I Suppose You're Wondering Why I've Called You All Here...

There may be spoilers for Demon God's Fane in this article. Reader beware.

They gathered in the gaming room, taking their places around the black table. Miniatures were arrayed atop the vinyl battlemat. Dwarven Forge dungeon pieces were arranged, at the ready, to either side of the large, homemade GM's screen. My Monster Manual was already open to one particular page -- the one with "Demon" written near the top.

This was going to be good.

Playtesting adventures isn't like playtesting rules. The goal here isn't as much to look for balance (although that's not ignored) as to look for playability. What makes something playable? Here's my short list:

1. Can the players follow the clues or cues to keep things moving?

2. Are descriptions and explanations adequate to interest the players and inform the GM?

3. Does the adventure devote enough detail to the things the players focused on during play?

4. Are the encounters structured in a way that makes it easy for the players to rest or retreat when they need to?

Now, those aren't nearly all the criteria for a good adventure -- just the things that playtesting can help a designer test.

In the playtest for Demon God's Fane, we had a nice mix of 14th-level characters. Chris Perkins played a human fighter, Bruce Cordell a halfling rogue, Jeff Quick a human cleric, and Sean Reynolds a diviner (based, apparently, on some old girlfriend of his -- I don't know, go to his site). Normally, I'd be leery of testing an adventure with a divination specialist, because it's not really a strong school, but for this adventure, with all its information-gathering aspects, I thought it would be a good test to see how the mysteries hold up to magical mystery-bashing.

The players came to Golden Lake to find a magic item that an old book told them was in the area -- one that they needed to help revive a cursed friend. When they started dealing with the events in the village (yes, I am being a little coy with facts, here -- I figure plenty of people might have missed or ignored the spoiler alert above), they used a mix of magic and interaction to deal with people and circumstances. To start with, they raised all the murdered victims. Not what every group would do, I imagine (that's a 2,500 gp bit of charity), but interesting -- and not a bad way to learn who the murderer is.

They used magic (detect thoughts) to pry into the minds of a number of the villagers to make sure that they were getting the whole story. Since some of the NPCs in town are strange characters, this led to a few light-hearted and humorous moments. When dealing with some of the more difficult NPCs, the (slightly off-kilter) wizard dominated one of them (Gesse Anatham), just to get her off their backs, which everyone also found pretty funny.

Which brings me to another good reason to playtest an adventure: To see whether it's fun. How do you test for fun? Well, beyond the obvious, you can look for these things:

1. Is there a mix of encounters, so that people who like combat, people who like interacting with NPCs, people who like solving problems/mysteries, and people who like to make plans are all happy?

2. Is the story gripping enough to hold the players' attention?

3. Do the players get to use their characters' powers in fun and useful ways (thus rewarding them for the character choices that they made)?

The first two of those points are pretty obvious, but a lot of designers miss the third one -- in fact, some actively design counter to it, making all their encounters in areas where the characters can't use their cool abilities. The whole first section of Demon God's Fane, for example, is a murder mystery. Now, many people know that mysteries can be spoiled by a lot of D&D spells -- but in this adventure, that's okay. Make the cleric happy he prepared zone of truth and commune, I say. Reward the wizard for thinking to ready detect thoughts (but don't be afraid to prepare a few surprises for them too -- if you've read the adventure, you know what I mean).

Ultimately, the characters came out all right. There was a single death (for a moment it seemed like two, but we managed to save one character by double-checking the range of the veritable volley of bodak gaze attacks. (In fact, there's a note in the module about that very fact for that very reason -- Bruce Cordell's rogue survived because of we checked the range again). The playtest also influenced other things in the adventure, too. A lot of the beginning notes regarding the Fane and its defenses reached their level of clarification after the playtesters assaulted the place. Exactly what victims of the Umbral Taint do and do not know, and what the murdered victims themselves know (if addressed with a speak with dead spell or if raised) was added to the adventure as a result of the playtest.

Really, what playtesting does for an adventure is help the designer cover all the bases -- it helps him remember all the little things. And it's the little things that can make the difference between a good adventure and a great one.