You were drunk and thinking about fertilization and the result you came up with was programming?!?
Playability is by far the most important thing to me. So the first thing I do is hack together a basic engine as a proof of concept for whatever game mechanics are swirling about in my head. So for Space Squadron, I'd already made similar games, so I simply tested to make sure I could do the same thing on Android.
After that, with Android, I did a performance test. So I upped all my entities to see what the platform was capable of. When I knew how many ships and rockets etc. I could support on the average phone, I then designed from Vental's kind of approach, thinking about cool mission ideas and a very basic story to tie it all together.
Then I worked out what programming/media tasks I'd set myself by designing missions. So by having a convoy escort mission, I knew I'd have to design convoy ships and AI, then also write opponent AI to target those ships. Initially I wanted the player to dock with the convoy, but I took that out because of the feature creep, and I was making the problem domain too big for myself.
With Fat Ball (my current android game), I did the same thing. I hacked together the basic engine to make sure it was fun and playable on the phone. After I was happy with that, I took a step back and planned all the elements - so training mode, quick game, tournament mode, and multiplayer. But the good thing about doing everything yourself is you can let it evolve in your head, without trying to communicate with others. For example, initially I was going to have bluetooth multiplayer, but I decided that was too much of a challenge and switched it to 2 player on the same handset. The tournament mode also is just being fleshed out as I go, without much design, because up until this point, I didn't know what characters/arenas/AI skills I would have in place, which are important to how it comes together.
So after all that waffle, I would say the most important thing to do is get the core gameplay mechanics working with little or no design. Make sure you have a fun concept. Then step back and look at the big picture, and design a game that works with those mechanics and complements them, but is within the scope of your abilities and time.