Trial conversion rate is the trials that became a first payment divided by the trials that started in the same period. A trial start is not a purchase and carries no revenue. A trial that has not ended yet is neither a win nor a loss.
Trial conversion rate formula
Converted trials ÷ trials started in the period. Count it per trial cohort: group by the date the trial started, not by the date the payment arrived.
| Denominator | What it answers | When to use it |
|---|
| All trial starts | Share of the cohort that has paid so far | The default. It rises until every trial resolves |
| Resolved trials (converted + churned) | Conversion among trials that already ended | A cohort that is still open |
| Installs | Trial start rate times trial conversion | A funnel metric, not trial conversion |
ActivationPal reads subscription events from the RevenueCat webhook, or the Superwall webhook, and builds the cohort from the trial starts in the selected date range.
Its trials card shows started, then converted, pending and churned, each as a count and a percent of started.
Why trial starts are not revenue
ActivationPal maps the events like this:
- A trial start is
rc_initial_purchase with period_type: "TRIAL" and price 0. It is counted as a trial and stays out of purchases and revenue.
- A trial conversion is
rc_renewal with is_trial_conversion: true. It is counted as new-subscriber revenue, not a renewal, so the revenue mix (new subs, renewals, one-time) puts it under new subs.
Each trial is matched to its later events by the original transaction id, falling back to the user id. A cancellation after the trial has converted does not rewrite the trial outcome.
The purchases count is every paid transaction. Trial starts are not in it.
Free trial to paid conversion: a worked example
An example cohort: 200 trials started in the last 30 days. 62 converted, 40 are pending and 98 churned (71 cancelled, 27 expired). The card reads:
- Started: 200
- Converted: 62 (31.0%)
- Pending: 40 (20.0%)
- Churned: 98 (49.0%)
The two rates:
- Converted ÷ started = 62 ÷ 200 = 31.0%
- Converted ÷ resolved = 62 ÷ 160 = 38.8%
If all 40 pending trials churn, the rate stays at 31.0%. If all of them convert, it reaches 102 ÷ 200 = 51.0%.
The card also shows 50% churn by, the median time from trial start to churn, and churn paths split into cancelled and expired. Each outcome opens the list of users behind it.
Pending trials and reporting windows
A pending trial has started and has neither converted, been cancelled nor expired.
ActivationPal counts a trial as churned the moment auto-renew is switched off, not at expiration. Turning auto-renew back on returns it to pending. Outcomes that land after the selected range still update that cohort, up to now.
Do not judge a cohort until it is older than the trial length. For a 7-day trial, last week's cohort is still open.
For trends, use the Cohorts view metric Trial to paid:
- It shows converted trials ÷ trials started in each cohort period, regardless of install date.
- It runs by day, week or month over the last 12 periods.
- Periods still in progress are marked partial.
Dashboard ranges are today, yesterday, this week, 7, 14 or 30 days, this month, or a custom range up to 366 days. The same numbers come from get_stats and get_cohorts in the agent API.
Low trial conversion rate: what to check
- Is the cohort resolved? A high pending share means the rate is not low yet.
- Cancelled versus expired. Cancelled means the user switched off auto-renew during the trial. Expired means the trial ended without a payment.
50% churn by. Churn in the first hours points at the paywall promise or onboarding, not at the product later on.- Open the churned users. Read their timelines against onboarding and paywall events. This needs the RevenueCat app user id passed to the SDK so purchases join devices.
- Sandbox purchases. They are stored as debug and stay out of release numbers, so test trials do not dilute the rate.
- The step before the trial. Check paywall shown to trial start with the funnel calculator.
- Cohorts before and after a change. Compare weeks in Trial to paid around a paywall or price change.
- Refunds. Check the refund rate in the refunds tab, because a conversion that is refunded is not kept revenue.
Once the rate is steady, ActivationPal shows it next to your revenue, and the app revenue calculator turns it into a revenue estimate.