Playtesting with strangers

The value of a stranger is that they have no investment in your feelings and no memory of the four earlier versions. Everything they do is a reading of the object in front of them, which is the only thing your eventual players will have as well.

The difficulty is that a badly run session with a stranger produces worse information than a good session with a friend. Most of the method below exists to stop the designer from contaminating the test.

Before they arrive

Write down what you are testing. One sentence, and specific: does anyone work out that the counters can be spent? is a test; is the game fun? is a mood. Write the sentence on paper and keep it in view, because in the room you will be tempted to test everything at once and will learn nothing precisely.

Decide what you will not do. You will not explain the rules beyond the sheet. You will not answer questions during play. You will not sit where they can read your face. Deciding this in advance matters, because in the moment every one of these will feel like ordinary politeness.

What to say at the start

Say three things and then stop. That you are testing the game and not them, so anything confusing is your fault and worth saying aloud. That you will be quiet during play and it will feel rude, and it is not. That you would like them to think out loud, even when the thought is I have no idea what this does.

Then hand over the materials and be silent. The first ninety seconds of a stranger picking up an unexplained game is worth more than the rest of the session, and it is the part designers most often talk over.

What to record

Record events, not opinions. Opinions arrive at the end and are usually a summary of the last three minutes. Events are the data.

Note the moment they first hesitate and what they were looking at. Note anything they pick up and put down. Note every question they ask, verbatim, including the ones you cannot answer. Note where they reread the rules. Note the first time they smile and the first time they sigh, with a timestamp; those two marks locate the working part of the design and its weakest join more reliably than any questionnaire.

Note what they do that the rules do not cover. A rule gap that a stranger walks into on their first play is a gap every player will find.

The questions afterwards

Ask them to describe the game to you as if you had never seen it. What they emphasise is what the design actually foregrounds, which is often not what you intended and always worth knowing.

Ask what they were trying to do on their last turn. If they cannot answer, the game lost its thread before it ended.

Ask what they thought would happen when they took a particular action, choosing an action where you saw them hesitate. The gap between expectation and outcome is where feedback design lives.

Do not ask whether they liked it. They will say yes, and you will have spent your last question on a courtesy.

Reading the results honestly

Treat one stranger's confusion as a hypothesis and two strangers' identical confusion as a fact. Resist the fix that occurs to you in the room; write it down and sleep on it, because the in-room fix is nearly always an explanation added to the rules sheet, and adding explanation is how a design gets heavier instead of clearer.

The strongest response to a confusion is usually to remove the thing that caused it rather than to describe it better. That is harder, slower and correct more often than not, and it is the reason to run these sessions early, while removing things is still cheap.