ARPAS (average revenue per active subscriber)
ARPAS tells you what the average subscriber is worth over a period. It sits between revenue per user and revenue per paying user, and for a subscription app it is often the most useful of the three to set against acquisition cost.
Updated
Most subscription apps have three kinds of users: people who installed and never started a trial or subscription, people who started one, and people who ended up paying. ARPAS averages revenue across the middle group, the users who entered your subscription funnel.
How ARPAS is calculated
Adapty Ads Manager calculates ARPAS as total revenue divided by the number of active subscribers. Subscribers here means users who activated a trial period or a subscription.
Revenue in Ads Manager is counted before store commission, so ARPAS is a gross figure. It is also available as a cohort metric, which ties revenue back to the subscribers acquired in a given period and tracks how it grows as they renew.
ARPAS vs ARPU and ARPPU
The three metrics share the same numerator and differ only in who is counted in the denominator. Using the same $60,000 of revenue makes the difference clear.
| Metric | Divides revenue by | Example |
|---|---|---|
| ARPU | All unique users | $60,000 / 5,000 users = $12 |
| ARPAS | Users who activated a trial or subscription | $60,000 / 1,500 subscribers = $40 |
| ARPPU | Unique paying users | $60,000 / 1,000 paying users = $60 |
ARPU spreads revenue across everyone who installed, ARPPU across only those who paid. ARPAS counts trial users who have not paid yet, so in a typical subscription app, where some trials do not convert, it sits between the two.
Using ARPAS in Apple Ads
ARPAS is useful because it lines up with a cost you can measure. If you know what an Apple Ads keyword costs per trial or per subscription, ARPAS for the same keyword tells you what those subscribers earn on average. When ARPAS over your payback window is comfortably above that cost, the keyword can take more budget.
- Compare keywords on subscriber quality. Two keywords with the same cost per trial can bring subscribers with very different ARPAS, for example if one attracts annual plans.
- Watch trial-to-paid conversion. A falling ARPAS often means more trials are ending without payment, which points at the paywall or the trial offer rather than the bid.
- Use cohorts for decisions. ARPAS for a period mixes old and new subscribers; cohort ARPAS at a fixed age compares like with like.
An illustration of the arithmetic: keyword A costs $25 per trial and its subscribers show an ARPAS of $40 at the cohort age you use for payback; keyword B costs $15 per trial with an ARPAS of $18. Keyword A leaves $15 per subscriber above acquisition cost, keyword B $3, so the more expensive keyword is the better one to scale.
What moves ARPAS
- Plan mix. More annual or higher-priced plans raise revenue per subscriber in the period they are paid.
- Trial conversion. Trials that end without payment add to the denominator but not to revenue.
- Renewals. The longer subscribers keep paying, the more cohort ARPAS grows with age.
- Refunds and pricing changes. Both change revenue without changing the subscriber count.
Related metrics
ARPAS pairs naturally with cost per trial, cost per subscription and trial-to-paid conversion on the acquisition side, and with ROAS, ROI and LTV on the revenue side.