
Advertising an app to obtain installs is easy to start and easy to do badly: many cheap installs may be worth nothing if people open the app once and never return. The right focus is on quality: activation, usage and value over time.
The essentials in a definition
App campaigns seek installs and in-app actions, and quality counts more than volume: installs that stay, buy and return. Measurement on iOS depends on SKAdNetwork and App Tracking Transparency, which limit user-level data, and in-app event measurement can rely on GA4 events. Choosing the right events and measuring retention is decisive.
Installing is not the goal
An install is only the beginning. What counts is that the user completes the first important action (sign-up, first purchase, first task) and returns. Campaigns optimised only for installs tend to bring low-quality users. Define a value event and optimise for it.
Cost per install alone is a misleading metric.
Prepare measurement
Configure in-app events that reflect the journey: sign-up, activation, purchase, subscription. Ensure data reach the ad platforms, respecting operating-system privacy and consent rules. Consider aggregated measurement where individual tracking is limited. Validate events before investing.
Without events, the system optimises blind.

Quality-focused app campaigns
- Define a value event
- Set up measurement and consent
- Care for the store listing and onboarding
- Optimise for action and value
Creatives and store listing
Creatives should show the app in use and the benefit. The app-store listing is part of the funnel: icon, screenshots, description, reviews. Test variations. Align the ad’s message with what the user finds on opening. Offer a quick, clear first experience (onboarding).
The first session decides a lot.

A cheap install that never returns is the most expensive of all.
Optimise for value
Move from installs to action events and then to value, when there is enough data. Analyse retention by cohort and campaign. Compare cost per active user and value over time. Eliminate low-quality sources. Scale gradually.
Scale only what retains.
Decision table
| Stage | Focus | Indicator |
|---|---|---|
| Launch | Ensure measurement and first-time experience | Installs and activation |
| Optimisation | Value event (sign-up, purchase) | Cost per value event |
| Retention | Coming back and continuing to use | 7- and 30-day retention |
| Monetisation | Purchases and value over time | Value per user |
Common mistakes to avoid
- Optimising only for installs. It attracts users who do not stay.
- Measuring without value events. The platform does not learn what matters.
- Ignoring the first-time experience. Many abandon on first use.
How to measure the result
Track cost per install, activation, cost per value event, retention and value per user.
Limits and exceptions
Mobile platforms’ measurement and privacy rules change and limit available data, and attribution is partial. Results vary by category and market. Validate with cohorts and tests before scaling.
Illustrative example
Illustrative example: a delivery app launches with install campaigns and, after two weeks of data, switches to optimise for first order. It tracks 7-day retention and value per user, and adjusts the first-time experience. On iOS, it uses the available aggregated metrics and interprets them with caution.
In practice: checkpoints
Test the first-time experience on several devices and networks before increasing budget, because technical failures waste paid installs. Provide clear store-page assets, short onboarding and a quick path to the first valuable action. Define cohorts by channel and creative, review retention at day 1, 7 and 30, and compare value per user. Respect platform privacy rules, ask for permissions at the right moment and explain why. Update the app regularly, respond to reviews and keep an eye on crash rates, since quality problems quickly reduce both ratings and conversion.
Frequently asked questions
Should I optimise for installs?
At the start it may serve, but move to value events.
How do I measure quality?
With retention, activation and value by cohort.
Does the store listing matter?
A lot. It is part of the funnel.
Which event should I choose to optimise?
The one closest to value with enough volume for the platform to learn.
How do I measure on iOS?
With the available aggregated metrics and models; individual data are limited.
In app campaigns, optimise for value events and retention, not just installs, and interpret limited iOS measurement with caution.
Sources and recommended reading
- Apple: SKAdNetwork
- Apple: user privacy and data use on the App Store
- Analytics Help: GA4 recommended events
Last editorial review: October 2026. Informational content; always confirm current conditions in the official sources before deciding.