Everybody designs games in their own way. Of the dozens of designers I've met at school, through work, and online, everybody has their own quirks and methods when they are working on an idea. I think it's important for readers of this blog to know where I'm coming from as a designer and so I've boiled down the general process into a couple steps. These are things that I did without even realizing it before, not something I've come up with ahead of time. They are:
Step 1 - The Idea
This is the first spark of creativity where you get the itch to make a game about something. This could also be called the inspiration step. Everybody comes get their ideas differently. Some people sit down and consciously brainstorm, other people just have it come to them, and other people just do drugs.
At this stage, my ideas come to me as some sort of hypothesis. Frequently, I have something I want to test or present to my players as a test. With a game like Decay, it was in the form of "Can I capture the feeling of being unsatisfied with your life?" With my recent L4D map decay it was much less grandiose, and stemmed from "I want a survival that feels like a real last stand, rather than the survivors being trapped and surrounded." With Mission 16 it was, "What could be a realistic conflict that would arise from the discovery of time travel?"
Not every game needs to start with a hypothesis like this, but that's how I do it. These things just come to me, for the most part. I got the idea for Decay from this tangentially related Calvin and Hobbes comic. When I'm driving, when I'm playing other games, whatever. I don't want to make it sound more mystical than it is, that's just how it works for me.
Step 2 - Flesh It Out
This step is where I start thinking about the answers to the questions I posed in Step 1. I think about what mechanics fit it, what sort of story should go here. I just let my brain run loose in every direction while I'm here, no matter how infeasible or the time constraint. I don't think of possibility here, just how far I can take the idea. To give you an example, with Decay there were parts where I was thinking that the whole game was a massive drug trip, or that the player was in some kind of twisted experiment. We considered having other characters in the game with you, etc. All of which were valid directions to take the game in. This part, for me, is just about filling a notepad with as many ideas as possible to flesh out the hypothesis, this can include raising new questions which even replace the old hypothesis, or modify it significantly.
A lot of people have very specific methods for doing this. They make flowcharts or start working on a design treatment at this point. I don't do that quite here, because that for me limits where my creativity goes. I quite literally will have notes on the back of napkins or scraps of paper or several notepad files on my computer with odds and ends of thought.
Step 3 - Focusing
This is where the practical side of my designer kicks in. Now I start working on a formal design as a concept doc or a design treatment. As I'm putting the ideas on paper I think of which ones really help progress the thesis and which ones are there just because I like them. I start peeling away at the design until I'm left with something that is feasible to build and has a much more focused design than if I had taken everything from stage 2.
That is not to say that everything is locked in at this point, more could be taken away or added, but at this point is where I decide what to start building.
Step 4 - Build It
Once I've got a really solid idea of what i want to do, I can start building it. What engine I use is largely dependent on the idea. I don't start with the engine and say "What idea works in this engine?" I start with my idea from above and say "What engine works for this idea?" The former is very valid, if you're given an engine that you have to use that's an ideal place to start. But when you're in my situation as a student you can be very adaptable. I use what works best, it so happens that most of my ideas have worked well in the source engine, but I've had ideas that would work on anything from The Witcher Engine through Unreal.
Step 4.5 - Iterate
This is really a part of the building phase. Once the game starts coming together, as soon as I have a first playable I get some fresh eyes on the game. I really learned the value of this while making Decay, where we had two playtest sessions through out the developement. We had people check it out after the first level was playable, and then those same people again when the second level was done as well as people who had never seen it before. This feedback ended up changing some of the fundamental design of some areas and puzzles that ultimately created a richer game experience.
This step brings me to what I think is one of the most important parts of being a designer, flexibility. I really think that being able to accept changes to the design part way through a project is key to making a successful game. Nobody gets it right the first time. Painters do it, poets and writers do it, directors do it. You have to be willing to make changes, cuts, and additions as you need. The Mona Lisa has several other Mona Lisa's under it, and probably years of footage have been left on the cutting room floor as well as scenes reshot at the last moment. That is the nature of creativity and you need to be ready to roll with that. Get feedback, and reiterate on your design as much as you possibly can.
So that's it. Get an idea, flesh it out, focus it, build it, and iterate on it. I think most designers go through this process, but everybody does it differently. That's how I do it and I hope that it may spark some of your own ideas.