From my ENWorld blog:
Musings on Skill Challenges (or: Three Questions You Should Ask Before You Run One)
First of all, here's a quote from a post I made [around the middle of April 2009], and which kind of sets the tone for the rest of this post:
To me, skill challenges strike a middle ground between a very free-form problem-solving approach which is almost entirely dependent on player skill/DM adjudication (e.g. solve this mystery) and a very mechanical, rules-defined approach to tackling common (but specific) problems (e.g. opening locks, noticing secret doors, finding and removing traps, following tracks).
Properly used, the skill challenge framework can give you the best of both approaches. An inventive player can come up with creative solutions to the skill challenge, or novel ways to use the skills he is good at, and the DM can allow the use of those skills to score successes, or even award successes without the need to make a skill check. On the other hand, a player who is more comfortable working within a fairly well-defined framework can simply run through his character's list of trained skills and pick one that seems appropriate to the challenge.
So, before you run a skill challenge, the first question you should ask is:
1. Should you use a skill challenge in the first place?
Just because you expect the PCs to make skill checks at a certain point in the adventure, it doesn't mean that you need to run it as a skill challenge. In my view, before you decide to make a series of skill checks into a skill challenge, you need to consider the Consequences and the Constraints.
That succeeding or failing at a skill challenge should have significant Consequences either way might seem so basic as to be obvious, but DMs sometimes fail to make the players care about whether or not they succeed. Sometimes, it is because the difference between success and failure is small. At other times, it is because the players think that the difference between success and failure is small. DMs should ensure that the Consequences of succeeding or failing a skill challenge are worth the time and complexity of running one, and find some way to plausibly communicate them (or at least hint at them) in play.
Skill challenges should also have some Constraints in order to be interesting. If the character with the highest modifier in the most relevant skill can just keep making skill checks until the skill challenge is passed or failed, then that portion of the game should not be run as a skill challenge even though it is technically possible to do so. Ideally, a skill challenge should have Constraints such as a time limit to encourage as many PCs to participate as possible (but more on this later), or should require the use of three or more different skills to overcome so that more PCs have to get involved.
The next question relates to the old post that I quoted:
2. Do you really need to ask for a skill check?
Your answer to this question is likely to be closely related to your views on whether the player of a low-Intelligence PC should be allowed to solve a puzzle in the game, or whether the PC needs to make an Intelligence check first.
Nonetheless, if you are inclined to allow player skill and creativity to solve problems in your games, the presence of the skill challenge mechanic should not prevent you from doing so. Feel free to allow players who come up with good ideas to overcome skill challenges without rolling for skill checks - just as if the skill challenge mechanic did not exist.
If you want to give clever and creative players an advantage, but are not prepared to let them solve the problem without rolling dice, then the granular nature of skill challenges (in that a certain number of "successes" are required before the skill challenge is overcome) allows you to reward good ideas with more successes if the PC makes the skill check, or even automatic successes (less than the total number of successes required, if you do want the PCs to make some skill checks).
Of course, none of the above is very useful if you are faced with a bunch of players who just go through their PCs' skill lists and roll dice. Under such circumstances, a DM who wants to encourage more imagination and out of the box thinking from the players should start hinting that such approaches would be more advantageous, and then actually reward the players' attempts, perhaps erring on the side of generosity, at least at first, to encourage more of such creativity in the future.
Finally, the third question:
3. Is the standard "three failures" model the best failure condition for this skill challenge?
While the standard "three failures" model makes sense occasionally (annoy the king too often and he'll have you escorted out of his throne room), it can make players hesitant to participate in a skill challenge if their PCs have low skill modifiers in the relevant skills as they are more likely to hurt than help the party's chances of overcoming the challenge.
There are ways to mitigate this effect: some skill checks may not count as failures for the purpose of the skill challenge if they are failed, the skill challenge may require a variety of skills so all the players may have to get involved eventually, etc. However, I think the simplest way to avoid the problem may be to have a different failure condition in the first place.
I must confess that recently, I've been quite taken with the idea of timed skill challenges (i.e. X successes within Y rounds or the skill challenge is failed) or the related idea of something bad happening to the PCs each round until they overcome the skill challenge (this is pretty much the approach taken by traps). In addition to the obvious advantage that every PC has the incentive to participate even if his chance of success is low, it also adds a (slightly) more tactical element to the skill challenge: if the PCs have not obtained the necessary number of successes by the final round of the skill challenge, some of them may think about spending action points to increase the chances of overcoming the challenge.
