I recall the reported costs of "junk rockets" vs those surface-to-air defenses, which I believe are a similar case of the attack being orders of magnitude cheaper than the defense
E.g., when there was a new bunker-buster bomb in the initial Iraq war, one was developed in 28 days. Similarly, the MOAB was developed in about a year. There are plenty of teams of highly skilled and well-equipped engineers who will happily undertake to solve that problem.
Also, weapons systems with higher expected rates of use can benefit from economies of scale-ed up production.
Unless they're sending a million balloons this week, it won't be that kind of problem.
The recovery might be something else, but then again, those groups are constantly training anyway, so I'd venture to guess the difference is negligible.