While there are many methods and tricks for designing NPCs, I am a great believer of the GTH technique.
GTH stands for, "a goal, a tactic, and a habit." It's an old writer's trick (I think it was credited to Samuel Delany last time I read about it) on creating characters. The idea of this method is that in order to create an NPC, while you can continue adding details, the essentials for using the character is to know one goal, one tactic, and one habit.
As a writer, I have collected over the years a number of quizzes and questions one's supposed to ask to develop characters. They talk about the last five years of the character's life, the physical details, the stories behind scars (emotional and physical) as well as motivations, family secrets, and the closest they've ever come to dying.
The problem is, they're of limited use. I know these answers, or I can make them up as necessary. (I once spent some time of an evening deciding which of my gaming characters would be most likely to shave their pubic hair. Don't ask.) The questions are basically there to help round out the characters... they don't necessarily tell you how to play them.
Three of the most important questions I try to ask my players of their characters are:
1. What are the character's greatest hopes and fears?
2. What circumstances would turn the character into a hero?
3. What makes this character interesting enough to play?
The first, their hopes and fears tie directly into my game design. I want the PCs' hopes to come true... and I want to use their fears against them. It sets up the first layer of conflict against the PCs. It tells me what kinds of threats are going to be effective, as well as what kinds of goals to develop.
The second, the circumstances for heroics is another plot-developing situation. The style of answer here tells me quite a bit, too: is the character already a hero? It gives me the follow-up question of, "Who are the character's heroes," as that also tells me what the character strives for, and sets a layer of stylistic development as I examine the environment necessary. Heroes rarely develop from simplistic success... even tricksters need to be hungry to create.
The third, what makes the character interesting in the player's mind gives me insight into the PC and the player. I sometimes follow it up with, "Why should I have the character in this campaign?" as the explanation tells me what the player knows (or thinks he or she knows) about the campaign as much as it gives me an idea of what they want from the campaign. In some ways, if I had to ask only one question, this would be it: in one sense, this tells me everything I need to know as a GM.
The truth is, asking these sorts of questions should be done at more than one point during the campaign. A character develops as much (if not more) in play than by deliberate design. While in the beginning it may be Damascus' goal to "get lucky," (isn't that true for all Bad Stuff characters? [grinning]) and take some much-needed rest, during the game it may become much more important for Damascus to learn who "the IBF" is, even if it interferes with her original character goal.
"How has this character changed?" is a measure of the impact of the story. One of the differences between diced and diceless games comes up right here: in diced games, sometimes legends are built on a significant die roll.
In the only real World of Darkness game I played, my character (who happened to be a WoD-version of myself, like the rest of the players) changed upon an unreal defensive roll against a very, very nasty blade coming towards me. The kind of, "It could happen in real life, maybe," situation, and the GM decided that the roll was significant enough to realize in-game the potential of my PC... [in this case, to trigger my transformation into a werespider.]
In diceless games, the process is a lot more a mutual development between GM and player. If the above scenario had been played diceless, I as a player probably wouldn't have determined it enough of a concern... and probably would have pushed for it to happen instead when my in-game loved ones had been threatened, or something suitably theatrical. Summed up, it reads:
Dice are a method of development in gaming characters.
...and not just in character creation, either.
Control of the development of a character can happen outside game circumstances as well. I didn't have any real idea how to play Damascus until I started trying to write some feats and exploits from her background. The character quiz got me started...but it didn't give me the necessary hook, the handle I needed.
In Damascus' case, it was a couple of quirks that actually took her into 3-dimensions for me: her "personal rules," and her "biting things at random to see how they taste."
At the same time, I knew Jinx-Jobina instinctively once I heard "her song." I hadn't been looking for it, but upon hearing it I mentally nodded and made a note. Whenever I needed to get into Jinx-Jobina's mindset, I could play the song, or think about how the song made me feel.
Creating characters is not the same as playing the characters. There are stress points, situations that the design wasn't intended to take. This goes for NPCs as much as PCs, and maybe doubly so, because there may be the tendency to make NPCs "one-trick ponies." (This is a flaw with the GTH technique: once the goal, tactic, or habit has been exhausted, I need to either destroy the character or develop further depth...sometimes "on the fly.")
This is also true of campaigns... but that's a discussion for another day.