About the client
An idle arcade title stuck below the profitability line
A hybrid-casual idle arcade game on Android, running $30K–$40K per month acrossTier 1 and Tier 2 geos over six months on a hybrid monetization model. The title was unprofitable and D30 ROAS sat well below the level needed to scale, with a hard goal of 100% ROAS by D60. Install volume was maintained throughout while the title turned profitable.
SOLVING ADVERTISING CHALLENGES TOGETHER
Spend was concentrated, and nothing was optimized for return
All budget sat on two channels — Meta and Google — with no bidding strategy behind either, and creatives were judged on cost per install, which said nothing about whether those installs ever paid back. We came in as a full growth partner, owning UA execution and feeding product signals back into the game.
WHAT WE DID
Channel & geo diversification, and creative scored on ROAS
- Expanded intoUnity and AppLovin alongside Meta and Google, with a dedicated bidding strategyper channel.
- Mapped thewinning combinations: AppLovin → Tier-1 (US / UK / FR), Unity → US, Google →Korea — and found a Southeast Asian Tier-2 market as a hidden profitabilitypocket.
- Used MMPattribution to measure each creative’s downstream ROAS, separating IAA ROASfrom IAP ROAS for the hybrid model.
- Surfacedcreatives with slightly higher CPI but ~30% higher ROAS, then iterated theirfirst 5–10 second hook to lift install rate.

RESULTS THAT EXCEEDED EXPECTATIONS
D30 ROAS trajectory over six months
Diversifying beyond Meta and Google, matching bidding to each channel, and judging creative on return turned an unprofitable title into a profitable, scalable one — D30 ROAS grew 2.27×, D60 ROAS hit 109% against a 100% target, and install volume held throughout.
