Prototyping

For a first prototype, do not spend any more than the absolute minimum amount of time on the design. It just needs to be playable. That's a big enough job as it is.

Game design is a process of experimentation with the rules. It is not a type of handicraft.

Everyone overdoes their prototypes, early on. I did. Beginners (including me) create prototypes as though they're creating a finished game. Experienced designers know that this is just the first revision of a hundred.

You're allowed to have fun, and just make things look nice, for no reason. I do that sometimes. I just don't tell myself that I'm actually doing something useful.

If a game survives its first playtest, I begin gradually upgrading the visual appearance, with each revision. The game ends up neat and presentable, with nice icons, appropriate art, and good typography.

Do not do art or graphics for something until it survives one revision.

For my designs, there's a greater than 50% chance the game is never played again after its first playtest.

I've spent weeks creating and printing/uploading a game, only to play it for a few minutes before abandoning it.

Even if a game survives, all the original design will be overwritten by the end of the process. No design or icon created in a first prototype has ever survived to the end of one of my game designs. (Even when I made very fancy first prototypes.) Don't waste your time.

Take the view that no revision is ever the last.

First prototypes

My first prototypes look very basic, and unappealing. They're usually just a bunch of plain cards with a title, some quick text, and primitive artwork.

I've found a great new source of early prototype icons: emojis! They're not really thematic, but they're clear. Also, they're text, so they're very easy to use. You can cut and paste them, and they flow in line with the rest of the text.

You do need card art, and by "art" I mean anything that's not text. Players do have to be able to differentiate between the cards. The cards shouldn't just be text. If your card just has a few big icons, that's fine. Make them huge, and they can be the art. Use other icons as art. Use art from another game. Use anything.

Do not create a whole game

In the first playtest of your game, you will learn very important things within a few turns. There won't be much point playing past that point. Then, you'll be back to the drawing board, and lots of things will change. (Or, if you're good, the game might go in the bin.)

You're going to be playing this game for ten minutes.

If you've created an entire game, you'll have to alter the whole thing, repeatedly — even the parts you're not using.

Create the first ten minutes of the game, if that's possible. If there are rounds, only create the first round. Just make enough of each component to enable that. This means extremely small decks of cards, which you just recycle if they run out. A deck of ten cards isn't very random, but it's enough to play the game. It doesn't matter that the game isn't random. You'll only play that deck once, and then you'll change it.

Create the most normal and boring actions you can think of. You want the first playtest to be "normal".

Only create the two-player version of the game.

If the game has a board, create a small corner of the board. Do not make the board modular or variable.

If there are different "characters" the players can be, just create two. Even better, leave them out. Every "extra bit" that the game doesn't necessarily need, can be left out completely.

Keep the game very small and non-variable, for many revisions.

Resist the urge to make content. Resist the urge to make content!

The fact that every game will be samey is actually a good thing. When the game goes bad, in playtesting, you know it's the rules, not just an unlucky draw or one wacky card.

Do create the full breadth of the game, or most of it, so you're really testing the proper game. The game should be very short, but it should be representative of the full game, and not be a simplified version of it.

Obviously, don't worry about balance, fairness, replayability, or any other such issues early on.

The game doesn't need a proper end, until you're many revisions in. You can often leave the objective out. I often tell players to "just do stuff." Or, you can just create a small, temporary win condition, like "gather 10 resources."

Make your boards and reminder sheet too big, and leave blank space. Constantly rearranging and resizing things, to take up all the space, is a colossal waste of time. I'll sometimes have dice with a blank face. I just tell the playtester to reroll.

Don't make things perfect each revision. Just solve your biggest problems, and leave everything else as it is. If you can't solve something, just leave it there for a while. Many unsolvable problems will be fixed when the rules of the game change.

My decks often contain obsolete cards.

In my pirate game, if I thought of a cool new die face, I used to create a nice new icon, and write an explanation on the rules sheet. Now, I just throw black text onto the die face itself, that says "MOVE BETWEEN HIDEOUTS" or "SELL CUBES FOR 2 FISH". If these work, I'll make nice icons and proper rules later.

Rules

Rules should become consistent and clear later in development. Do not worry about edge cases or interactions until then. Those rules will likely change anyway.

Just arbitrate things during the game.

Just create a simple rules card. Players know that numbers in corners are probably costs, and that played cards are probably discarded. A rules/reminder card also lets you know what the game was about, when you look back at it, if you take a break from it.

Don't bother with the exact theme yet, and just use placeholder names like "Wizard" or "Factory".

You can do all this creative stuff later. It's still fun.

Physical testing

A game will be experienced physically, once it's printed. Many publishers will therefore create (or demand) a physical version of the game, to match that effect.

Honestly, I don't waste my time. Most games are physically similar to other games.

I've never created a physical copy of my farm game. There are plenty of other farming games where you move tokens around, and have your own tableau. They work fine. However, my game has trees, that are stacks of tokens. I did have to get out my game pieces, and see how manageable stacks of tokens were.

If you're investigating how much work your game creates for the players, getting out some bits is also useful.