How to write a book proposal.
In non-fiction, the proposal is what sells the book — editors buy from it long before a finished manuscript exists. Here is what goes in one, what each section actually has to do, and where each most often fails. It's the same structure our proposal builder walks authors through.
Why a proposal at all?
Novelists finish the book and then sell it. Non-fiction works the other way round: you sell the book first, on the strength of a document that argues for it. Nobody is asking you to finish it. They are asking you to prove it.
A proposal is a business case wrapped around great writing. It has to convince an editor of three things at once — that the book matters, that readers are waiting for it, and that you are the person to write it. Every section below carries part of that argument. A section that carries none of it is padding, and editors read enough proposals to know the difference.
Seven sections, and the job each one does.
Lengths vary between publishers and agents, and nobody will reject you over a page count. The specs below are what we ask for, and they are typical.
-
The hook
One to two sentences.
It tells an editor what the book is, who it's for, and why now. The framework we hand authors is a useful starting point: most people believe [common misconception]; actually [reframe]; and it matters more than you think because [stakes]. Adapt it freely — the test is whether it sounds like something an editor would quote back to a colleague.
Where it fails: naming a topic instead of making a claim. "A book about leadership" is a subject heading. A hook is an argument somebody could disagree with.
-
The target reader
One to two paragraphs.
Their occupation or life stage, the specific problem that brings them to your book, why they would choose yours over the alternatives, and why it's relevant to them right now. Describe the category of person who will buy the book — not an invented named character, which reads as a substitute for knowing.
Where it fails: vagueness. "Adults interested in business" is not a reader. "Mid-career professionals facing their first leadership role who have never managed a team" is — and a marketing department can do something with the second one.
-
Comparative titles
Three to five books, published in the last three years.
Comps show where your book lives in the market. You aren't competing with these titles, you're joining a conversation they're already having. For each: what it covers, what it does well, and how yours differs. Then a short market-gap paragraph — the angle or method none of them offer. Older titles can appear as secondary comps if you flag them as such.
Where it fails: comps picked to flatter. Listing the three biggest bestsellers in your category tells an editor you hope to resemble them, which is not an argument. Sales figures are never required and are better left out.
-
Author bio
200 to 300 words. One page at most.
Non-fiction publishers buy the author first and the idea second, so this section is doing more work than its length suggests. It needs credibility, a bit of humanity, and a clear throughline from your experience to this particular book.
Where it fails: merging with your platform. Follower counts and mailing-list numbers belong further down. Mixed together they weaken both — the bio stops being about judgement and the platform section loses the numbers it exists to present.
-
Chapter outline
25 to 100 words per chapter. Usually eight to ten, plus an introduction and conclusion.
This is where an editor sees whether the book works — not what it's about, but how it delivers on its promise from beginning to end. Structure is argument. The order of your chapters is a claim about how the reader's understanding should change.
Where it fails: a list of topics with no momentum. If the chapters could be reordered without loss, the book doesn't yet have a shape. Each one should make the next one necessary.
-
Sample writing
3,000 to 10,000 words. One to three chapters, usually the first plus one or two others.
Often the last thing an editor reads, and the section most able to override everything above it. Weak writing kills strong proposals; strong writing has rescued weak ones. There is no room to hide here, which is the point of including it.
Where it fails: sending the most informative chapter rather than the most compelling one. Pick the writing that pulls a reader in and sounds like the rest of the book.
-
Marketing and platform
Two pages or more, in three parts.
Your current platform, in real numbers: mailing list, social, site traffic, speaking. Then opportunities — named podcasts, publications, conferences, associations, people who might blurb it. Then your commitment: what you will personally do to sell this book.
It helps to know the arithmetic this section is answering. Breakeven for most publishers sits around 3,000 copies; 10,000 means they have made their money back and then some. Everything here is evidence about which of those is likelier.
Where it fails: promises nobody can check. "I'll promote it heavily on social media" is not a plan. No platform yet? Lead with engagement quality and growth — a small, active audience argues better than a large, silent one.
The sections are not equal.
When we assess a proposal we weight the sections rather than averaging them, and the order surprises most authors.
Marketing and platform carries more weight than any other single section — more than the concept. Then the overview and concept. Author credentials and the writing sample sit together after that. Comparative titles matter, but less than authors expect. The outline and sheer completeness carry the least: necessary, rarely decisive.
Read that as a warning rather than a scoring guide. The section most authors write last, fastest, and with the least enthusiasm is the one that moves the verdict most. If you have limited time, that is where it should go — and if your platform is genuinely thin, it is better to know that now than after six months of writing.
None of which makes the idea unimportant. A brilliant platform attached to a book nobody needs still doesn't sell. But of two good ideas, the one with evidence behind it wins, every time.
You don't have to write all of it first.
The most common way this goes wrong is writing forty pages before anybody has read the idea. Months of work, on spec, and the flaw that sinks it was usually visible on page one.
So start with a single page that answers five questions:
- What problem does your book solve?
- Who is your target reader — occupation, life stage, the thing that keeps them up?
- Why is your book different from what's already out there?
- Why are you the right person to write it?
- How do you plan to market it?
Those five are the spine of everything above. Answer them clearly and the proposal has somewhere to grow from. Struggle with them and forty pages will not rescue it — which is worth finding out in an afternoon rather than a year.
They're also exactly what our free author portal asks. It takes about twenty minutes, and a working literary agent reads what you write and sends back pages of notes on it. There's nothing to pay, and you don't need an agent to start.
See these sections in practice.
What each one looks like when it works, and when it doesn't — or start your own one-pager and have an agent read it.