What actually happens at $5 a day
$5/day is roughly $150/month. Depending on your category and platform, that buys somewhere between 30 and 150 installs over the course of a month. That sounds like something. It isn't.
Ad platforms don't optimize on installs. They optimize on volume of the one event you tell them to chase. For a subscription app, that event is a trial start or a paid conversion, not a download. The platform needs to see enough of those events, consistently, to figure out who to find more of.
At $5/day, you're mailing the algorithm a postcard now and then. It can't learn from that. You end the month with a handful of installs, a cost-per-install number that means nothing yet, and no clearer picture of whether the channel works.
The question to ask instead
Flip it. Don't ask: "is $5 enough?" Ask: "what's the smallest budget that gets me past the learning phase, on one platform, on the one conversion event that actually predicts revenue?"
That question has a real answer. It depends on your app category, your platform, and your target conversion event. It's usually higher than $5/day. But it gives you something to aim for instead of something to doubt.
Most small budgets don't fail because they're small. They fail because they're scattered. Two platforms at $2.50 each. Three ad sets splitting $5. A budget that can't carry one channel ends up proving nothing on any of them.
Android vs iOS at a small budget
The category matters. On Android, in a cheaper app category, a small budget can generate enough conversion signal to learn something real. Not fast, but real.
On iOS, you're already starting at 3 to 5 times the install cost for comparable traffic. $5/day on iOS, for most subscription apps, mostly buys noise. You're paying for impressions in a system that can't optimize because you can't afford to feed it enough events.
This doesn't mean iOS is the wrong channel. It means $5/day on iOS is the wrong experiment.