--- id: "dashboard-and-metrics/charts/trial-cancellation-chart" title: "Trial Cancellation Rate Chart" description: "The Trial Cancellation Rate chart measures the number of trials that started in a given period, and the portion of them that canceled within a selected timeframe of their trial start, so that you can understand how often customers opt out during the trial and how that changes over time and by various dimensions." permalink: "/docs/dashboard-and-metrics/charts/trial-cancellation-chart" slug: "trial-cancellation-chart" version: "current" original_source: "docs/dashboard-and-metrics/charts/trial-cancellation-chart.md" --- > **AI agents:** This is the Markdown version of a RevenueCat documentation page. For the complete documentation index, see [llms.txt](https://www.revenuecat.com/docs/llms.txt). The Trial Cancellation Rate chart measures the number of trials that started in a given period, and the portion of them that canceled within a selected timeframe of their trial start, so that you can understand how often customers opt out during the trial and how that changes over time and by various dimensions. A trial is considered *canceled* when it expired without converting to paid and the customer turned off auto-renewal at some point during the trial (while it still granted access). Cancellations are timed by the customer's **last** in-trial opt-out, so a customer who cancels, re-enables, and cancels again is measured by their final cancellation. Customers who opt out but later convert are **not** counted as cancellations. ### Available settings - Filters: Yes - Segments: Yes - Cancellation Timeframe: Yes ### Cancellation timeframe The Cancellation Timeframe selector sets the maximum amount of time each period's cohort has to cancel within, so that historical periods stay comparable. The available options are 1, 2, 5, and 7 days (the default), plus Unbounded. Each "N days" option is an N×24-hour window measured from the trial's start timestamp, not calendar days, and a trial counts as canceled within the timeframe when its last in-trial opt-out falls inside that window. Unbounded counts trials that opted out at any point during the trial. ### Trial cohorts This chart is cohorted by a trial's start date. As a result, all dates which contain trials that have not yet ended will have an incomplete cancellation rate, since those trials may still cancel but have not yet had the opportunity to. :::info Incomplete period styling Support for incomplete period styling on this chart will come in the future. ::: ## How to use Trial Cancellation Rate in your business Tracking when customers cancel during the trial helps you spot early voluntary churn — customers who decide not to continue before they are ever charged. A spike in cancellations within the first day or two of the trial, for example, can signal an onboarding, expectation, or pricing problem. This chart can be segmented by product dimensions like Product Duration and Offering to see which purchase types drive the most trial cancellations; or by customer dimensions like Apple Search Ads Keyword, to understand how a customer's acquisition source influences whether they cancel. Pair it with the [Trial Conversion Rate](https://www.revenuecat.com/docs/dashboard-and-metrics/charts/trial-conversion-rate-chart) chart to see the full picture of how trials resolve. ## Calculation For each period, we measure: 1. Trial Starts: The count of trials started in the period. 2. Trial Cancellations: Trials that expired without converting and whose last in-trial opt-out happened within the selected Cancellation Timeframe. 3. Trial Billing Failures: Trials that expired without converting where the payment was in a billing-retry or grace period at expiry. 4. Trial Elapsed: Trials that ended without converting and without an observed opt-out or billing-failure signal. This includes lapsed trials (see below) and trials that are past their end date but have not yet recorded a terminal expired event. Cancellation, billing failure, and elapsed are the three reasons a trial can end without converting, so together they account for every non-converting trial: ``` Trial Cancellations + Trial Billing Failures + Trial Elapsed = trials that started in the period and did not convert ``` ### Formulas 1. \[Trial Cancellations] / \[Trial Starts] = Cancellation Rate ## Lapsed trials For some trials, RevenueCat never receives an event that records a voluntary cancellation, even though it correctly reconstructs the trial's final outcome. We refer to these as **lapsed trials**: trials that ended without auto-renewing, but for which there is no opt-out signal to attribute them to a cancellation. This happens when no opt-out is observed during the trial — for example, when RevenueCat only ingests the underlying receipt at or after the point of cancellation. In these cases the trial's final state (expired) is correct, but the intermediate opt-out step is lost. Lapsed trials are not dropped from the chart: because they expired without converting and without an observed opt-out, they are surfaced as **Trial Elapsed**. This keeps the breakdown complete — every non-converting trial lands in exactly one of Trial Cancellations, Trial Billing Failures, or Trial Elapsed, which together reconcile with the trials that did not convert in the [Trial Conversion](https://www.revenuecat.com/docs/dashboard-and-metrics/charts/trial-conversion-chart) and [Trial Conversion Rate](https://www.revenuecat.com/docs/dashboard-and-metrics/charts/trial-conversion-rate-chart) charts. ## FAQs | Question | Answer | | ------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | What is the relationship between Trial Cancellation Rate and other conversion charts? | Learn more about the relationship between conversion charts [here](https://www.revenuecat.com/docs/dashboard-and-metrics/charts#understanding-conversion-rates). | | How does this chart relate to the trials that didn't convert in the Trial Conversion charts? | Every non-converting trial is attributed to exactly one reason: Trial Cancellations, Trial Billing Failures, or Trial Elapsed. Together these reconcile with the trials that did not convert in the Trial Conversion and Trial Conversion Rate charts. | | If a customer cancels, re-enables auto-renewal, and cancels again during the trial, is it counted? | Yes. A cancellation is timed by the customer's last in-trial opt-out, so this trial is counted by its final cancellation. A customer who opts out but later converts is not counted as a cancellation. | | Are cancellation timeframes measured in calendar days or hours? | Each "N days" timeframe is an N×24-hour window measured from the trial's start timestamp, not calendar days. | | What are lapsed trials, and where do they appear? | Lapsed trials are trials that expired without converting and without an observed opt-out signal. They are surfaced in the Trial Elapsed measure. See [Lapsed trials](#lapsed-trials) above. | | Does the Trial Cancellation Rate chart support incomplete period styling? | Not at this time, though support for incomplete period styling in this chart will come in the future. |