How we spent $43,560 in Apple Ads100,000+ installs and 56% day-7 ROAS

We spent $43,560 through Apple Ads and generated more than 100,000 installs. The useful part starts further down the funnel: 21,633 registrations, 4,213 first deposits, 15,610 repeat deposits and 56% ROAS by day seven.

Installs were only the top of the funnel. If they do not turn into registrations, deposits and returning users, a beautiful CPI screenshot does not make the campaign profitable.

The task: capture hot demand, not a bag of installs

ASA reaches people who are already searching inside the App Store. They are closer to action than a cold social audience, but dumping every keyword into one campaign still does not create magic.

We separated brand, category, competitor and discovery from the start. Exact terms had their own structure, while broad and Search Match worked as scouting layers. This showed where real demand lived and where Apple had simply found another fast way to spend the budget.

The numbers

• $43,560 in spend;
• more than 100,000 installs;
• 21,633 registrations;
• 4,213 first deposits;
• 15,610 repeat deposits;
• 56% ROAS on day seven.

Roughly one in five installs reached registration, and almost one in five registrations reached a first deposit. Repeat activity pushed the picture much further: 4,213 first deposits were followed by another 15,610 repeat deposits.

Where the buying stalled

ASA can produce volume quickly. The harder part is not confusing volume with quality. Some terms delivered cheap installs but fell apart further down the funnel. Others cost more at the start and produced much healthier registrations and deposits.

We never killed or scaled a campaign on CPI alone. The full chain mattered: tap → install → registration → FTD → redeposit. If a setup did not survive to revenue, it was cut even when the top of the dashboard looked attractive.

How we tuned the setup

The work ran in short iterations:

1. Clean search terms and add negative keywords.
2. Move proven demand into dedicated exact campaigns.
3. Rebuild bids and budgets around post-install quality.
4. Match product pages and opening screens to the user's intent.
5. Give strong cohorts enough time to produce repeat deposits instead of judging one day.

The App Store page is part of the setup. Gaming demand needs a clear game loop and a fast route into the product. Subscription demand needs clear value before the paywall. One generic visual for every query usually produces one generically average result.

Why 56% ROAS on day seven mattered

Day-seven ROAS is not the final payback of the cohort. It is an early read, but it already shows that the traffic is not empty: users install, register, make the first deposit and return with repeat activity.

We were not trying to force a 100% screenshot into the first week. The goal was a controllable acquisition system where the team knew which setup deserved scale and why it returned money.

What moves into the next scale

The next step is straightforward: more budget for clean exact setups, new markets only after product-page and event checks, separate CPPs for different intent and constant reconciliation between Apple Ads and product data.

For the operating model, read our Apple Ads guide. To see whether ASA fits your product, explore the ASA service or show the app to our team.