Case study · Google Ads · ROI
Why paying more per install can raise ROAS on Google
Moving App campaigns from installs to a downstream in-app event raises the cost per install from $2 to $5, yet this hypothetical app ends up with 30% more subscribers from the same budget.
Illustrative scenario. The company and all figures are hypothetical and show how we’d approach the problem. This isn’t a client result, and your numbers will differ.
- $2.00 to $5.00
- Cost per install
- $100 to about $77
- Cost per subscriber
- 1.20 to 1.56
- First-year ROAS
All figures are hypothetical.
The situation
Imagine an app spending $20,000 a month on Google App campaigns with an install goal. It gets 10,000 installs at $2.00, which looks efficient. But only 2% of those users subscribe, and a subscriber is worth $120 in the first year.
What the goal tells the system
A campaign set up for installs looks for people likely to install, and it doesn't care what they do next. Switch to an in-app event that signals value and the system looks for different people, usually at a higher price. Install volume drops and the dashboard appears to get worse, which is why many teams reverse the change too early.
The math
| Install goal | In-app event goal | |
|---|---|---|
| Monthly spend | $20,000 | $20,000 |
| Cost per install | $2.00 | $5.00 |
| Installs | 10,000 | 4,000 |
| Install to subscriber | 2.0% | 6.5% |
| Subscribers | 200 | 260 |
| Cost per subscriber | $100 | About $77 |
| First-year revenue at $120 | $24,000 | $31,200 |
| First-year ROAS | 1.20 | 1.56 |
How we'd make the switch
- Pick an event with enough volume. The system needs a steady flow of conversions to learn. If subscriptions are too rare, start with an earlier event that predicts them, such as a started trial.
- Run the new goal alongside the old one. Don't change everything at once. Split the budget, then compare on cost per subscriber.
- Agree on the success metric first. Decide in advance that cost per install isn't the decision metric, and allow for the learning period and conversion lag.
- Check the value goal later. Once revenue events flow reliably, test a target ROAS approach.
The takeaway
Judge a campaign by the event you actually get paid for. A rising cost per install is only bad news if the cost per subscriber is rising with it.