Part 3 of 3 Plan, Write, Edit — an arc about the three jobs a book needs and the seams between them.

Part one argued that a book is written by making three moves — planning, drafting, revising — constantly and in no fixed order. Part two argued that when the plan is not reachable from the draft, the revision you end up doing is the sentence-level kind, which makes prose better and leaves structural faults exactly where they were.

This part is the practical one. What does it actually take to plan, write and edit a book in one place — what to ask of book writing software before you trust it with all three — and how do you work once you have it?

"One place" is not a count of apps

The phrase is usually sold as a number. Five tabs became one. That is a real improvement to your afternoon and it is not the thing that helps the book, because you can consolidate three programs into one and still keep the three jobs as far apart as they ever were — three sections, three separate screens, the same reorientation every time you cross between them.

A more useful definition, and the one worth holding any setup to:

Planning, writing and editing happen in one place when the shape of the book and the record of its decisions are addressable from inside the draft, in the same session, without leaving the passage you are working on.

Addressable is the load-bearing word. Not present somewhere. Not openable in another window. Reachable at the moment the question arises, which is almost always mid-sentence, and answerable without the reorientation that makes you decide the question was not that important.

Three tests tell you whether a piece of book writing software qualifies, and they are worth running on whatever you use now before you consider changing anything.

The three tests
  • The capture test. Your character just turned out to know about the letter. Can you record that decision where it will be found again, without leaving the paragraph? If the honest answer involves another window, the decision is going in your head instead.
  • The shape test. Mid-chapter, can you see what the surrounding chapters do — not reread them, see them — in a few seconds? If checking the shape means a reading session, you will not check it.
  • The consistency test. When you need to know what you decided in chapter three, does the answer come from a place that has been kept current, or from rereading chapter three?

Most setups pass the first test and fail the other two. That is the ordinary state of things, and it is why so much revision lands on sentences.

How to work once the three are together

Consolidating helps only if you change the habits the separation taught you. Four that matter.

Record the decision at the moment you make it, not at the end of the session. The end-of-session sweep is the most reasonable-sounding habit in writing and it does not survive contact with a real session, because by then you have made a dozen decisions and can reconstruct three. Every decision you write down in the second after making it is one you never have to excavate from your own chapters later.

Keep the shape as one line per chapter. Not a synopsis — a synopsis is a second draft that nobody reads. One sentence saying what changes. That list is short enough to read whole, which is the only property that matters, because seeing the book at a glance is exactly the capability a draft denies you.

Run the structural pass before the line pass, and never in the same sitting. They use different attention and the line pass always wins when they compete, because it offers immediate visible progress and the other offers uncomfortable questions. Separate them in time and the structural one survives.

Let the plan change in the draft's favour. When the scene contradicts the outline, the scene is usually right — it is the thing that actually exists. The plan is a map, and a map that argues with the territory gets redrawn. Writers who treat the outline as a commitment stop updating it, and an outline nobody updates is an outline nobody consults. There is a fuller treatment of keeping a plan current if that is the part you keep losing.

A first week
  1. Day one. Write the one-line-per-chapter list from memory. Forty minutes. The gaps are the finding, not a failure.
  2. Days two to five. Draft as normal, with one rule: every decision about the book that is not the sentence in front of you gets recorded where the shape lives, immediately.
  3. Day six. Reread only the list. Mark chapters where nothing changes, adjacent chapters doing the same work, and anything that arrives unprepared.
  4. Day seven. Act on exactly one mark — move, merge or cut. One structural change is worth more than a week of tightening, and it is the change a line pass could never have found.

Where book writing software comes in

All of that is doable by hand, and writers have done it by hand for a long time. What book writing software changes is the cost of those habits.

This is the part Write in a Click was built for: one home for your entire book, where planning, writing and editing are the same place rather than three destinations. The decisions live in Story Hub — world, characters, places, glossary, chapter notes and free notes — alongside the chapters instead of in a document of their own, which is what makes a chapter-three decision still true in chapter nineteen. AI Sync pulls world-building details out of the scenes as you write them, so the record stays current without a maintenance session you will not run.

And Robin, your co-author, knows the whole book — not a passage pasted into a chat box at the start of the conversation. That matters specifically for the pass part two says most writers cannot run: asking whether the middle third earns its place is a question about eleven chapters at once, and it can only be asked of something that has read all eleven. Robin answers about your book, by name, with the decisions you made in it — and does not rewrite you, because your voice is the thing the work is for. Your drafts are yours and are not used to train models.

It runs on the web, iPhone, iPad and Android — the same book on all of them, with offline drafting on mobile — so the shape of the book travels with the draft rather than living on one machine.

What this does not fix

Worth saying plainly, because a page like this is usually where the honesty stops.

Consolidation does not tell you whether the middle third is any good. It makes the question cheap to ask and gives you something to ask it of; the judgement is still yours and still hard. It does not make structural revision comfortable — it will always be the pass that threatens more work, and that is a fact about the job. It will not fix a book with nothing at its centre, and no arrangement of windows ever has.

What it does is narrower and worth having. It removes the reason the three jobs drift apart, so the small planning moves get recorded, the shape stays visible, and the revision you run is the one the manuscript needs rather than the one your setup left within reach.

That is the whole argument of this arc, in one line: a book does not move in three phases, so the question to ask of a writing setup is not how well it does each of the three, but how far apart it keeps them.