Case study · Multi-channel · ROI
When three platforms claim the same revenue
A subscription app sees a healthy 1.90 reported ROAS. Measured against sales that would have happened anyway, the real figure is 1.30, and that changes where the budget should go.
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.
- 1.90
- Reported ROAS
- 1.30
- Incremental ROAS, before
- 1.50
- Incremental ROAS, after
- +$12,000
- Added incremental revenue per month
All figures are hypothetical.
The situation
Imagine a subscription fitness app spending $60,000 a month: $15,000 on Apple, $30,000 on Meta and $15,000 on Google. Each platform's dashboard looks fine. The growth lead reports a blended ROAS of 1.90, yet the finance team can't find that money in the revenue line.
Why the numbers disagree
Each platform counts conversions by its own rules and has no view of the others. A person who sees a Meta ad, then searches the App Store and taps an Apple ad, can be counted by both. Some of those people would also have subscribed with no ads at all.
| Channel | Spend | Reported revenue | Reported ROAS | Incremental revenue | Incremental ROAS |
|---|---|---|---|---|---|
| Apple | $15,000 | $36,000 | 2.40 | $24,000 | 1.60 |
| Meta | $30,000 | $54,000 | 1.80 | $36,000 | 1.20 |
| $15,000 | $24,000 | 1.60 | $18,000 | 1.20 | |
| Total | $60,000 | $114,000 | 1.90 | $78,000 | 1.30 |
The platforms claim $114,000, which is 46% more than the $78,000 the spend actually added.
How we'd get to the real number
- Start from backend revenue. Paid subscriptions from the billing system are the only figure every team can agree on.
- Remove the baseline. Use a holdout to estimate what organic demand would have delivered without paid media.
- Split what remains. Test channel by channel with geo holdouts and brand search pauses, starting where the budget is largest or the claim looks most generous.
What would change
Apple keeps the highest incremental return, so some budget moves to its category and competitor campaigns. Meta's spend stays level, but with a stronger creative pipeline. Google's App campaigns move from install goals to in-app value. Spend stays at $60,000, and incremental revenue rises from $78,000 to $90,000, for a 1.50 incremental ROAS. That's 15% more revenue for the same budget.
The takeaway
Reported ROAS is still useful for day-to-day optimization inside a single platform. It shouldn't decide how budget is split between platforms. Keep one incremental number for that decision, and refresh it whenever the mix changes materially.