Playtesting

Why it matters

Designers are the worst judges of their own games. You know the rules, the intended strategies and what each card is "for", so you cannot experience the game fresh. Playtesting is how you find out what the game actually does to people, as opposed to what you hoped it would do. Professional studios playtest constantly, from paper prototypes to finished builds.

The core idea: observe behaviour, then ask

The most reliable data in a playtest is what people do: where they hesitate, what they misunderstand, when they lean in and when they check their phone. What people say is useful too, but less reliable: testers are polite, guess at causes, and suggest fixes that may not be the right ones.

A useful rule from many designers: players are usually right about where the problem is and often wrong about how to fix it. Listen to the "where"; design the "how" yourself.

Plan the test

Before anyone plays, decide:

• Goals: two or three questions the test should answer. "Is it fun?" is too broad; "Do players understand trading by turn 3?" is testable.
• Testers: at least one person who has never seen the game. Friends and family are fine early, but they are kinder than strangers.
• Setup: materials ready, rules sheet printed, a notes sheet with columns for time, behaviour, quotes and interpretation.
• Your role: host and observer, not teacher or salesperson.

Run the test

1. Brief: "I'm testing the game, not you. Please think aloud. I'll mostly stay quiet."
2. Watch: let them struggle a little. Write down exact quotes and timestamps.
3. Intervene only to unblock: if they are completely stuck for minutes, give the minimum help and note that you had to.
4. Debrief with open, neutral questions:
   • "Talk me through what you were thinking when you chose to trade."
   • "When were you most and least engaged?"
   • "If you could change one thing, what would it be?"

Avoid leading questions such as "Did you like the trading system?" They invite polite yes answers.

Worked example: from notes to changes

Raw notes from a test of the dungeon prototype:

Prioritise by frequency and severity:

1. Ending drags (both testers, big impact) -> try: the exit opens once 3 treasures exist in total.
2. Mid-game risk too high (one tester, medium) -> try: oil cards more common in rooms far from the start.
3. Light rule unclear (both, small but easy) -> fix the rules sheet.
4. Keep: the monster card moment.

Change one or two big things for the next test, so you can tell what worked.

Types of playtests

• Internal tests: the team plays; good for broken rules, bad for "is it understandable".
• Friends and family: quick and easy; kind feedback.
• Blind tests: new players learn from the rules alone; best for clarity.
• Targeted tests: players from your intended audience; best for appeal and difficulty.

Common mistakes

• Explaining and defending. If you argue with feedback, testers stop giving it.
• Leading questions. "Wasn't that cool?" tells you nothing.
• Testing without goals. You come away with a pile of opinions and no decisions.
• Taking every suggestion. Look for patterns across testers; one person's idea is a data point.
• Waiting until the game is "ready". Test the ugly prototype; big problems are cheapest to fix early.

Quick recap

• Playtesting shows what the game actually does to people.
• Observe behaviour first, ask open and neutral questions second.
• Plan goals, take timestamped notes, and stay quiet unless a tester is completely stuck.
• Prioritise problems by frequency and severity, and change one or two things per round.

Try it

Open the Do tab and run a short playtest of your prototype.