The last and simplest — you'll see soon how the pun is intended — of Krugman's rules of research is "Simplify, simplify".
... always try to express your ideas in the simplest possible model. The act of stripping down to this minimalist model will force you to get to the essence of what you are trying to say (and will also make obvious to you those situations in which you actually have nothing to say). And this minimalist model will then be easy to explain to other economists as well.
In the same way, it helps to think about what you're trying to do in the simplest possible way, both as an explanatory device and as a disciplinary one. Explanatory first.
Most of us know the joy of walking around with some grand design in our heads, and the subsequent embarrassment of trying, and failing, to explain it to other people. "Well, it's about this guy who — no wait, first there's this dog, and — " If they haven't already walked away from you by then, they're being charitable.
The "elevator pitch" is both respected and reviled, because it forces you to give someone the ten-second, standing-on-one-foot version of the story. But people squirm at it because they feel like they're doing violence to their precious little concept by boiling it down to its bones. My point is twofold:
a) If your story doesn't have good bones to begin with, then there's a good chance it won't survive such a boil-down anyway, and
b) The reason you boil it down is not to do violence to it, but to intrigue other people. I emphasize this because people often feel they are doing their audiences a disservice by "misrepresenting" their work. But you're not — if your work is good and has things in it that will draw the attention of others, then don't worry about them getting bent out of shape if you left out a few details on the way in. I cringed when I heard how the Brothers Schrader had pitched The Yakuza to Warner Brothers as "Bruce Lee meets The Godfather", but it got the project sold.
Sometimes you have to go to someone else to figure out what can be said most succinctly about your work, as when Steven Savage summed up my Flight of the Vajra as "A more responsible version of Tony Stark finds he's got to save the galaxy - and his team consists of a circus acrobat, a futuristic Dali Lama, Jim Gordon, Seven of Nine, and David Bowie." I could not have thought that up in any number of lifetimes.
OK, now disciplinary. By this I mean having some succinct idea of what your story really is, under it all, keeps you from meandering. Or, if you do meander, you know how to get back home. Everything that gets put into a story gets put there for a reason, and if you know the reason like your own name, you never get lost. It also keeps you from confusing length with depth, and complexity with convolution.
I've noticed people often have trouble with the idea of not putting anything in that doesn't really need to be there. The problem is that they fall in love with things — a moment, a beat, an idea, a flashy something — and they want to give it a home. Not a bad impulse, but that's why you write it down somewhere else and then take a good hard look at it. One nice thing about keeping such a goodie bag is that you're never short of things to toss in to liven a story up — and if you've been diligent and saved plenty of those things, you increase the odds of finding the thing that matches the story best and — how's this for irony? — helps you keep it simple.