Mobile Game And App UA: The Complete Guide
Playad Team
Last updated: Aug 30, 2026
The short version
App user acquisition is not ecommerce advertising with a different conversion event. Four structural differences drive almost everything:
- An install is terminal. You acquire a user once and they leave the addressable pool permanently. Scaling is a depletion problem.
- iOS measurement is coarse and delayed by design. Attribution is aggregated, lagged, and systematically incomplete.
- Creative buys access, not just efficiency. On networks that rank by IPM, weak creative loses volume, not only margin.
- The install is not the outcome. Retention and revenue arrive days and weeks later, so acquisition decisions are made before the information exists.
Everything below follows from those four.
Start here
- CPI: Why A Falling Cost Per Install Can Be Bad News — the metric everyone optimizes and the trap inside it
- Retention Rate: Read The Curve, Not The Day — whether the install became a user
- Scaling App UA Without Breaking Your Cohorts — why quality degrades by construction
If you read three things, read those. They contain the reasoning the rest depends on.
The metrics
| Metric | Answers |
|---|---|
| IPM | How hard does this creative pull, independent of bid |
| CPI | What did an install cost |
| Retention Rate | Did the install become a user |
| ARPU | What is that user worth, and over what window |
| Payback Period | When do you get the money back |
Read them in that order. Each one is meaningless without the one after it — CPI without retention is a cost with no counterpart.
iOS measurement
The part that surprises people arriving from web advertising.
- SKAdNetwork (SKAN): What iOS Attribution Can And Cannot Tell You — the four constraints that matter
- The ATT Prompt: Why Your iOS Numbers Are Incomplete By Design — and why the missing data is not randomly distributed
- Why Your iOS And Android Numbers Do Not Match — mostly measurement, not performance
The practical consequence: never compare iOS to Android at a point in time, and expect to underfund iOS if you allocate on reported numbers.
How the buying actually works
Worth reading even if you never touch a mediation setting. It explains why a network stops scaling when more budget changes nothing, and why creative quality determines how much inventory you can reach.
Creative
Where most of the leverage is, and where the operational constraint usually sits.
- Creative By Game Genre — monetization model sets the brief
- Playable Ads: What Makes One Work — the format that qualifies hardest
- Why UA Burns Creative Faster Than Ecommerce — why the pipeline is the strategy
- Soft Launch: What Creative Testing Should Actually Tell You — test angles, not executions
- Creative Production As A Pipeline
The step everyone forgets
Every paid install passes through the store listing, and improvements there lift every campaign, network, and creative at once. It is usually the only step nobody has tested, because it sits between the UA team and the product team.
The order to work in
- Get measurement right first. An MMP connected, and honest expectations about what iOS can tell you.
- Establish what a user is worth before setting any acquisition target.
- Find an angle that works, then build a pipeline to produce variations of it.
- Fix the store listing, because it multiplies everything else.
- Scale on marginal cohort performance, not blended averages.
Most struggling UA operations are stuck at step 2 or step 3 — either paying without knowing what a user is worth, or producing creative one asset at a time.
Practical rule
Judge everything on cohorts, not on installs. Install volume is the metric that is available first and matters least.
Automate marketing in less than 2 minutes