How to Write a Book Proposal Editors Actually Buy

I sold my first nonfiction book off a forty-one-page document, and I had not written a single chapter of the actual manuscript. Somewhere in a drawer I still have the printout, coffee-ringed at one corner, with an agent's pencil marks crowding the margin of the market analysis section and three question marks next to a comp title she thought was too recent to prove anything. That document took me four months to write. The book itself, once the contract was signed, took fourteen.

This is normal, and it surprises almost everyone who has not done it. Editors in trade nonfiction do not buy manuscripts. With rare exceptions -- memoirs from people already famous, narrative nonfiction from writers with a long magazine track record -- they buy proposals. The proposal is the sales document. The manuscript comes after the check clears. A proposal is not a lesser draft of the book you are going to write. It is a different document, built to answer different questions, and the writers who treat it as a rough sketch of chapter one are the ones whose proposals get form-rejected without a note.

What a Proposal Argues, Beyond the Book's Subject

A book proposal runs 30 to 60 pages for most trade nonfiction and does three jobs at once. It proves the book has a shape an editor can see all the way through. It proves an audience exists and will pay for it. And it proves you, specifically, can write the thing, on time, at the length promised. Miss any one of the three and the other two do not save you. I have read proposals with a gorgeous premise and no evidence the writer had thought about who buys the book. I have read proposals from writers with enormous platforms and no argument, nothing an editor could picture as a finished table of contents. Both get passed on.

Robert Caro's memoir Working describes his own decades inside this system: proposals, advances tied to word counts he blew through by years, and an editor's patience tested by how long the actual writing took once the proposal had done its job.

The proposal bought him the time.

It is worth remembering that even at that level of the profession, the document that gets you the contract and the document that gets you the finished book are not the same act of writing, and do not require the same kind of certainty. A proposal argues that a book will exist and what shape it will take. A manuscript is the shape, argued out in every sentence, and you are allowed to discover parts of it that the proposal did not predict. Editors expect some drift. What they do not forgive is a proposal so vague that no drift could ever be measured against it.

The Eight Sections Editors Expect, In Order

Trade nonfiction proposals follow a structure that has stayed remarkably stable across imprints and genres, memoir included. Deviate from it and you make an editor work to find the information they are used to finding on page one.

1. Overview -- the pitch, the argument, the hook, why now, why you
2. About the Author -- credentials, platform, previous publication, personal stake
3. Market Analysis -- target reader, comp titles, category trends
4. Marketing and Promotion Plan -- what you will do to sell the book
5. Chapter Outline -- the shape, chapter by chapter
6. Sample Chapter or Chapters -- proof you can write the book
7. Practical Details -- word count, delivery date, special production needs
8. Table of Contents -- a one-page skeleton, sometimes folded into the chapter outline

Some agents reorder these or fold two sections together. None of them skip any of the eight outright, and an editor who has read a thousand proposals notices the gap immediately when one is missing.

Memoir, History, and Prescriptive Nonfiction Ask Different Questions

The eight sections stay the same across categories. The weight each one carries does not, and writers who apply the same emphasis regardless of category tend to over-invest in the wrong pages.

Memoir lives or dies on the sample chapters and on the author section, because the author is the subject. Voice has to be proven on the page, not asserted in the overview -- an editor cannot buy "compelling and honest" as a description of your own voice, only as a demonstration of it. Cheryl Strayed's Wild is the book most people reach for as an example of memoir selling on authority and voice together, and it is worth noticing how little the premise alone explains its success; plenty of memoirs share a comparable premise. The platform section matters less here than in prescriptive nonfiction, though it still matters. The marketing plan can be lighter. The sample chapters cannot.

History and biography lean hardest on the chapter outline and on access -- to archives, to living subjects, to primary sources nobody else has used the same way. An editor reading a history proposal is asking a quieter version of the question a journalist's editor asks before a big feature: what do you have that nobody else has? Comp titles in this category need to demonstrate shelf life as much as recent sales, since serious narrative history keeps selling for years after its pub date in a way some other categories do not.

Prescriptive nonfiction, the how-to and the business book and the book organized around a single framework, sells on platform and on the framework's legibility more than on the writing itself, which is not a popular thing to say to writers but is close to universally true among the acquiring editors I have spoken with. Add a short section articulating the central framework in graspable form, almost a one-page executive summary of the idea, because that page is often the one that gets forwarded to a sales conference before anyone reads chapter one.

The Overview Carries More Weight Than Any Other Page

The overview runs three to five pages and it is the only section every reader of the proposal reads start to finish. Agents skim chapter outlines. Editors sometimes hand sample chapters to a junior colleague. The overview gets read by whoever is deciding whether the rest of the document is worth their afternoon.

Open with the hook: a scene, a statistic, a question, the thing that made you need to write this book rather than merely wanting to. Then state the argument or the arc in language specific enough that an editor could repeat it back to a colleague in one sentence. Then answer "why now" -- a cultural shift, a news cycle, a gap on the shelf that has opened recently rather than always existed. Then "why you," briefly, because the fuller case belongs in the author section. Susan Orlean's The Orchid Thief and Erik Larson's The Devil in the White City both open their respective books the way a strong overview opens a proposal: a specific incident, narrow enough to picture, that turns out to be the entry point into something much larger. If your overview cannot name that specific incident, the book probably has not found its shape yet either.

Comp Titles Are a Credibility Test, Not a Formality

The market analysis section names your target reader specifically -- not "anyone interested in history," but readers of accessible narrative history who buy hardcover and follow a particular kind of true-crime-adjacent story -- and then backs that claim with four to six comparable titles.

Good comps are recent, usually published within the last three to five years. They sold respectably without being runaway bestsellers; comping your debut to a book that sold two million copies reads as either naive or desperate, and comping it to a book that flopped reads as not having done your homework. They come from your actual category, not your general vibe. An editor reading four well-chosen comps learns your genre, your reader, your positioning, and your reading habits in about ninety seconds. An editor reading two comps to blockbusters they have heard of on the subway learns nothing except that you have not spent much time in the nonfiction section of a bookstore lately.

William Zinsser's On Writing Well, decades old now and still assigned in journalism programs, makes an argument that applies directly here: clarity is a form of respect for the reader's time. A comp-title list is where that respect is easiest to fake and easiest to catch faking.

Writers whose book genuinely has no close comp face a harder version of this problem, and the instinct to write "there is nothing else like this" should be resisted almost every time. Nothing is ever quite that alone on the shelf. Widen the search to adjacent categories -- a different subject with a similar structural device, a different decade with a similar reporting method -- and use two comps to establish two different axes of comparison instead of one comp to claim a false uniqueness. An editor trusts "this combines the reporting depth of one book with the narrative pace of another" far more than "there has never been a book like this," which every acquisitions meeting has heard and mostly stopped believing.

Platform Decides Whether Prescriptive Nonfiction Sells

For memoir and narrative nonfiction, the sample chapters and the overview carry the most weight. For prescriptive nonfiction -- the how-to book, the business book, the book built around a framework -- platform often decides the deal before an editor finishes the first sample page. Editors want a specific number: email list size, average newsletter open rate, podcast downloads, speaking engagements booked for the coming year, prior media coverage with actual outlet names attached. Round numbers with no source read as invented, even when they are not.

I once watched a proposal for a genuinely good business book die in acquisitions because the platform section said "a significant social media following" with no number attached anywhere on the page. The agent had loved the framework. The editor who championed it internally needed a number for the sales memo, and there was not one to give her. When nobody could produce it in time, the book went to a smaller press for a smaller advance, and the framework inside it, which I still think about more than the frameworks of several bestsellers that year, reached a fraction of the readers it deserved.

The Chapter Outline Proves You Know Where the Book Ends

Each chapter entry needs a working title, an approximate word count, and one or two paragraphs describing what the chapter accomplishes -- the argument it advances or the events and emotional turn it covers, plus the sources or scenes it draws on. This section is a planning document as much as a sales document; once the contract is signed, you draft from it. Writers who pad this section with vague, interchangeable descriptions ("this chapter explores the theme further") are usually the same writers who stall at chapter six, because they never actually knew what chapter six was going to do.

Mary Karr's The Art of Memoir makes a case that applies well beyond memoir: a book only earns its length when each section is doing work the others cannot do. Write the chapter outline as if you had to defend, chapter by chapter, why that chapter exists at all. If you cannot defend one, cut it before an editor asks the same question in a rejection letter.

The Sample Chapter Is the Only Proof That Counts

Most proposals include one to three completed chapters, ten to twenty thousand words total, usually the opening chapter or introduction plus one representative chapter from later in the book. This is the section that gets compared against published work, not against your other unpublished drafts, so it needs to be genuinely finished -- not a placeholder, not a rough pass you plan to fix after the deal closes.

David Grann's books read like the reporting was locked down long before a sentence of the proposal was written, and that order matters more than it looks like it does. A sample chapter that promises access or evidence you have not yet secured is a promise an editor cannot verify until it is too late to back out cleanly. Write the sample chapter last, after the overview and outline have forced you to know what the book actually is. Writers who draft the sample chapter first often end up rewriting it twice, once to fit the book they eventually decide to write.

Choosing which chapter to include depends on the category. For memoir, choose the chapter that proves the voice can carry difficult material without tipping into either sentimentality or distance -- usually not the most dramatic chapter, which can read as manipulative in isolation, but the chapter that shows the most control. For history or biography, choose the chapter that best demonstrates your research method, the one where a reader can see you moving between sources and drawing a conclusion they could not have drawn from any single source alone. For prescriptive nonfiction, choose the chapter that best demonstrates the framework actually working on a real example, not the chapter that explains the framework in the abstract.

Format and Length Details Editors Check First

Total word count for trade nonfiction usually lands between 70,000 and 100,000 words; memoir runs shorter more often than history or biography, which can run longer when the research demands it. State a realistic delivery date, one that accounts for your actual schedule rather than the schedule you would choose in an ideal year -- editors have heard every version of the eighteen-month proposal that becomes a four-year book, and a proposal that shows you have thought honestly about your own pace reads as more trustworthy, not less ambitious. Note photographs, illustrations, or an index if the book needs them; production complexity affects the offer, sometimes by more than a first-time author expects.

None of this is where a proposal wins. It is where a proposal quietly loses points, the kind of detail an editor's assistant flags before the manuscript reaches anyone who cares about the writing. A proposal that gets these numbers wrong, a word count that does not match the category or a delivery date eighteen months out for a book that plainly needs three years of reporting, reads as a writer who has not yet thought like a publishing professional, regardless of how good the premise is.

Submission Strategy: Query the Idea Before You Finish It

Most nonfiction writers query agents with the overview, the author section, and a short pitch letter before the full proposal is polished, sometimes before the chapter outline exists in final form. This is the opposite of fiction, where the manuscript has to be complete before you query at all, and new nonfiction writers often waste months finishing a document nobody asked to see yet. Query five to eight agents at a time who represent your category specifically -- check recent sales, not just client lists, because an agent's stated interests shift and their website often lags behind it. Once an agent offers representation, the proposal gets one more editing pass with their notes before it goes out to editors, and that pass is frequently the one that turns a good proposal into a sold one.

A smaller number of nonfiction writers, usually ones with an existing platform substantial enough that editors already know their name, submit directly to editors without an agent, though even then most publishers expect one before a deal is finalized. Do not read a handful of rejections as a verdict on the book. Read them for pattern. If three agents independently flag the same weak section, usually the market analysis or the chapter outline, that is not bad luck, that is a proposal telling you what to fix before agent four reads it. I have revised a proposal after eleven rejections and sold it, essentially unchanged from what agent twelve saw, to the first editor that agent submitted to. The proposal was not the problem for most of those eleven. The list of agents was.

Pair the proposal with our book proposal template to build each section against a working structure instead of a blank page, and once you are ready to submit, our manuscript submission checklist catches the formatting and packaging details that get otherwise-strong submissions pushed to the bottom of an agent's pile.

The Document That Buys You the Time to Write

Guy Kawasaki's old rule for pitch decks -- ten slides, twenty minutes, thirty-point font -- was written for entrepreneurs, not writers, but the instinct underneath it is the same one that makes a book proposal work: say the true thing as plainly and as soon as possible, then earn every page after that. A proposal is not the place to be literary. It is the place to be legible. The literary part comes later, in the fourteen months after the check clears, when nobody is grading your sentences except you.

I still have that first proposal in the drawer. The pencil marks have faded more than the coffee ring has. Every book I have proposed since started life exactly that unglamorous way -- forty-some pages, a market analysis somebody argued with, and a chapter outline I mostly kept and occasionally broke.

Draft your book proposal in Plotiar. Keep the overview, market analysis, comp titles, and chapter outline in one project, searchable and revisable as the book itself takes shape. Try it free.