Most first games die in a notebook. They get written as a setting, a backstory, a list of cool factions, and never become something two people can sit down and play. The fix is not more inspiration. It is changing the order in which you make decisions. This guide walks through that order, from the first idea to the first prototype someone other than you can actually test.
Start with a verb, not a theme
A theme is what your game is about. A verb is what players do. New designers almost always lead with theme: a game about pirates, a game about running a coffee shop, a game about colonizing Mars. Theme is fun to talk about and tells you nothing about how the game plays.
Start with the verb instead. Players bid. Players bluff. Players route trains across a map. Players draft cards and combine them. The verb is the thing your hands repeat for an hour, and it is what people will actually remember.
Pick one strong verb and let the theme attach to it later. Worker placement is a verb: you place a limited number of pieces to claim actions before opponents take them. Push-your-luck is a verb: you keep rolling until you decide the risk is too high. Hand management is a verb: you spend and rebuild a hand of cards over time. Notice none of these need a setting to make sense. Pirates can come after.
If you cannot say what players physically do in one short sentence, you do not have a game yet. You have a mood.
Find the one decision that matters
Every game that holds attention has at least one decision players genuinely agonize over. Find it. This is your core loop, and everything else is scaffolding around it.
A good decision has tension on both sides. Do I take the resource I need now, or block the move my opponent obviously wants? Do I push my luck for one more card, or bank what I have? Do I spend my strong card this turn for a sure gain, or hold it for a bigger swing later? If a choice has one clearly correct answer every time, it is not a decision. It is a chore.
Try to name your one interesting decision out loud before you build anything. “The interesting decision in my game is whether to spend money fast for early board position or save it for the endgame.” If you can finish that sentence, you have a spine. If you cannot, add tension before you add components. More cards and more rules will not rescue a game that has no choice worth making.
Behind the counter at our store we watch players every day, and the pattern is consistent: people lean in when a decision feels close, and they check their phones when it does not. The close decision is the whole job.
Build the ugliest prototype that can be tested
The biggest mistake we see is designers spending weeks on art and graphic design for a game that has never been played. Reverse it. The first version of your game should look like garbage and be playable in days.
Index cards. A Sharpie. Spare cubes from a game you already own. Scrap paper torn into tokens. Write card text by hand. Draw the board on a sheet of A4. Use coins for money. The point of this prototype is not to impress anyone. It is to answer one question fast: does the core loop produce the tension you promised yourself?
Ugly prototypes have a hidden advantage beyond speed. When a card looks finished and laminated, you resist changing it because it cost you effort. When a card is handwritten, you cross it out and rewrite it mid-game without a second thought. Cheap components keep you honest and keep you fast. You want to be on your fifth rough version before most designers finish their first pretty one.
Test it yourself first, playing all sides, just to catch the obvious breaks. Then get it in front of other people quickly, while it is still easy to change.
Playtest with strangers, shut up, and watch
Your friends will be kind, and kindness ruins playtests. They fill in the rules you forgot, ignore the dead turns, and tell you it was fun because they like you. You learn almost nothing.
Blind playtesting is the real test. Hand someone your rules and your components and say as little as possible. Then sit on your hands and watch. Where do they hesitate? Which rule do they read three times? What do they try to do that the game does not allow? When do they look bored? Every confused pause is a design note, and you will be tempted to jump in and explain. Do not. If you have to explain it at the table, the rules do not work yet, and you will not be sitting next to every future player.
Write down what you see, not what you wish you saw. The hardest playtest skill is staying quiet while someone misplays the game you spent months on. That silence is where the real feedback lives. Watch their hands and their faces more than you listen to their words. People are polite about what they say and honest about where they look.
Cut, don’t add
When a game feels flat, the instinct is to add: another resource, another card type, a special action, a bonus phase. Resist it. Adding is easy and almost always makes the game worse. The skill that separates finished games from forever-prototypes is cutting.
Look at your game and find the rule that does the least work. The faction power nobody uses. The third currency that just slows setup. The phase players rush through to get to the part they like. Cut it and play again. Most of the time the game gets cleaner and the core decision gets sharper, because there is less noise around it.
A tight game with one great decision beats a sprawling game with five mediocre ones. Players do not admire your rulebook. They feel the loop. Trim until what remains is only the things that make your one interesting decision interesting. If a component can be removed without anyone noticing, it was never pulling its weight.
Know when it’s “done enough” to show someone
Your game is never going to feel finished, so stop waiting for that feeling. The bar for showing it to people is lower than you think, and the bar for calling it done comes later than you hope.
It is ready to show when strangers can play a full game start to finish without you correcting the rules, when the ending arrives at a sensible point instead of dragging or stopping abruptly, and when people ask to play again. That last one matters most. A second request is worth more than any compliment. It means the loop landed.
It is not done when you are still tuning numbers, balancing one runaway strategy, or smoothing the rough edges of a turn. That work is normal and never fully ends. The trick is separating the two states. “Done enough to show” means the bones are solid and someone else can experience the game you intended. “Polished” comes after, through many more rounds of small adjustments.
When you reach the point where a clean prototype plays well, strangers want a rematch, and you are mostly fine-tuning rather than rebuilding, you have crossed from idea into product. That is usually the moment a designer starts thinking about art, components, manufacturing, and getting the thing onto a table beyond their own. It is also the moment a studio earns its keep, turning a proven prototype into something that can be produced, localized, and put in front of the right players. If you reach that point, that is a conversation worth having. Until then, keep it ugly, keep it close, and keep testing.