How to prototype a card game from a spreadsheet

A prototype's only job is to be wrong quickly. Everything below exists to shorten the distance between changing your mind about a number and holding the card that number is printed on.

Your game is a table before it is a game

Almost every card game is, underneath, a table. One row per card; columns for the things a card has — a name, a cost, a type, a line of rules text, how many copies are in the deck. Designers reach for a drawing program first because cards are visual, and then spend the next month moving text boxes instead of balancing costs.

Open a spreadsheet instead. It is the only view in which you can see all sixty cards at once, sort them by cost, notice that eleven of them do roughly the same thing, and fix the game rather than the layout.

The columns you actually need

Start with fewer than you think. A workable first sheet is six columns:

• name — what the card is called.
• type — the category the rules care about.
• cost — or whatever number gates playing it.
• text — what it does, in one sentence you would say out loud.
• copies — how many of it are in the deck. This one saves you from pasting the same row eight times.
• art — a filename. Leave it empty for now; a grey box plays exactly as well as a painting.

Design one card, not a hundred

The layout is a separate thing from the data, and keeping them separate is the whole trick. Draw the card once — where the name sits, where the cost sits, how the rules text behaves when it is long — and say which column feeds each place. Then the deck draws itself, and a change to the layout is a change to every card at once.

In Boardsmith that link is a token: put {name} in a text field and it reads the name column. The card in front of you redraws as you type, so you are judging a real card instead of imagining one.

The first print run is paper and scissors

Do not order anything. Print the deck at true size on ordinary paper, cut it with scissors, and if you want it to shuffle, sleeve the slips in front of cards from a game you already own. That is a real prototype and it costs nothing.

True size matters more than it sounds. A deck printed slightly small feels wrong in the hand, and you will spend a session blaming the rules for it. Boardsmith's print-and-play PDF lays cards out at their actual millimetres on A4 or Letter, with cut marks — and that part is free, permanently.

What the first playtest is for

Not "is it fun". Nobody can answer that about a grey-box prototype, and asking puts your friends in the position of having to be kind. Ask narrower questions:

• Which card did you never want to play, and why?
• Where did the game stop and someone have to ask how something works?
• Did anyone know they had lost long before the game ended?
• What did you reach for that did not exist?

Then watch instead of listening. Where hands hesitate is where the rules are unclear.

Keep the loop short

This is the part that decides whether a game gets finished. After a session you will have eight changes: four costs, two rewordings, one card deleted, one added. If applying them means editing eight layouts by hand, you will do it twice and then stop.

With the data separate, it is eight cells. Save, export, print the pages that changed. The prototype you can revise on a weeknight is the one that gets to version twelve, and version twelve is where games start being good.

When it survives

Two things happen to a prototype that keeps working, and it is worth knowing both are a short step away:

• It goes online, so you can test with people who are not in your kitchen — the same spreadsheet becomes a Tabletop Simulator deck.
• It goes to a printer, on real card stock with a real finish — which needs bleed, DPI and one file per card.

Neither is a rebuild. The table you started with on the first evening is still the deck.