MRR is recurring proceeds revenue normalized in to a monthly amount. It's calculated as a sum of the monthly fee paid by each paying customer with a deduction of Apple comission.
For example, If you have 10 customers paying $10 per month. MRR will be equal to:
MRR = 10 subscribers × $10 per month - 30%(or 15%) Apple Comission = $70.
All non-monthly subscription plans are being normalized in to a monthly plan. For example, annual subscription that costs $120 per year will be normalized in to $10 per month.
Gross revenue
Cohorts: No.
Gross revenue is a total amount billed to customers for purchasing subscriptions prior to refunds, taxes and Apple’s commission.
Sales
Cohorts: No.
Total amount billed to customers for purchasing in-app purchases. Sales = Gross Revenue - Refunds.
Proceeds
Cohorts: No.
Estimated revenue developer receives after deducting Store commission and VAT.
Refunds
Refunds amount during selected period. This includes both refunds through Customer Care and refunds during subscription upgrade with partial refund.
Refund Rate
Cohorts: No.
The percentage of the total transactions that were completely refunded during a period of time.
Formula is the following:
Refund Rate = Refunded Transactions / All Transactions by the Period * 100%
ARPU
Cohorts: Yes.
APRU is Average Revenue Per User.
Calculated on a cohort basis. The cohort is users, that have installed the app within the selected period. They can be segmented and filtered by country, products, marketing campaigns (depending on what was selected in segments and filters).
We calculate ARPU for the users’ lifetime. This means that ARPU for the given cohort will be calculated by dividing the summarized users' revenue by the total number of users who have installed the app within the selected time frame.
You can choose whether to calculate ARPU using Sales or Proceeds.
Calculate users by events
User created – count all new users in the selected cohort for ARPU calculation.
For example, we have 100 new users created within the selected period. These users have paid $10 by now. ARPU is equal to $10 / 100 = $0.1 (per user).
Trial started – calculate an average revenue only among those new users, who have started a trial.
For example, we have 100 new users created within the selected period. And 50 of them have started a trial. These users have paid $10 by now. ARPU is equal to $10 / 50 = $0.2 (per trial user).
Cohort period
Default value - Max. Number of the days (0-365) or Max (no limitation). Day 0 is the 24 hours from the moment when user is created. Day 1 is the next 24 hours and so on.
This setting might be used to understand the average revenue to the Nth day of user life. So you can estimate the moment when the paid installs are paying off.
For example, if ARPU for new users by the Day 3 is $0.1, this means that an average user brings $0.1 to the 4th day of his/her lifetime.
Why ARPU is important?
One of the main goals of any mobile business is to maximize revenue. ARPU is one of two critical values for calculating your ROI along with CPI (cost per install).
Knowing the ARPU of different groups of users will let you optimize your marketing efforts and focus on high-performing user acquisition campaigns. ARPU-driven measurement help to point out which media sources could bring the top paying users and which are under-performing and should be stopped.
Understanding who is your high LTV users can help to find more like them through lookalike campaigns and leads you to boost total app revenue.
ARPPU
Cohorts: Yes.
ARPPU is Average Revenue Per Paying User.
It's pretty much the same as ARPU, except there are only paying users counted within the selected time period.
You can choose whether to calculate ARPPU using Sales or Proceeds.
Calculate users by events
Purchase – count all new paying users in the selected cohort for ARPPU calculation.
For example, we have 100 new paying users created within the selected period. These users have paid $200 by now. ARPU is equal to $200 / 100 = $2 (per user).
Trial converted – calculate an average revenue only among those new paying users, who have converted from a trial.
For example, we have 100 new paying users, created within some selected period. And 50 of them have converted from a trial. These users have paid $200 by now. ARPPU is equal to $200 / 50 = $4 (per free trial user converted into customer).
Cohort period
Default value - Max. Count of the days (0-365) or Max (no limitation). Day 0 is the 24 hours from the user’s creation. Day 1 is the next 24 hours and so on.
This setting might be used to understand the average revenue to the Nth day of the paying user life. So you can estimate the moment when the paid installs are paying off.
For example, if ARPPU for new paying users by the Day 3 is $1, this means that an average paying user brings $1 to the 4th day of his/her lifetime.
Additional notes
Both ARPU/ARPPU are calculated without refunds. Downgrades (partial refunds) difference is counted.
ARPPU will be usually much higher than ARPU because only paying users included in cohorts (a smaller part of all users in the app).
Cumulative LTV
The Cumulative LTV chart shows a cumulative (accumulative) revenue for cohorts up to 365 days. So this is a cumulative dynamics of how the average profit from a selected group of users grows within the selected cohort period. Cumulative means that if you set the cohort period 5D, the metric is calculated by adding 0D + 1D + 2D + 3D + 4D.
CHURN
Subscriptions churn
Cohorts: No.
This metric shows how many subscriptions were lost during the selected period.
Subscriptions Churn = (Number of subscriptions expired during the period) / (Number of paid subscriptions at the start of the period) * 100%
Number of paid subscriptions at the start of the period means that any new subscriptions, that could happen inside the period are not included.
Churn also can be negative. This may happen if your app allows customers to have more than one subscription at the same time and the user adds a purchase to existing ones. It's called "expansion". Negative churn is a signal that your app has a strong value to your customers.
Churned revenue
Cohorts: No.
This metric shows how much revenue was lost during the selected period.
Churned revenue = (MRR Lost to Downgrades + MRR expirations in the period) / (MRR at the start of period) * 100%
MRR at the start of the period means that any new MRR inside the period is not included.
Churned revenue especially helpful to ensure that you're not losing the most valuable customers (subscriptions churn can be relatively low while churned revenue is high).
Like subscriptions churn, revenue churn also can be negative if users upgrade subscriptions or subscribe more within your app. And, of course, it's a good signal too.
USERS
New users
Cohorts: No.
Yes, these are just new users of the app. By the way, you can filter them, using custom User Properties (if you send any).
Pay attention that new users counted when they were recorded by Apphud SDK (not actual install date). In 99% of cases, these dates are nearly identical.
CONVERSIONS
All charts in this group represented as funnels with particular steps (based on events).
New users are calculated as cohorts. It means that users are "grouped" by the initial event date (i.e. users, who have installed the app on a particular date).
For example, if we have a row with a date of Jan, 28, then trial conversions are summarised by users who had installed the app on Jan, 28 and then converted to regular subscribers any day after install and trial start (not only on Jan, 28).
Note that all conversion events are grouped by new users.
So, if your app allows two simultaneous subscriptions (say, with a trial period), then it will be counted as the single user who started a trial (not two trials).
The same works for trial conversion events – if there are two trials that were converted to regular subscriptions by the user within the app, we'll show this as a single user who converted from the trial(s) to subscriber.
Trial conversion
Cohorts: Yes.
New users, who started a trial and then converted to regular subscribers. Users who converted without a trial period are not counted.
Regular subs. conversion
Cohorts: Yes.
New users, who converted to regular subscribers (trial conversions are not included).
Paid intro conversion
Cohorts: Yes.
New users, who started a paid intro offer and then converted to regular subscribers.
Promo offer conversion
Cohorts: Yes.
New users, who started a promo offer and then converted to regular subscribers.
Non-renewing purchase conversion
Cohorts: Yes.
New users, who purchased a non-renewing in-app product.
Segments
You can segment conversion charts by the second additional parameter to get more meaningful results.
View parameters
For example, you have 100 new users, 10 of them started a trial and 2 were converted to a subscription.
% of total
Shows drop-off in each funnel step calculated as a percent of total (first step) count.
In this case, the total count (new users) will be 100%, the trial started – 10%, and converted – 2%.
% of previous
Shows drop-off in each funnel step calculated as a percent of the previous step.
New users will be 100%, the trial started – 10% and converted – 20% (because started trials are considered as 100% in this case).
User count
Shows absolute users count in each step.
Exclude refunds
Default: disabled.
Note that, we calculate all values including transactions with refunds by default. If you want to analyze only "successful" conversions, turn on Exclude refunds option.
Keep in mind, that you can see lower conversions in this case.