How to Write a Nonfiction Book

Start with one reader and one promise

Most nonfiction books that stall do so because the author is writing about a subject rather than for a person. “A book about leadership” gives you no way to decide what goes in. “A book for people promoted to their first management job in a small nonprofit, who need to run a performance review, a budget meeting and a difficult conversation in their first six months” tells you what to include, what to cut and which examples to use.

Write that reader down, then finish this sentence: “After reading this book, my reader will be able to…” The answer is the book's promise. It should be concrete enough that a reader could tell whether the book delivered it. “Understand money better” is too soft. “Build a household budget, pay down debts in a sensible order and choose between retirement accounts” is a promise someone can check.

Keep the promise where you can see it while you work. When you are unsure whether a story, a chapter or a whole line of research belongs, the test is simple: does it help this reader reach this result? If not, it may still be interesting, and it may belong in a different book.

Decide what kind of nonfiction book you are writing

Nonfiction covers very different kinds of books, and each has its own organizing logic. Mixing them by accident is a common reason a draft feels shapeless: a how-to book that keeps drifting into memoir, or an argument book that reads like a reference manual. Pick the main type first. You can borrow techniques from the others, such as a short story to open a how-to chapter, once the spine is clear.

Outline the chapters before you draft

An outline is the cheapest place to find a structural problem. Moving a chapter in an outline takes a minute; moving it in a drafted manuscript can mean a week of rewriting transitions and cross-references. For most nonfiction, a useful outline is a list of chapters, each with a working title and one sentence stating what the reader will be able to do or understand by the end of it.

Read only those chapter sentences, top to bottom. They should build in a sensible order, with no chapter depending on something that comes later, and together they should deliver the book's promise. Gaps and repeats show up at once. If a chapter's sentence will not come, that chapter is not ready to draft; fix it in the outline first.

Expect the outline to change as you draft. That is normal, but change it on purpose: update the outline whenever you move material, so it stays a true map of the book rather than an abandoned first plan.

Research in proportion to your claims

Not every sentence needs a source, but every factual claim a reader could check does. A practical rule: anything you could not defend if a knowledgeable reader challenged it needs a source, your own documented experience, or softer wording. Figures, dates, quotations, health or legal statements and claims about named people need the most care.

Capture the full citation the moment you use a source, including author, title, publisher or site, date and page, and store your notes so you can find the source behind any sentence later. Re-finding a half-remembered quotation in the month before publication is one of the most expensive mistakes in nonfiction. Our guide on how to research a book sets out a system for planning research, keeping notes and checking facts.

Avoid the opposite trap too: researching indefinitely because it feels safer than writing. Give each chapter a research budget, start drafting with what you have, and mark what is missing so you can fill it in one focused session.

How to write a nonfiction book draft without stalling

The first draft has one job: to exist. Its sentences can be clumsy and its examples thin, because revision will fix both. What revision cannot fix is a chapter that was never written. These habits keep a draft moving week after week:

1. 1Set a session target you can hit on a bad day, either a fixed time block or a word count, and protect the same slot every week.
2. 2Write the chapter you can see most clearly, not necessarily chapter one. The introduction is easier to write last, once you know what the book actually says.
3. 3When you hit a missing fact, type a placeholder such as [TK: year of first edition] and keep going. Search for the placeholders later in one research session.
4. 4Do not reread and polish yesterday's pages before writing today's. Read the last paragraph only, then continue.
5. 5End each session with a two-line note on what comes next, so tomorrow starts with a decision already made.
6. 6Track finished chapters against the outline, not word count alone, so you can see the book taking shape.

Revise in passes, from structure down to sentences

Revision works best as a series of separate passes, each looking for one kind of problem. Fixing commas in a chapter you later cut is wasted effort, so work from the largest scale to the smallest. Before the first pass, give the manuscript to a few people who match your intended reader, not only friends or colleagues who already know the subject. Ask where they got lost, bored or unconvinced; those are your revision targets.

Look for them where your intended reader already gathers, such as a professional group, a class or a workplace, and ask each one the same questions so their answers can be compared. A professional editor is most valuable after these passes, once the obvious problems are already fixed. Then work through the manuscript in this order:

• Structure: does each chapter deliver its promise, in the right order, with nothing essential missing?
• Chapter: does each chapter open with a reason to keep reading, move in clear steps and close with what the reader should take away?
• Evidence: is every factual claim sourced or softened, and every example accurate and fair to the people in it?
• Line: are sentences clear, specific and in your voice, with jargon defined the first time it appears?
• Proof: spelling, consistent names and terms, and correct numbering of figures and cross-references.

Choose how the book will reach readers

Decide your publishing route early, because it changes what you do next. Traditional publishers often buy nonfiction on the strength of a proposal, usually an overview, a chapter outline, an analysis of comparable books, a description of your platform and one or more sample chapters, before the whole book is written. If that is your route, the proposal may come before the full draft; see how to write a book proposal.

Self-publishing puts you in charge of editing, design, distribution and marketing, and you keep control of rights and timing. It suits authors who already reach their readers through their work, teaching or an existing audience. The full sequence of steps is in how to self-publish a book. Either way, think beyond the printed book: talks, workshops and courses can carry the same material to readers who learn better by doing.