I'm going to be lazy and repost some of the Q&As from my teenlitauthors stint this past week in lieu of any original writing.
When you receive a requested noncontractual revision from an author, what is your process for reading it? How many revisions will you go through before you let something go? When do you judge it ready to take the next step?
When I like a manuscript a lot, I often (even usually) require revisions from authors before I share a manuscript with the rest of our editorial staff. This is partly a safeguard on my part, because I really want to know an author can revise well -- and that we're on the same page about the book and can work together well -- before we go forward on that journey.
I read a revision the exact same way I read an original manuscript -- beginning to end, making notes in my head as I go along. I've talked to authors before who say "You read the first draft, I sent you the revision with that new chapter two weeks ago, why haven't you responded yet?" It doesn't work like that -- I don't (I *can't*) read a book in pieces (and nor can any other editor that I know of). Every ms. has to be considered as a whole to see how its individual parts function, and whether the new parts are working in the old machine . . . or, better yet, if the parts have been so fully integrated that we have a whole new shiny gleaming machine, and it's ready to rock and roll.
Once I've given it a first read and made notes, I look back at my revision letter and compare the notes to the letter to decide whether the author fixed the problems I identified in a satisfactory manner. (Usually I will remember the ms. I invite for revision enough that I have the major problems in my head, but the letter is a useful reminder.) (Also, the author doesn't have to have taken my specific suggestions for how to fix the problems, certainly, but the problems have to have been fixed by *some* method.) And I hope the answer is "yes, the author HAS fixed the problems" -- though usually I'll know this just from my first read, whether those same problems were occurring to me again. Sometimes the revisions create new problems -- one character is now so fully rounded that everyone else seems cardboard, or the author has bulked up the action plot SO much that the emotional plot is getting lost; but if the author has intelligently fixed the problems I identified originally, then I will usually give it at the least another revision letter.
The most common problem with revisions is that an author will fix one part and not integrate it sufficiently with the whole -- introduced a subplot to add emotional depth, say, not realizing that the new emotional depth should affect every single aspect of the character's behavior in the main plot as well. In good revisions, the author has clearly thought through all the implications of the change s/he is making and implemented those throughout, not just in the places where problems were explicitly identified.
All of this also depends on the problems of the individual ms. -- it's much easier to fix plot editorially than it is to fix character and voice/writing, as those have to come from deep within the author; so if the problems revolve around plot, or if the writer shows ability to make a character deeper and more interesting in a revision, that's worth investing more time in. But if the author can't improve flat characterization from one revision to another, the ms. probably won't ever go any deeper or better than it is, and it's probably wisest for me to let it go.
I try not to go through more than three revisions with an author: If it hasn't gotten to a workable place by that point, even if I love the project, I start to lose enthusiasm for it (and I think the author does as well). But if the author DOES revise well, I judge a ms. ready to go when it satisfies me enough -- and I'm excited by it enough -- that I want to share it with other people.