Play tested components will be appreciated far more than untested ones, and are more likely to be played by others. No matter how good a design looks onscreen, it is hard to predict how it will perform in practice. Thorough play testing will help uncover subtle issues like strange interactions with rarely used rules. Play testing can’t catch everything, because it only tests the situations that come up in the actual games you test with. But over several games it will uncover the biggest issues. The most important question that play testing can answer, though, is “is this concept actually fun in practice?”
For play testing to be most effective, you need to be able to put your own ego aside. If possible, don’t participate in the game yourself, but observe others as they play. Make copious notes and include enough context that you can understand the game situation later. Is the game slowing down at a certain point? Does it keep happening every turn? Are the players confused or playing incorrectly? Do they seem to be losing focus on the game?
Treat the relationship with your testers with care and respect. They are doing you a favour. Answer any rule questions they have (and think about how to improve your wording!), but do not defend your design. Your goal is to work together to improve the design. This means you need to admit to yourself that it can be improved, because it always can. Thus, your goal is not to get compliments on how great your ideas are, but to discover which parts of the design work. And your goal is not to get into a shouting match about who is right, but to discover which parts of the design don’t work. The role of the testers is to help uncover these facts, not to find or fix the causes. (If they have ideas you should listen and write them down, but their solutions are likely to be surface-level. ) They are the foot feeling ahead in the darkness, looking for the edge of the cliff. Your role at this moment is to listen and record their findings. So when they tell you they feel a cliff, write it down and acknowledge their contribution by saying, “OK, thanks”. Remember that some people lack tact or may not be good at reading nonverbal cues and other social signals. And some people are jerks. When you feel a sting, work to strip the comment of its emotional charge. In your notes, rewrite the comment in neutral language but don’t change the substance.
Once you have enough time and distance to think calmly and objectively, you can start the search for the underlying reasons for the problems and ways to fix them. Your notes often won’t be useful in raw form. The specific events that you observed may only be a sign of a larger problem. Always look for different approaches that will make your design cleaner, more elegant, more coherent, and closer to the feeling you are trying to evoke. One rule change that solves three problems is better than three new rules: look for ways to reduce the total number of rules or special cases. Take advantage of the things that worked well during testing: figure out why they worked and push the things that didn’t work in that direction. If you can’t, rework them or get rid of them.
Think everything is perfect now? Good. Time to playtest it again.
Designs are not an externally valid measure of your worth as a human being. Be brutal. Never hang on to an idea just because you think it is clever and you have fallen in love. It’s an idea, you can’t marry it. And always allow for the possibility that the entire implementation is wrong and that you need to go back and find a completely different approach. The more you are able to acknowledge the flaws in your own work, the better you will be able to make the next version, and the better your subsequent designs will be. Narcissists are crap at design.