A short book finished in a few months is worth more than a comprehensive book abandoned at chapter six. What follows is the whole arc in seven stages, in the order that keeps most manuscripts moving. It is not a schedule and it is not a formula — it is the sequence in which the decisions get cheaper.
1. Write the plan before you write the book
Not an outline. A plan: what the book is, who it is for, how long it will be, roughly how long it will take, and what it should leave the reader able to do that they could not do before. Two pages is plenty.
The reason this comes first is economics of effort. Every decision you defer gets more expensive as the manuscript grows. Changing the intended reader at the planning stage costs an afternoon. Changing it at forty thousand words costs the book.
2. Define the reader narrowly enough to be uncomfortable
"People interested in small business" is not a reader. "A tradesperson in their first two years of working for themselves who has never priced a job" is a reader. The narrow definition feels like it is shrinking the market. It is not — it is the only thing that lets you write a sentence that lands, and books that land get recommended.
Write one sentence: This book is for ___, who is stuck on ___, and by the end they will be able to ___. If you cannot finish it, you are not ready to draft, and no amount of writing will substitute for finishing that sentence.
3. Build the table of contents as a structure, not a list
The table of contents is the book's architecture and the cheapest place to fix it. Lay the chapters out and check three things: does each chapter do exactly one job; do they run in the order the reader needs rather than the order you learned them; and does the last chapter deliver the thing the one-sentence promise said it would.
Expect to reorder. Most first-draft contents pages are organised around the author's history with the material, which is almost never the order a newcomer needs.
4. Choose a working title, then stop thinking about it
A working title is a decision-making tool, not a marketing asset. Its job is to keep the draft on scope: when a section does not serve the working title, you notice. The final title is chosen much later, when you know what the book turned out to be. Confusing the two costs weeks of unproductive agonising at the exact moment you should be drafting. The title techniques page covers the final decision properly.
5. Draft to a shell, not to a mood
Build the repeating chapter template — title, opening, headed body, summary — and fill it. The first-chapter method is exactly this, and it applies to all of them. Drafting sessions then have a definable end: "write the second heading of chapter four" is a task; "work on the book" is not.
Two practical rules that survive contact with real life. Do not edit in the drafting pass — you will lose the day and gain three paragraphs. And write down the chapter you are not writing yet: keeping a running "later" list stops good ideas from derailing the current session.
6. Revise in separate passes
Revision is not one activity. It is at least four, and doing them simultaneously is why revision feels endless: structural revision (is the argument in the right order), line editing (does each paragraph earn its place), copy-editing (grammar, consistency, style sheet) and proofreading (typos, spacing, page furniture). Do them in that order, in separate passes. Fixing commas in a chapter you later delete is the most common waste in the whole process. The editing and proofreading page goes through each pass in detail.
7. Treat release as a start, not a finish
Publication day is the beginning of the book's working life, not the end of yours. Before it, three practical matters are worth settling rather than discovering:
- Copyright. In the United States, copyright exists on creation, but registration is what enables enforcement and, if made in time, certain remedies. The Copyright Office's literary-works guidance sets out what to file and when. Do not rely on a summary — including this one.
- Cataloguing and discoverability. Libraries find books through catalogue records. The Library of Congress explains what it offers publishers, including the Cataloging in Publication programme and its eligibility rules, which surprise a lot of independent publishers.
- Reviews and endorsements. If you give copies away in exchange for reviews, or a reviewer has a connection to you, disclosure rules apply. The FTC's endorsement guidance is written in plain language and is worth twenty minutes.
The honest summary
Six of these seven steps are decisions, and only one is writing. That ratio is not a rhetorical trick — it is the actual shape of finishing a non-fiction book, and it is why the people who finish are not always the best writers in the room. They are the ones who made the decisions early and cheaply.
Next: the first step in detail, or the publishing decisions if the draft already exists.