may use a different cohort range. | Variable | Match the cohort date range in the Amplitude retention chart. |
| Time zone. Amplitude buckets days by the project time zone; Vortex IQ aligns to the merchant reporting time zone. | Marginal | Confirm the project time zone matches your reporting time zone. |
| Filter or segment scope. Profile-level filters (platform, B2B, test users) change which cohorts feed the curve. | Variable | Apply the same segment to the Amplitude retention chart. |