> ## Documentation Index > Fetch the complete documentation index at: https://docs.vortexiq.ai/llms.txt > Use this file to discover all available pages before exploring further. # Retention Curve, Mixpanel > Retention Curve for Mixpanel stores. Tracked live in Vortex IQ Nerve Centre. How to read it, why it matters, and how to act on it. **Card class:** [Non-Hero](/nerve-centre/overview#card-classes-explained) • **Category:** [Retention](/nerve-centre/connectors#connectors-by-type) ## At a glance > **Retention Curve** plots the full day-by-day decay of a cohort, showing the percentage of users who come back on each subsequent day after they first acted. Where single-point cards such as D7 and D30 give you a snapshot, this curve shows the whole shape: the steep early drop in the first few days, the slope through the middle, and whether it flattens into a loyal base or keeps falling toward zero. The shape is what matters. A curve that plateaus means you have found a habitual core; a curve that never levels off means you are renting users rather than keeping them. | | | | --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **What it counts** | The percentage of a cohort that returns and performs the qualifying event on each day after first action, plotted across the retention window. | | **Sample type** | Backend API data from Mixpanel retention reports, computed per cohort over time. | | **Why it matters** | The curve's shape reveals product stickiness in a way a single day cannot. A flattening curve signals a loyal base; a continuous slide signals churn you need to address. | | **Reading the value** | Read it as a line over days. Steep early decay is normal; what matters is whether the curve flattens (good) or keeps falling (a churn problem). | | **Currency** | percent | | **Time window** | `90D` | | **Alert trigger** | `-` | | **Sentiment key** | `mix_retention_curve` | | **Roles** | owner, marketing | ## Calculation Vortex IQ reads Mixpanel's retention report and renders the return rate of a cohort for each day after their first qualifying action, across a 90-day window. A cohort is the set of users who first performed a defining event in a given period; retention on day N is the share of that cohort who returned and performed the qualifying return event on day N. The result is a line that starts at 100% on day zero and decays over time. Because it is a curve rather than a single point, it captures both the early drop-off and the long-run plateau, letting you judge the durability of engagement rather than just one milestone. ## Worked example *A representative reading of **Retention Curve** for a typical merchant on Mixpanel.* Imagine the cohort of shoppers who first purchased in the week of 14 Jun 26. By day 1 about 38% return, by day 7 roughly 19%, by day 30 around 11%, and from there the line flattens near 9% out to day 90. That flattening tail is the encouraging part: it means you have a loyal core that keeps coming back rather than a curve sliding to zero. When you later compare a cohort exposed to a new loyalty email flow, its curve plateaus closer to 14%, a clear lift in the durable base. For deeper investigation, use Vortex Mind to trace upstream causes; for natural-language exploration, ask Ask Viq. ## Sibling cards merchants should reference together | Card | Why merchants reach for it | | --------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------- | | [`mix_cohort_retention_d7`](/nerve-centre/kpi-cards/mixpanel/cohort-retention-d7) | The day-7 point on this curve, a fast read on early retention. | | [`mix_cohort_retention_d30`](/nerve-centre/kpi-cards/mixpanel/cohort-retention-d30) | The day-30 point, showing whether retention holds into the medium term. | | [`mix_worst_decaying_cohort`](/nerve-centre/kpi-cards/mixpanel/fastest-decaying-cohort) | Surfaces the cohort whose curve falls fastest, the one to investigate first. | | [`mix_stickiness`](/nerve-centre/kpi-cards/mixpanel/stickiness-daumau) | A complementary habit measure; a flat curve usually pairs with healthy stickiness. | | [`mix_returning_users`](/nerve-centre/kpi-cards/mixpanel/returning-users) | New versus returning split, the front line of whether retention is improving. | ## Reconciling against Mixpanel **Where to look in Mixpanel's own dashboard:** Open Mixpanel's Retention report, pick the same defining and return events the card uses, and set the range to 90 days. The retention table and its curve view should match the shape on the card, day for day. Confirm whether you are looking at unbounded retention (returned on or after day N) or bounded retention (returned exactly on day N), since the two produce different curves. **Why the Vortex IQ value may legitimately differ:** | Reason | Direction | What to do | | --------------------------------------------------------------------------------------------------------- | ---------------------------- | ----------------------------------------------------------------------- | | **Bounded versus unbounded retention.** Mixpanel offers both definitions, which produce different curves. | Curves diverge by definition | Match the retention type before comparing the two shapes. | | **Event selection.** A different qualifying return event reshapes the curve entirely. | Variable | Confirm the defining and return events match the card. | | **Immature cohorts.** Recent cohorts have not lived long enough to fill the later days. | Tail reads incomplete | Read the long tail only from cohorts old enough to have reached day 90. | **Cross-connector reconciliation:** if the retention curve flattens but your ecommerce platform shows few repeat orders, the return event being tracked may be lighter than a purchase, so align the events before drawing conclusions. For divergence investigations, use Vortex Mind. ## Known limitations / merchant FAQs **Q: How often does Retention Curve update?** It refreshes on the Nerve Centre's normal cadence over a 90-day window. Recent cohorts keep maturing, so the later days of the curve continue to fill in as time passes. **Q: Why does the most recent part of the curve look incomplete?** The newest cohorts simply have not aged enough to have a day-30 or day-90 figure yet. Read the full curve from cohorts that are old enough to have reached the end of the window. **Q: My curve keeps falling and never flattens. What does that mean?** A curve that slides toward zero without a plateau means you are not building a habitual base. Compare cohorts before and after product or lifecycle changes to find what nudges the tail upward. **Q: Can I customise the alert threshold?** This card is a curve with no default alert, but you can configure thresholds per profile in the Sensitivity tab, for example to be notified when a day-N point on the curve falls below a chosen level. *** ### Tracked live in Vortex IQ Nerve Centre *Retention Curve* is one of hundreds of KPI pulses Vortex IQ tracks across Mixpanel and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. [Start for free](https://app.vortexiq.ai/login) or [book a demo](https://www.vortexiq.ai/contact-us) to see this metric running on your own data.