Part one mapped the scatter. Part two priced it. This part is the method, and it starts with a claim I want to make precisely, because it is the whole argument:
Some manuscript editing passes cannot be run on a scattered book at all. Not are harder. Cannot. If the whole text is not in one place, a whole-book pass has nothing to operate on, and the writer substitutes the only thing available - a chapter-by-chapter pass - which finds a completely different and much smaller class of problem.
That is why consolidation is worth ninety minutes of your life. Not for tidiness. Because it unlocks work you currently cannot do.
Consolidate once, deliberately, in about ninety minutes. Then run the four passes that need the whole book at once: continuity, voice, structure, line.
Step one: the ninety-minute consolidation
The reason most consolidation attempts fail is that writers try to do them properly - reorganising, retyping, cleaning up as they go - and a full reorganisation of a half-finished book is a week of work that produces no pages. It gets abandoned around day two, and the abandonment is taken as evidence that the writer is disorganised, which starts the cycle again.
So do it badly and quickly instead. The goal is not a beautiful archive. The goal is one address.
- Pick the home first (5 min). One place where the draft will live from now on. It matters far less which one than that there is exactly one.
- Move the draft, in order (30 min). Every scene and chapter, in reading order, into that one home. Do not edit. Do not fix the thing you notice on page forty. Move it.
- Move the apparatus you actually use (25 min). Not everything - use part one's test: the things you leave the page to look up mid-sentence. Characters, places, invented vocabulary, timeline, the rules of your world. Everything else can stay where it is.
- Make one decisions file (20 min). A single running list of the things you have decided and must not contradict. Most writers find these were never written down anywhere - they were in your head, which is why you kept guessing.
- Archive the rest (10 min). One folder called archive. Nothing gets deleted, so nothing is at risk, and nothing outside the new home is live any more.
Two rules make this survivable. Do not edit while moving - the urge is enormous and it is what turns ninety minutes into a week. And do not fix what you find; write it on a list and keep moving. The consolidation produces one artefact - a book with an address - and one list of things to fix later.
From then on: the one-home rule. Anything you have to consult while drafting lives where you draft. New note about the world, new character, new decision - it goes in the home, not in whatever was open at the time. This is the part that actually holds, because it is one rule rather than a system, and it is enforced by convenience rather than willpower.
Step two: the four passes
Now the work that was previously impossible. Run these as separate passes. The temptation to do them all at once is strong and it is the single most common way revision goes wrong - your attention cannot hold four questions and will silently answer only the easiest, which is always line-level fiddling.
Pass one: continuity
Read the whole book with the decisions file open, and check only for contradiction. Names, ages, relationships, the geography, the rules of your world, the timeline. Nothing about quality. You are looking exclusively for places where the book disagrees with itself - which, if you read part two, are concentrated exactly where you guessed under momentum.
This pass is straightforwardly impossible on a scattered book, because contradiction is a property of the whole text. You cannot see it a chapter at a time.
Pass two: voice
Read for sound only, ideally aloud, ideally in one or two sittings. You are listening for drift: the point where the prose stops sounding like the writer who started the book. Mark the boundaries where it changes rather than trying to fix sentences.
Then a decision most writers skip - decide which voice is the real one. Often the later chapters are better, and the correct fix is to bring the opening forward rather than dragging the ending back. That judgement requires hearing both ends of the book in one sitting, which is the specific thing scatter takes away from you.
Pass three: structure
With the whole draft in one place, read only for shape: where the book accelerates, where it sags, what each chapter is actually for. A chapter that exists to deliver one piece of information is a chapter to merge. A stretch where three chapters do the same work is a stretch to cut.
Structure is the pass most improved by consolidation, because shape is invisible at close range. Chapter by chapter, everything looks locally reasonable. That is exactly the trap I described in turning brilliant scenes into a finished novel - individually excellent scenes that do not add up to a book.
Pass four: line
Last, and only last. Sentences, word choice, rhythm. This is the pass everyone wants to start with, and starting with it means polishing prose in chapters that pass three would have cut. It is the most enjoyable and least valuable pass, and it belongs at the end for exactly that reason. If you have caught yourself unable to leave it alone, I wrote a whole method for escaping the revision loop.
Two signals, both within a fortnight. Your sessions start faster - the reassembly at the top of a session shrinks toward nothing. And your "check this later" tally from part one drops, because the answer is now on the same screen as the sentence.
Where we fit, honestly
Full disclosure, as always on this blog: consolidating a book into one home is the problem Write in a Click was built around. One home for your entire book - planning, drafting, editing, the world-building in Story Hub, version history so the old paragraph is still attached to the thing it is a version of, and Robin, an AI co-author who has read the whole manuscript rather than the paragraph you pasted in.
That last part is the piece that only works once the book is whole. An assistant reading one chapter can help with sentences. An assistant that can see the entire manuscript can tell you that the brother's age changed in chapter fourteen - and that is a difference in kind, not degree. It is also, precisely, the continuity pass above, which is why the argument in this series and the thing we build are the same argument.
But this method is not a product tour, and I would rather it were useful than persuasive. Everything above works in any tool that holds the whole book: one long document, a single project file, a folder you keep discipline over. The consolidation is the load-bearing step, and it costs you ninety minutes wherever you do it. If you take one thing from these three parts, take that - the ninety minutes, not the recommendation.
The end of the series
Three parts, one argument. A book manuscript is the draft plus the apparatus plus the history, and when those live apart, you pay in re-entry minutes, in continuity you have to repair later, and in a voice you cannot hear well enough to hold. The fix is not discipline and it is not better notes. It is distance: making the gap between where you write and where the answers live equal to nothing.
Run the inventory from part one. It takes twenty minutes and it is the only part of this you cannot skip, because you cannot consolidate a book whose pieces you have not found.
