The Stages of Game Development

So, you want to design tabletop games. You have an awesome idea for a new game? Great, you might be at the first stage of fifteen to getting that game into publication! Here's how I think of the different stages of a tabletop game:

1. The elevator pitch. This will come to me at random times, and is the idea that so many people think is the most important part of the process. It's not, of course: ideas are easy, execution is hard. But, this is also the bit where the ball gets rolling. "Dungeon crawling with a D6 system, using the dice as markers", "Texas Hold 'Em, with spells", "Adventure board game, but you're the people hiring the adventurers, not the adventurers themselves", and "standard playing cards, but with information on the back" are all examples of games I've been involved with. At the end of this stage, about all that exists is a note somewhere in a document and whether or not I keep turning it over in my head.

2. The description of play. During this stage, I've taken something far enough to have an idea of some basic mechanic. This isn't a rules set by any stretch, but it's simply the basics of how you accomplish something in my design. It's probably not even fair to call it a game yet, it's more just a concept of how state transitions might occur. It's at this stage that I start to get an idea for what resources are available to players, how some of those resources are expressed and manipulated, and at least some idea of what the goal is. I'll sometimes trot out some bits and pieces at this point, roll a few dice, but I won't always do that physically. Usually things are still simple enough for me to just visualize things. At the end of this stage, what I have is a larger note somewhere in a document, which is probably standalone at this point.

3. Proto-prototype. During this stage, I've got something that is arguably a game. I've probably fleshed things out enough to have a very rough sequence of play, some really primitive cards, maybe a crude map. I've put just enough effort into things to decide if there's anything even vaguely resonating with the design. While I can eliminate a lot of dumb ideas for games long before this stage, usually because I can't conceptualize how to turn the elevator pitch into something like a description of play, many designs reach this stage without any real idea if it's going to be at all fun. What seems like an elegant, clever manipulation in my brain turns into boring drudgery on the table. Or the solutions are blindingly obvious. Or something else. What I'm looking for at this point isn't so much what works, but to disqualify a design. If I can't see where the potential fun is, there's no point in proceeding any further. At the end of this stage, what exists is a very, very rough pile of components that are useless to anybody that doesn't have direct access to my brain. Things fail frequently at this point.

4. First draft prototype. During this stage, I have a thing that can be accurately described as a game, for the first time. I've put down enough rules on paper to feel like they're a relatively complete description. There are enough parts defined that you can set up the game and take turns (or whatever). Crucially, there is some kind of endgame condition expressed in the rules, which makes this something other than undirected wandering (which the previous stage often is). This isn't a prototype that I'm comfortable showing to other players yet, necessarily, but instead something I can set up solo and execute some turns on. The goal of this stage is to take the learning I made in the previous stage and see if it tests out with more structure. Is the fun still there? Does it still seem like the original idea can be expressed into a structure like this? Can I see the outlines of a real game beginning to emerge from the primordial game-stuff? For me, at least, this stage involves a fair bit of computer prototyping, as I'll whack together some scripts in Python just to test some probabilities to test if my ideas are at all sound. This isn't true for everybody, but this is often a time savings for me. At the end of this stage, I have a rules set, some super basic components, and a subset of the final game's content.

5. Second draft prototype. At this stage, I have a thing that I'm willing to show to my best gaming friends, people who will continue talking to me even after I hatch a dumb idea on them. This is even easier for two-person games, because I only have to hassle one other person. I've gone over the rules based on what happened with my first draft, and fixed any of the truly glaring stuff that I discovered. I've created enough content in the game to give the game a fair shake, even if I think most of that stuff will likely change later. I've got components in enough of a form that someone other than me can make heads or tails of them at the table. The goal for this stage is simply to see if other people can see the same spark that I can see in the design. Everybody at the table knows that this thing is far away, but can everybody also recognize the potential? The first outing for this game likely won't terminate as the rules specify, since there is probably something sufficiently jacked up that I'll need to call it off early. Nevertheless, I can see what people are reacting to, what they're trying to do, what they like, and what parts are making them crazy. I'll end up with a huge list of things to fix, and I'll possibly just shelve the thing at this point if things look really hopeless. At the end of this stage, I have a set of stuff that I can keep in a box or baggie and plausibly call a game, complete with placeholder designs on components.

6. Rapid rules iteration. At this stage, if I've made it this far, I have a thing that a group of my good friends think is worthy enough of my time to keep fiddling with. This might actually be the most fun stage in the whole process. I'm making changes very rapidly, and I can frequently move from "total mess, with some promise" to "hey, that's pretty good" pretty rapidly. A handful of playtests is sometimes all I need to quickly zero in on something playable and decent. If my game is being difficult or just has a lot of moving parts, this stage can take longer. But, regardless, I can move things along really fast compared to previous stages. I'm changing things quickly in all parts of the design, and most of my stuff makes the game better, because there aren't that many parts I have to worry about making worse. My regular gaming group has probably played things quite a bit at this stage. At the end of this stage, I have more refined stuff in a box, and maybe some initial thoughts about functional design for components, in response to things encountered at the table.

7. Rapid content iteration. At this stage, I've got the rules far enough along that I'm feeling pretty comfortable about what sorts of content I'm going to need. It's time to take a stab at most of that stuff. I'll set some initial limits for how much stuff I'm planning on putting in, I have a good idea of what the definition of each piece of content is, and it's time to strap on my thinking cap and fill that stuff out. My aim here is to create enough stuff in the game that I'll have an idea if the scale of things I have envisioned is going to bear out. If my idea calls for 300 monsters to be created, I need to know if that's realistic. Can I make that much stuff? Can I reasonably expect to balance that much stuff? If I can create a subset of my content without too much trouble, it gives me some idea of how hard the full set is going to be (remembering potential network effects, of course). I want to feed as much of this content into my home playtest group as possible so I can get an idea of where the trouble spots for content generation are going to be. At the end of this stage, I have stuff in a box that is more voluminous than it used to be.

8. First outside playtest. During this stage, my home gaming group has expressed the thought that my game is in pretty good shape. It's time to get an outside opinion. To get to this stage requires a rulebook that has enough in it that the game can be learned without me in the room. The content and materials need to be complete enough that players can play things. The content needs to be complete enough to be able to call it at least the basic game - in other words, it should be a sufficient quantity of stuff that I'd feel comfortable with shipping that amount of content even if nothing were added. This is, in short, a complete game, albeit one that still needs a ton of balancing, re-writing, and tuning. My first outside playtest absolutely doesn't need to be a bunch of strangers. It should probably be to a group that I know, that trusts me not to waste their time, and one that I know is going to provide solid feedback and not just ego-stroking (or existential horror). Is the affection my home group showed for the game just driven by my own energy and passion (and their affection for me), or can the game stand on its own merits and entertain an outside group? I don't need a bug free product here, far from it, but I need to believe that at this point my game's merits will be evident to the world. At the end of this stage, I have a complete game that an outside group can take and play. It might still be pretty spartan, but it's functional.

9. Rapid rules iteration 2. The exposure to outside play groups will cause another explosion of feedback, as I encounter different ways of looking at the design. Tons of rules questions are going to pop up on things that were perfectly clear to my home group. Pieces of content I thought were pretty good are going to trip people up and/or annoy them. Subsystems that I thought were in good shape, which were not causing any trouble, are going to spring leaks. It's ok! This ends up being another pretty fun phase, because by now, I have a strong feel for the design and what works, and I can usually patch most of this stuff up pretty quickly. I'm aiming to make my outside play group happy with the game, and I only get so many shots with them, but my rapid improvement here can make that happen pretty fast. At the end of this stage, I have a complete game that I'm comfortable with exposing to even more outside groups.

10. Multiple outside, friendly playtests. This stage involves discovering all the amazing ways my rules can get misread. I'm going to re-write all of them, and all of my content while I'm at it. I basically assume that I'm going to have to write twice as much content as the final design calls for. I'm going to go around in so many circles that I'm pretty sure that I'm never going to get past this stage. I'm pretty sure I've made a good game, but it's trapped under the cumulative confusion of a couple dozen gamers attacking it at the same time. If I persevere, remember that these are groups that are inclined to like me and be helpful, and keep taking their feedback and improving, I'll notice a huge amount of improvement in my game. At the end of this stage, I have a complete game that is ready to go to strangers.

11. Content iteration. During this stage, I have to get my ducks in a row on content. I've got a lot of confidence in the rules and the content I've already created, but I haven't necessarily finished all the parts. Since my rules are pretty baked, it's pretty clear how to fill out the remaining material now. I've been working on the game long enough that I have a good feel for things, so filling out the rest shouldn't be too hard. At the end of this stage, my game is complete.

12. First blind playtest. During this stage, I'm looking to see what the reaction would have been if I had released that last stage into the world. These are going to be folks that aren't necessarily inclined to see my game in a positive light. What's the reaction going to be? Will people have fun? Are people going to just say "I'd rather be playing [x]"? This is a real acid test for me. It's at this point that I actually like to have somebody else running the playtests - a developer if I've got one, a local volunteer who doesn't mind collating feedback if not. This helps the blind playtesters give honest, unfiltered information that they might hold off if they knew the designer was directly watching. The goal here is to try and ascertain what the remaining problem spots in the design are. You can't really count on blind playtesting groups to last for the long haul, so each blind playtest is pretty precious. Make great use of them! At the end of this stage, I have a game that is very close to a releasable state.

13. Remaining blind playtests and polishing. During this stage, I've fixed the stuff from the previous stage. Most of the issues at this point are probably more around the physical designs of components and other similar concerns, and not fundamental issues. If they are, I've really screwed things up. I'm attempting to really put a bow on things. I'm trying to clear up the last little bits of troublesome content, reword that sticky rule, fix the layouts of cards, and otherwise just perfect things. This kind of polishing is painstaking and difficult, but this effort really shows in the end product. At the end of this stage, I have a game that could be released, if the art and component design was complete.

Publication is its own set of steps, of course. Getting the physical systems right can be huge, and some of those steps should ideally be executed in parallel with the rest of this stuff. Blind playtests are crucial for hamering out the design of components as well as the systems, and you need to perform blind playtests with the final components just as much as the final rules. But those stages are for another day.