Lesson 16 · Part 5: Making It
Prototyping
How to answer a design question in an afternoon with paper, a cardboard box, or somebody else's game.
Prototyping
You and a friend are inventing a card game, and you disagree. You think each player should hold five cards. Your friend says seven. You could argue about it all evening, and you’d both have perfectly good reasons.
Or you could deal five cards each and play for ten minutes, then deal seven and play for ten more. Twenty minutes later nobody’s arguing, because you’ve both felt the difference. Seven was too many to keep track of, and everybody kept forgetting what they were holding. Question closed. On to the next one.
That little experiment is the single most useful habit in game design. Designers start doing it the moment they have an idea, and many of them will tell you the early rounds are the most important stage of the whole project.
A question you can touch
A prototype is a small, rough, usually playable piece of a game, made to answer one question. Should a turn be one move or two? Should the board be a grid or open space? Does this drawing style suit the world better than that one? You can’t settle questions like these by thinking hard. You have to touch something, try it, and see how it feels.
So you take that one piece and make the simplest version that lets you decide. Then you play with it while imagining the rest of the game around it. The rest doesn’t need to exist. One token on a blank sheet of paper is enough to tell you whether moving diagonally feels good. A single test cupcake tells a baker whether the recipe needs less sugar, long before anyone bakes two hundred of them for a wedding.
“We’re going to build the simplest possible thing we can that will help us answer that question.”
The output of a good early prototype is one clear sentence: we go with A. Back in Level 2 you met the design tree, the huge spread of branching decisions between a vague idea and a finished game, and pruning, the job of cutting branches off. A prototype is how you prune. And here’s the tough-love part: if a prototype can’t tell you which branch to take, it failed as a prototype, however nice it looks.
Two things follow from that. A prototype should be cheap, measured in hours or days, because its only product is a decision. And you should know your question before you make anything. “Let’s knock something together and see” tends to produce something that answers nothing.
The questions change as a project grows up. Early ones have crisp answers. Later ones ask whether a newcomer can figure out how to play, and later still whether people find the thing absorbing. Those answers are fuzzier and need more interpretation, which is what Level 18 is for.
Cheap comes in many shapes
Beginners assume prototyping means programming. Often it means scissors.
A paper prototype is any prototype that lives off the computer. You cut up slips of paper and call them armies, or farmers, or fish. You write a rule that a piece moves two spaces a turn. You and a friend push them around a hand-drawn grid, and within minutes you know whether the basic contest has any tension in it. The name is loose: coins, toothpicks, buttons, and bottle caps all count. Tailors have done this for centuries, sewing a test garment out of cheap cotton before cutting into the expensive fabric. Nobody gets attached to the test version, and that’s what makes it useful.
A second kind tests how an idea lands with people. Some designers make two or three different mock box covers for the same game, set them on a shelf, and ask friends which one they’d want to open. Every box holds the same game. Each one describes it differently. If everyone reaches for box B, you’ve learned something about how strangers first understand your idea. That’s a communication prototype. It tests the first impression and leaves the rules out entirely. Publishers do the same thing when they mock up several book covers before choosing one.
The cheapest kind of all is one you don’t make. Say you can’t decide whether your game should be played in turns or with everyone moving at once. Someone has already made games that do both. Spend an afternoon playing a slice of each, paying attention only to that one question, and picture your own game in their place. That’s an off-the-shelf prototype: a borrowed piece of someone else’s work that answers your question for the price of a board-game night.
Throw it away
When a prototype does need software, most designers don’t care what it’s made with. Use whatever lets you move fastest. You’re going to play with it for a few hours, make a decision, and toss it. Sometimes a rough version grows into the next prototype. Either way, nobody polishes it. Polishing a prototype is like ironing a paper napkin.
From sweet spots to a spine
Make enough prototypes and a few will stand out. One action is just plain pleasant. One moment makes everyone at the table grin. Those are sweet spots, and the next move is to ask how far you can spread them through the rest of the design.
Spore is a good example: it’s a computer game where you guide a species from a single cell all the way up to flying spaceships. It was in real danger of feeling like five different games stapled together. Early on, the team made a prototype tool for building your own creature, and it was so much fun that they tried one for buildings, then one for vehicles. All of them worked in much the same way. Those building tools became the spine of the game, the core activity that runs through every stage and holds everything together.
A spine pays off twice. A player who’s learned to build creatures already knows most of what they need to build vehicles. You see the same trick in a deck of playing cards: learn the four suits and the ranks once, and you can play hundreds of different games without learning anything new. Call it design leverage, getting a lot of game out of a little teaching.
Notice that a prototype is also a cousin of the tangibility from Level 15. It’s your idea made into something you can grab.
The slice of cake
Eventually the questions get bigger than any one piece can answer. Does the whole thing work together: the look, the sound, the rules, the feel? To find out without building everything, teams make a vertical slice: one small piece of the game, finished properly through every layer, top to bottom.
Think of a layer cake. You don’t need the whole cake to know whether the recipe works. One slice shows you every layer, the sponge, the jam, the cream, the icing, all stacked the way they’ll really be. A vertical slice of a board game might be one full round, played with finished cards, real art, and the final rulebook wording. If that one slice is good, you’ve got real evidence the rest can be good too. If it isn’t, you’ve found out before baking the other ninety-nine slices.
A vertical slice comes later than the scrappy prototypes, and it costs more. It’s still a prototype, though. You’re still asking a question, it’s just a bigger one.
What this means for you
You don’t need permission, software, or special skills to prototype. You need a question and twenty minutes. The discipline is to keep asking what you’re trying to find out, and to stop building the moment you know.
Most of what you make this way will get thrown out. That means it’s working. Every discarded prototype is a branch you never have to wonder about again, and the next level explains why a big pile of them is one of the best signs of progress a design can have.
Try it · Pocket Game, step 16One question, twenty minutes
Write down one thing you’re unsure of in your Pocket Game, phrased as A or B. “Should a turn be one move or two?” “Should the pieces start in the middle or at the edges?” Set a timer for twenty minutes and make the cheapest thing that answers it. Often that just means playing two short rounds, one each way, with a friend or against yourself. Write the answer in one line. If you can’t write it, your question was too fuzzy. Sharpen it and go again.
Carry forward: Know the question before you build, and stop building once you have the answer.