Designing airships
The game pretty heavily revolves around designing airships, so if that fails to be fun, there's not so much point to everything else. To be interesting, there needs to be a complex landscape of different viable designs in rock-paper-scissor-like relationships to one another. If, on the other hand, there turns out to be just one or two ideal ship designs, everyone will gravitate there, and the whole ship editor may as well not exist.
Obviously, if one design is dominant, I can adjust things to nerf it. But this doesn't help if it just makes a different design dominant. The design landscape needs to have designs that are situationally dominant to be interesting.
Fighting
This game kind of violates my ideas about granularity. While the commands you give are high level, like "fly here" or "target this ship", their execution is far more detailed. This means you have to take into account details you don't have control over. At worst, this can lead to an experience like trying to use a mobile phone in oven gloves: your granularity of control is frustratingly insufficient.
What hopefully saves the game is that you do have fine-grained control during the ship construction stage. If that's not enough, there are roughly two options: increasing or decreasing granularity of control. The latter may seem counterintuitive, but this kind of frustration is partially borne from a feeling that you ought to be able to have more influence. By "zooming out", this feeling may be removable. The other option is to let you control things at a finer-grained level - say by being able to set settings on individual modules. (Like turning them on and off, prioritizing their repair, etc.)
Externalities
This leaves externalities - things that go wrong outside of the game itself. It's always possible that I can't finish the game for some reason. So far it's been really fun to make, but it's always possible that I lose my motivation - or just get too busy to work on it. That's not really something I can predict, so the best course of action is to steer towards having something playable as soon as possible. Then, even if it never gets "finished" in the sense of having all features I want to add, and polished to completion, you still get to play it, and it's still A Thing.
For now though, onwards! It's time to write that single-player campaign mode...