I'm finishing up the first draft of a new assignment, a low budget action movie (for more info on low-budget (but higher budget than mine) action movies read the excellent Bill Martell's blog).
One thing about assignments is that the outline and structure is far more fixed than when you're doing things yourself. The people buying it want to know pretty precisely what they're getting.
I'm also working with the director on this (in fact, it's his idea I'm developing) and he wants it to have theme and character arc, some real moment of drama that we can pull out of it -- the heart of the story. So we bounce around some ideas and lay in the sketch of that for the outline.
But it's not until I'm halfway through the first draft that what feels like the correct heart starts to show itself. And it's not until I'm fully through the first draft that I hit the full theme and character drama that makes me feel this can really work.
And these are things that just never emerge, even from a detailed outline (for Marlowe the outline was over 20 pages, and the script was only 95!).
Which makes me think that the heart of a story is an emergent thing. It is not so much something that you put in there the way you put set pieces and characters and reversals in there. It is something that arises in the interstices between those bigger pieces of structure -- small moments of character, small changes in dialog, and so on. And when I did hit on the heart of the story, what I mainly did to bring it out was mainly go back and make subtle alterations to scenes.
In a real sense, the plot didn't change at all -- just some flavor bits around the plot. Which is frankly a bit nerve-wracking because I think the story felt quite a bit worse before, and that means that the difference between a good story and a mediocre one could be nothing more than a pageful or two of words out of the 100 pages of script.