OK, quick definition time: ‘playtesting,’ in roleplaying game (RPG) parlance, refers to the practice of taking an RPG adventure or campaign and trying it out before it makes its debut as a published product (or at a con game, or something else that’s similarly more formal than a Saturday night game session). Sometimes you’re testing things to destruction; sometimes you’re just making sure that everything will run as smoothly in real life as it did on paper. Either way: IF YOU SELL RPGS FOR MONEY, PLAYTEST YOUR PRODUCTS OR I WILL GLARE AT YOU. Also: if you don’t playtest stuff that you’re planning to sell for money, you may very well be reincarnated as a newt. And if that happens, the rest of us will laugh at you—and you’ll deserve it.
Moving along, there are two ways to participate in a playtest: either you’re going to gamemaster (GM) it, or you’re going to play in it. If you’re playing in it, your job is going to be relatively straightforward. The GM will—or at least should—tell you what stuff needs to be checked on or checked out; and then it’s your job to do that checking. Aside from the usual fully-socialized human stuff (don’t be rude to other players, don’t hog the spotlight, don’t be a jackass generally), you should also probably be a bit more cavalier about your character’s chances for survival than you would normally be. After all, the purpose of player-characters (PCs) in playtests is to be used up. If the PC gets used up too soon in the game, tell the GM and have him make a note of that. If the PC breezes through the game easily and without worries, have the GM make a note of that, too. After all, it’s hardly an adventure if there’s no possibility of failure, right?
If you’re a GM, on the other hand, you have some more prep work to do. First off: if you’re playtesting the adventure for somebody else*, obviously you should read whatever texts and game materials the writer has provided you. If there’s something missing, make a note of it. You’ll be making a lot of notes, by the way: coming up ahead of time with broad categories of the types of notes that you make, and keeping all the notes about one category in one place, will save you time later.
Once you’ve gotten the notebook and pencils that you’ll need for your notes, the next step is to get players. Always get more players than you need, because people often can’t make it to a playtest after all. If you end up doing a playtest with more people than you need, have the extras run adversary characters, roleplay out non-player characters (NPCs), or simply hang out and kibitz. This last can often be surprisingly useful; people who aren’t in the middle of play can sometimes see stuff that the actual players cannot.
Speaking of kibitzing… as a practical matter I suggest that you don’t let discussions over rules interpretations last for too long. If something’s confusing, and something is always going to be confusing at one of these things, see if it can be resolved in a minute. If it can, make a note in the “This needs clarity” part of your notebook, and keep playing. If it can’t, make a snap one-time judgement, make a note in the “This needs fixing” part of your notebook, and keep playing. Your job isn’t to fix the adventure, it’s to find the stuff that may need fixing. Another practical tip: when you solicit feedback from the players, just write down what problems they had with it and don’t try to explain those problems away (whether you wrote the adventure yourself or not). Making changes based on feedback is a completely different issue and article.
Finally: if you do a playtest, and send over copious notes, and some of those notes do not provoke changes in the final edition of the game or adventure… then they do not, that’s all. It’s not a personal slight on you or the other playtesters. In the end, the author of a game is the final arbiter over which advice he is going to accept, and that is the way of it. If that rankles, why, the solution is simple: go and write your own adventure, and have someone else playtest it for you. You will no doubt gain a remarkable amount of perspective about the whole process; and in record time, to boot.
*Whether or not you should playtest your own adventures is, in my opinion, dependent on what the final venue is. If you’re testing to make sure that your convention game is going to play out in less than four hours, then by all means: run the playtest yourself. If you want to see whether something you’ve written will fall apart when it’s actually played, then: get somebody else to do it.