Paid Ad Metrics That Support Profitable Scaling for Subscription Apps
Cheap clicks do not prove an app can grow profitably. A campaign can look efficient on cost per click or click-through rate while bringing in people who cancel quickly, spend little, or were already subscribers. The stronger approach is to judge acquisition by what happens after the click: whether trials become purchases, whether new users stay engaged, how much paying customers actually generate, how quickly acquisition cost is recovered, and whether the people acquired are genuinely new.
Start with the balance between acquisition cost and lifetime value
Scaling is a balance between two numbers. Lifetime value describes what a customer is worth across the relationship. Customer acquisition cost describes what it takes to bring that customer in. Their relationship shows whether additional spend has room to work.
As budgets rise, acquisition cost often rises too. Early spending usually reaches people already aware of the app or actively looking for a similar product. Later spending reaches colder audiences, so simply demanding a lower acquisition cost can become a losing chase.
Raising realized lifetime value is often the more productive lever. It can support growth without requiring the same number of new customers, and it protects margins when media costs keep climbing. The practical question is therefore not only, “Can we buy users more cheaply?” but also, “Can we make each acquired user worth more?”
Measure acquisition cost by purchases, not trial starts alone
For apps with free trials, ad platforms often report trial starts because many trials do not convert inside the attribution window, particularly on iOS. Trial starts are useful as an early signal, but they are not the same as purchases.
True acquisition cost should be measured per subscription purchase. Where possible, include the wider cost of running acquisition: team or partner time, attribution tools, creative production and other recurring expenses. A monthly check of total costs against acquired users helps capture fixed costs that platform dashboards miss.
Always pair trial volume with trial-to-purchase conversion. A cheap trial is not necessarily a valuable trial.
Use trial-to-purchase conversion to judge sign-up quality
Trial-to-purchase conversion reveals whether sign-ups have commercial intent. Weak conversion often points to weaker lifetime value, although it should be read alongside engagement and retention rather than treated as a complete quality score.
Segment the result by audience, creative angle, individual ad and country. One useful pattern is that younger audiences may produce inexpensive trials while older audiences convert at higher rates and have greater capacity to pay. In that situation, excluding the cheaper segment can still lower cost per purchase.
Send converted-trial outcomes back to the advertising platform where available, then inspect the same breakdowns in your own dashboard. The aim is to stop judging a campaign by trial price when purchase price tells a different story.
Treat early engagement as an early quality signal
Waiting for a trial to end can be slow, especially with long trials. Early engagement can provide a faster indication of whether someone is likely to remain. Examples include completing a core feature, reaching an activation milestone, or returning for a defined number of sessions during the first week.
Timing matters. One product example found that users became much less likely to churn after completing six core actions on six different days, but that milestone usually occurred after seven days. Optimizing directly for it would have been difficult because attribution windows were short. The lesson is to choose an engagement signal early enough for the ad platform to use, while still being meaningful.
Depending on the product, teams may optimize for trials, an “activated trial” event sent after confirming the trial has not been cancelled, or a combined trial-plus-core-action event. The combined event is more expensive and needs enough volume, but it can reduce spending on people who cancel immediately. Apps without trials should be cautious: adding an engagement goal can reduce event volume too far for the platform to learn. Monitor engagement first, then test it as an optimization event once spend is established.
Prefer realized lifetime value over projected lifetime value
Forecast lifetime value is not cash that can be reinvested today. Realized lifetime value per paying customer, also called average revenue per paying user, uses actual revenue minus refunds for a cohort divided by the number of customers in that cohort.
Compare this figure across campaigns and ads. It shows which sources produce customers who generate more revenue, not merely more sign-ups. One example involved separating business users from individual creators after onboarding data showed much higher realized value from business users. Sending that higher-value signal back to the ad platform helped focus spending on the more profitable group.
Use realized value to improve the relationship between revenue per paying customer and acquisition cost. That is a more reliable basis for deciding whether to expand a campaign than projected lifetime value alone.
Use payback period to pace growth
Payback period is the time needed to recover acquisition cost. A shorter period allows the growth loop to move faster: spend brings in subscribers, subscribers generate revenue, and that revenue funds more acquisition.
Forecasting payback can be difficult when plans, durations, trial conversion and renewal behaviour vary by cohort and season. A practical alternative is to track the blended relationship between average revenue per paying user and acquisition cost at set points such as day 0, day 3, day 7, day 30, day 90 and day 365. Historical cohorts provide a forward-looking estimate for newer cohorts.
Long payback periods carry risk if realized revenue falls short of expectations. When payback is too slow, prioritize segments that naturally generate higher realized value instead of simply increasing spend.
Compare paid-user churn with overall churn
Churn shows how many subscriptions are lost during a period. Compare churn among paid-acquired users with overall churn. Paid users may churn more than organic users, but a large gap can indicate the wrong audience, weak onboarding or expectations set too high by the campaign.
Churn may sit outside a performance marketer’s direct control, yet it still affects lifetime value and therefore the bids a platform can support. For annual plans, churn is a lagging indicator, so early engagement and monthly renewal rates provide faster warnings.
Check that spend creates new subscribers
A favourable acquisition-to-value ratio can be misleading if much of the spend reacquires existing subscribers. Reacquiring someone may cost as much as acquiring a new customer while adding little incremental growth.
Track cost per new trial or subscriber by ad. One operator used a dashboard joining ad-platform data with app data to watch cost per new user alongside revenue per paying user divided by acquisition cost. Rising cost per new user served as an early warning that purchase cost was likely to follow.
Treat cost per new user as directional rather than definitive. Check incrementality and attribution quality, and use shorter click-based attribution windows where view-through attribution may overstate a campaign’s contribution. The central test is whether more spend produces more genuinely new subscribers and more total revenue.
A practical scaling checklist
Before increasing budget, work through this checklist:
- Calculate acquisition cost per purchase, including wider operating costs where possible.
- Segment trial-to-purchase conversion by audience, ad, angle and country.
- Define an early engagement or activation event that occurs soon enough to be useful.
- Compare realized revenue per paying customer across campaigns, not only projected lifetime value.
- Track payback using historical cohorts at consistent day intervals.
- Compare paid-user churn and renewals with overall retention.
- Track cost per new user and verify that growth is incremental.
- Send the strongest downstream quality signals back to the ad platform.
At smaller scale, cost per trial, trial-to-paid conversion and payback provide the fastest read on whether the funnel works. As scale grows, realized value, retention and contribution margin per cohort become more important. Cheap traffic can start a campaign, but sustainable scaling comes from acquiring the right new customers and retaining enough value from them.
Frequently asked questions
How do you use new trialists and subscriptions in practice?
Use cost per new trial or subscriber, not total trials or purchases. Even with subscriber data connected to an ad platform, some spend reaches existing users, so identify which ads actually bring in new subscribers before increasing budgets.