Book Proposal and Outline Template
An overview, a chapter outline, and the grid that says which of your chapters six comparable titles have already written.
Free download · No account needed
Comparable Titles Analysis
[Working title], [genre]
The grid
| Cells, chapters times comps | — |
| A chapter on the same subject elsewhere | — |
| A section or a passage elsewhere | — |
| Nothing | — |
What falls out of it
| Chapters touched by four or more comps | — |
| Their share of the manuscript | — |
| Chapters touched by none | — |
| Their share of the manuscript | — |
The sentence this section exists to produce
Not that yours is more narrative. Three chapters by number, their share of the manuscript, and the one book on the list you actually have to argue against, named, with the overlap counted.
The claim check
Every chapter the overview calls new ground, tested against its own row. Then the reverse pass: the untouched chapters nobody claimed, which is usually where the finding is.
The competitor
Comps ranked by chapters overlapped rather than by prestige, with first-year sales and the provenance of every figure beside them. The section opens with the top row.
Platform, in copies
Each audience number converted into copies somebody can evidence, then set against the comps’ median first year. That percentage is the one an acquisitions committee can use.
A publisher asks for two separate things. Princeton University Press wants a proposed chapter outline with a paragraph on each chapter, and, further down its own list, comparable books with “a brief statement outlining how your book differs”. New Mexico asks for three to five competing titles. Neither asks you to cross the two, and crossing them is the whole of what this pack does.
Fourteen chapters against six comps is 84 cells. On the worked proposal, 17 hold a chapter on the same subject elsewhere, 25 hold a section or a passage, and 42 hold nothing. Five chapters are touched by four or more of the six and carry 37.6% of the manuscript. Three are touched by none and carry 21.3%. That 21.3% is the market position, stated in chapters rather than adjectives. It was not the 21.3% the author expected: the overview claimed four chapters as new ground and two are covered by a comp.
Meanwhile the only untouched chapter in the first half had never been claimed, because the author had it filed as procedural background. The second join is sales against overlap. The comp listed fifth overlaps on 9 of 14 chapters, hits every crowded one, and outsold the rest at 9,800 first-year copies, so the book proving the market is the book to argue against. The third join is platform: 36,540 in reach converts to 374 copies anyone can evidence, 9.1% of a median comp’s first year.
What's in the pack
Chapter and Comp Coverage Matrix
Every chapter against every comp, one cell each, holding a chapter, a section, or nothing at all.
Comparable Title Register
Six to seven comps with first-year sales, the source of every figure, and the chapters each one overlaps.
Platform Metrics
Reach in one column and copies somebody can evidence in another, with the record held for each.
Comparable Titles Analysis
The section acquisitions reads first, opening with the book you overlap hardest and arguing against it by chapter.
Positioning Note
The worked arithmetic end to end, from 84 cells to the four sentences that changed in the proposal.
Overview
The first two pages, with the novelty paragraph written after the grid exists rather than before.
Chapter Outline
Numbered, with what each chapter does, its word target, and one line on the material behind it.
Author Platform Section
Leads with copies rather than reach, and does the division an editor was going to do anyway.
Marketing Section
Commitments ordered by what has already converted, so paragraph length follows evidence rather than enthusiasm.
How it works
- 1
Open or download
Edit with AI installs the pack as a private Space and asks for your numbered chapter list, four to seven comps, and their tables of contents. Download gives you the blank files, no account.
- 2
Fill the grid
Each chapter against each comp: a chapter on the same subject, a section or a passage, or nothing. Cells you have no evidence for stay blank and get counted separately.
- 3
Read the two shares
Chapters touched by four or more comps, and chapters touched by none, both as a share of the manuscript. The second share is the market position and the first one stays out of the pitch.
- 4
Test every claim
Every novelty claim in the overview against its own row, then the untouched chapters nobody claimed. Then each audience number converted into copies against the comps' median first year.
Frequently asked questions
What does it need from me?
A numbered chapter list with one line on what each chapter does, and a word target per chapter even if it is a guess. Four to seven comparable titles with their tables of contents, which is what most cells get filled from. Every audience number you can count today. Interview material belongs in a source register instead.
What format are the downloaded files?
Six Word documents and three CSV sheets, zipped. The documents open in Word, Pages and Google Docs; the sheets open in Excel, Numbers and Sheets. Nothing needs converting. Add a PDF query string to the download if you want to send somebody the analysis rather than fill it in.
What does Edit with AI actually do?
It installs this pack as a private Space, then fills the grid, which is the pass that costs the most by hand. You get the coverage matrix, the crowded and untouched chapters with their shares, the comp you actually compete with, and every novelty claim in your overview tested against its own row.
Is it not risky to point out the overlap?
The overlap is already there, and an acquisitions editor has read the comps. Princeton's guidelines say you may claim nothing comparable exists, then ask you to be absolutely certain. Naming 37.6% as crowded is what makes the 21.3% believable, and a proposal claiming total originality across fourteen chapters reads as unread rather than as bold.
Where do sales figures for the comps come from?
An agent or editor with a sales-system lookup first, then the publisher's own catalogue, then the author saying a number in public. Each figure carries its source and a yes or no on whether a sales system produced it. On the worked list two of six were not from one, and the register says so rather than quietly dropping them.
My chapters are not written yet. Does the grid still work?
Yes, and this is the stage to do it. The grid needs what each chapter will do and roughly how long it will be, not the text. Learning at outline stage that two crowded chapters carry 13,700 words is cheap. Building that outline, and checking what sits behind each chapter, is the long-form outline pack.
Is any of this free?
The download is free and needs no account: the real documents and sheets, blank, as Word and CSV. Edit with AI creates a free account and fills them in from your own outline and comps. Commissioning an outside read of the sample chapters is a developmental brief, and answering a copyeditor's queries without losing track of a decision is a copyedit query response pack.
Find out which of your chapters somebody already wrote
Download the blank Chapter and Comp Coverage Matrix, Comparable Title Register and Platform Metrics as Word and CSV, or have River fill the grid from your own outline.
Edit with AI