in Editing
Knowing when to stop revising
Revision has no natural end point, so you have to choose one. Here is how to tell improvement from mere change, and when to put the pen down.
Revision is the part of writing that never announces its own end. A first draft is finished when you reach the last page. A revised draft is finished when you decide it is, and that decision is harder than it looks.
Some writers stop too early, sending out a manuscript that still has a structural problem they half know about. Others never stop, circling the same chapters for years. Both are avoiding the same thing: a clear judgement about what the book needs.
Improvement and change are not the same
The simplest test is to ask what each round of revision actually did. Early passes tend to make large, visible differences. A subplot goes. Two characters merge. The opening chapter moves to page forty. Each change solves a problem you could name.
Later passes are quieter. You swap one adjective for another, then swap it back. You move a paragraph and later move it home. When your edits start cancelling each other out, you are no longer improving the book. You are rearranging it.
A useful habit is to keep a short log of what you changed in each pass and why. If the reasons become vague, or you notice the same sentence appearing in three successive entries, that is information.
Know what kind of problem is left
Not all remaining problems are yours to solve alone. There is a point where the author has become the least reliable reader of the book, because they can no longer see the page without seeing their intentions behind it.
Say a manuscript has a middle section you suspect is slow. You have cut it twice and it still feels slow to you. It may genuinely need restructuring. It may be fine, and you are simply tired of it. You cannot tell the difference from inside, and more revision will not settle the question. A fresh reader will.
This is also the moment to separate problems by scale. Structural questions, such as whether the book starts in the right place or whether the ending pays off what came before, deserve attention before anyone else sees the manuscript. Sentence-level polish can wait for a copyedit, where someone will read every line anyway.
Signs you are close
A few signals suggest a draft is ready to leave your desk. You can state what the book is about in a sentence or two, and every chapter serves that. You have read it through in one or two sittings without stopping to fix anything large. The changes you want to make are ones a copyeditor or proofreader would catch.
Another sign is that your notes have shifted from problems to preferences. A problem is something a reader would stumble on. A preference is something you would do differently on another day. A manuscript can carry plenty of the second kind.
Setting a limit
Some writers find it helps to set a limit in advance: a fixed number of full passes, or a date by which the draft goes to readers. The limit is arbitrary, but that is its value. It turns an open question into a closed one and forces you to prioritise.
If you reach the limit and something important is still wrong, you will know, and you can extend it for that specific reason. What a limit prevents is the drift, the months spent touching sentences that were already fine.
Stopping is not the same as declaring a book perfect. It means the book is ready for the next person who needs to read it.