Trial Conversion Analytics

View how many prospecting users are converting into paying users from your free trials

Updated over a week ago

To help you understand how your subscription trials are converting to payment, we've implemented a new trial conversation analytics graph that allows you to track how many people are in trial, and out of those, how many converted or have an expired trial.

To access the In Trial Overview graph, navigate to Subscription, then click on See Breakdown next to In Trial.

This will take you to the In Trial page where you will see a graph displaying each month's results. Below it, you will have a breakout of each month. The data collected is divided into four metrics:

  • Trialing (Blue): a new user started a trial for one of your subscription plans, and the trial hasn't yet finished. This includes subscribers whose trial conversion payments failed and the subscription status is currently On Hold - Their invoice is Unpaid, but not yet Overdue.

  • Converted (Green): a user went from being In Trial to an Active user.

  • Expired (Red): a user canceled their plan before the trial was over or the credit card was declined after several attempts and the system marked the invoice as Overdue, which means the user didn't convert to an Active user.

  • Conversion rate (Black line): the percentage of the users that converted from In Trial to Active users. Trial conversions are calculated by dividing the number of conversions in the given month by the number of expired trials:
    Converted / (Converted + Expired) x 100

NOTE: Each person appears only one time in the month when they started a trial.

You can compare each month how many people are converting into new paying users or canceling their plans by hovering over the graph. When hovering over a month, you will see the actual number for each metric—just like in the breakout area—and a percentage that compares the selected month with the previous month.

For example, when I hover over March, I had 43.91% fewer expired trials when compared to February. That means that either more people are still in trial or had more conversions than last month. In the same month, I also had 15.58% more new users in trials than the previous month.

When comparing data on your Subscription tab with the In Trial Overview graph, know that the data we are collecting for the graph is in the In Trial column. The Total column is to show you how many are in trial plus how many are already paying users.

NOTE: The graph's data will be provided on a rolling 12-month period. It will start from the month the feature was released along with the previous 11 months.

Possible reasons for discrepancies in trialing analytics

Due to the calculation and population of trial data in different parts of the Admin Area, you may notice some discrepancies at times.

For example, you may notice that the number of "Trialing" users in the In-Trial graph for a given month is different than the number of "In Trial" users in your People page.

The In Trial graph shows you the number of new trials for a specific month in the Trialing row. These are users who entered into a trial in that specific month, but they may have created an account with your store before the month they went into the trial.

When you look at the People page and filter for In Trial, you get a list of all users who are in trial over a certain date range. This date range will only show users who were created during that time frame and went into the trial. This may result in a discrepancy between the Trialing number in the In Trial graph and the In Trial category in People.

When you consider the scenarios above, you could have situations where the number of Trialing users in the In Trial graph for one month is greater than the number of In Trial users over the same month filtered in the People page.

This could be explained by the fact that the users who entered the trial during a specific month had created an account in a prior month, but failed to enter the trial.

The information provided in these graphs cannot be downloaded as reports; however, you can create a feature request if this is something you would be interested in.

Did this answer your question?