Over the last few years, a few folks have asked me how I knew when I was done with a game. For me, it actually isn't super-complicated, because I generally don't start actively working on a game until I have a very clear idea of what the end product is going to be, and then I just get to it and I just keep at it until the game resembles that idea. I go into it with my parameters already defined: what the game is going to feel like, what tensions are going to be present, what ideas are going to be expressed.
While the process of game design and testing is inherently iterative, creating a sort of informal "brief" for myself greatly decreases the number of iterations required. I know there are other designers who approach the process very differently - you sometimes hear people talking about "finding the fun" or even "learning what my game was really about". Approaching it that way as a matter of course would drive me absolutely bonkers.
Though on the other hand, I probably do approach it that way, only all the experimentation and fumbling and discarding of ideas all happens in that preliminary stage where I'm forming that picture in my head. Sometimes it takes several months; in some cases, years later, I'm still waiting for things to come together. So probably it's not any quicker, and may even be slower, than just trying one darn thing after another. It does however require a lot less energy, and since I often have a half-dozen or more projects in the hamper at once, it lets me focus that energy in a more immediately useful direction.
It also forces me to set one game aside for a while so I can shift my focus to another, and this is an invaluable part of developing and testing a game, because when you come back to it, it is with fresh eyes. It is so easy to get too close to the thing, and that can compromise your ability to see the thing clearly. And naturally as a designer (and publisher) one is eager to get the game onto tables. No designer sets out to make a bad game, or one that's chock-a-block with errata, but that heady mix of passion and impatience can cloud their judgment.
Patience. It's a hard thing. The Lord has given me many blessings, but patience was never one of them. I think if I only worked on one design at a time, I would be a very bad designer, because I wouldn't have the patience to wait and wait and wait until that picture formed in my head: my games would be poorly conceived. There's no way I'd ever be able to set something aside and wait and wait and wait before coming back to it with fresh eyes: my games would be poorly developed and tested. And then, when they were published, they would be poorly received. For a designer like myself and for a company like ours, that would be the kiss of death - this whole crazy improbable ludicrous wonderful thing is built on our reputation.
Thank goodness I'm slightly scatterbrained, my attention always flitting back and forth between different projects. It doesn't really feel like waiting when I have something else to fixate on. That gives me the time and distance I need from a project so that I can evaluate its triumphs and faults more objectively. I'm so impatient, it seems, that I end up being patient. Especially when I think the thing is "done" and there's no more work to do, it's important that I give it a few weeks or months before doing one more round of testing. And if I still think it's done at that point, then we get it ready for production, and then we get a proof copy. By the time the proof arrives, I'll have been away from the game for a few more weeks or months, and so I approach it again, set it up again, play it again to confirm that it's ready for your table. My games as a designer, and our games as a publisher, generally don't have much errata or post-release fiddling, and a lot of that has to do with approaching the thing very deliberately.
I first work out a skeleton outline of the game in my head. The important points here are the game’s general goals, its outlook, the subject matter and it’s treatment. I then write up a draft and make a rough playtest kit. Then I leave the design alone for a while. I’ll repeat this process a few time and throw in some playtesting solo and with others. Once the design matures, I’ve decided no further modifications are needed, and I’m ready to defend every design decision I read the draft again and again until I’ve read it twice in a row without thinking of any changes that are needed. I also check to see what can possibly be misunderstood by a reasonably experienced gamer. Once that’s done I’m ready to submit the design to a publisher.