Galley
FIELD GUIDE NO. 02

Creating a Nonfiction Book Outline

Turn one clear reader promise into a chapter-by-chapter plan you can actually draft.

Eight steps · An eight-part scorecard · A 45-minute sprint

Most abandoned manuscripts do not run out of words. They run out of structure.

The book started as a topic. Chapters accumulated the way notes accumulate: whatever came to mind, in the order it came. Around chapter four the overlaps appear. Around chapter six the writer is explaining the same idea a second time and cannot say what the next chapter is for. The draft stalls, and it stalls for a structural reason that the first hour of planning would have caught.

An outline prevents that, and it is not a formality. It is the promise of the book broken into an ordered set of deliverables, each one small enough to draft in a sitting. This guide is the method for building one. Work through it once and every writing session that follows starts with an assignment instead of a blank page.


01 THE SHORT ANSWER

What a nonfiction outline must accomplish

A useful outline is not a storage place for everything you know. It is a route from where the reader is now to a result the book can credibly help them reach. The chapters are the major stages of that route, and the sections explain, demonstrate, or help the reader complete each stage. An outline earns its place only if it reduces expensive decisions later, and it does that by performing four jobs.

01

It protects the promise

Every major part helps deliver the result implied by the title, subtitle, and positioning. Material that does not serve that result belongs elsewhere.

02

It creates progression

The reader moves through a sequence: understand, decide, prepare, act, troubleshoot, continue. They should never merely accumulate information.

03

It controls scope

An outline shows what the book will not cover. A focused book that fulfills one useful promise is more credible than a broad book that resolves none.

04

It exposes risk early

It reveals where you need research, expert review, permissions, or careful qualification, all cheaper to find before thousands of words depend on them.

Before drafting, the outline should be able to answer five questions. Who is the book for? What is true for that reader before the book, and what should be true after? What must the reader understand or do, and in what order? What evidence and tools make each step trustworthy? The governing principle behind all five: a nonfiction outline is a promise broken into an ordered set of deliverables.


02 DEFINITION

A topic list is not an outline

Put the same book side by side in both forms and the difference is immediate. One names subjects. The other assigns work. The test is whether each line tells you what the reader will be able to do once the chapter has done its job.

A TOPIC LIST · NAMES SUBJECTS

Mindset. Productivity. Communication. Meetings. Feedback. Every subject is relevant, but there is no order and no job, so the list cannot tell you what any chapter must accomplish or when the book is done.

A TRANSFORMATION OUTLINE · ASSIGNS JOBS

Diagnose why the current system feels scattered. Establish agreements that remove avoidable uncertainty. Build an asynchronous rhythm. Redesign meetings around decisions. Create a feedback loop. Install the system over 30 days.

The second version gives each chapter a job and implies an order. It also quietly defines the finish line: when the reader has installed the system, the book has kept its promise.


03 STRUCTURE

The anatomy of a useful nonfiction book

Most practical nonfiction can be planned through four structural zones. They are functions, not mandatory labels. A workbook and a narrative guide will wear them differently, but each zone still has a job to do.

THE OPENING · ESTABLISH THE CONTRACT

Help the intended reader recognize themselves and trust the journey. Identify the reader and situation, clarify the cost or opportunity, state the result, set boundaries, and show how to use the book. Your story belongs here only when it earns trust.

THE BODY · DELIVER THE TRANSFORMATION

The major stages between Point A and Point B. A chapter may help the reader decide, learn a principle, complete a task, avoid a failure, or integrate earlier steps. The number of chapters follows the promise, not a universal formula.

THE CONCLUSION · CONVERT KNOWLEDGE INTO CONTINUATION

Do not reopen the whole subject. Reconnect the reader to the outcome, summarize the few ideas that must survive, name the next action, address the likely relapse, and offer an appropriate next resource.

THE BACK MATTER · PRESERVE TRUST AND USEFULNESS

Sources, worksheets, a glossary, recommended tools, and the next step. Plan these early. A source list assembled from memory at the end is rarely reliable.


04 THE METHOD

The eight-step Promise-to-Page method

Run these in order. Each step converts an abstract book idea into a more testable plan, and each one uses the output of the step before it.

THE PROMISE SENTENCE

This book helps a specific reader move from a recognizable present condition to a useful, bounded outcome by a distinct method.

A promise is not a guarantee. Avoid outcomes the book cannot control. This sentence becomes your validation filter for every structural decision that follows.

1

Lock the reader and the promise

Do not outline for a vague audience such as people interested in productivity. Choose a reader whose situation changes what advice is relevant, then complete the promise sentence above. Keep the outcome bounded, and avoid promising results the book cannot control.

2

Make Point A and Point B observable

Confused to confident is emotionally clear but structurally weak. Describe what the reader can and cannot currently do, in terms you could verify.

  • Point A: present behavior, the key uncertainty, the constraint that makes generic advice unsuitable, and the cost of staying stuck.
  • Point B: a decision made, a system created, a skill demonstrated, or an action completed.
  • A weak Point B: the reader understands remote management. Stronger: the reader leaves with documented agreements, an update rhythm, redesigned meetings, and a 30-day plan.
3

List the reader's required questions and obstacles

Before inventing chapter titles, list everything that must be resolved for the promise to be true. Write each item on its own line and do not organize yet. The purpose is coverage, not order.

  • What must the reader understand first, and what decisions must they make?
  • What must they prepare or gather, and what actions must they complete?
  • What is likely to go wrong?
  • What changes for beginners, advanced readers, or special circumstances?
  • Which claims require evidence or expert review?
4

Cluster the material into stages

Group related questions and actions by the job they perform for the reader. Do not make every cluster a chapter automatically. Merge clusters that serve the same decision, and split a cluster that contains two outcomes able to stand alone. Most practical books cluster into four kinds of work:

  • Orientation: what the reader must understand before acting.
  • Diagnosis and decision: how to assess their position and choose between options.
  • Preparation and execution: what must be in place, then the core method.
  • Troubleshooting and integration: predictable failures, then how to sustain the result.
5

Sequence by dependency, not excitement

Authors put their favorite material first. Readers need material in the order that makes the next action possible. If two chapters can swap places without affecting comprehension, their roles may be unclear, or the book may really be a reference book rather than a sequential guide. Three tests expose a weak sequence:

  • Dependency test: does the reader need Chapter 2 before Chapter 3 can work?
  • Decision test: is the reader asked to choose before they understand the options or criteria?
  • Momentum test: does the book produce an early, meaningful win without hiding an essential warning?
6

Give every chapter one job

Write a one-sentence job statement for each chapter: by the end, the reader will be able to decide, explain, create, perform, diagnose, or avoid something specific. Two chapters with nearly identical job statements should be merged or redefined.

  • Strong, observable verbs: choose, compare, diagnose, calculate, design, practice, install, troubleshoot, evaluate.
  • Weak, vague verbs: know, appreciate, explore, become aware of.
7

Build sections around questions, evidence, and action

Each section should explain a necessary idea, provide evidence or a credible example, guide a decision or action, or prevent a predictable mistake. For each chapter, identify the questions it answers, the claims that need support, the example that makes the idea concrete, the reader action that turns reading into progress, and the transition that makes the next chapter feel necessary.

8

Pressure-test the whole book

Read only the title, subtitle, introduction promise, chapter titles, and chapter jobs. A reader should understand the journey without seeing a single paragraph. Then test the outline against five risks and revise before assigning word counts.

  • Promise gap: is any benefit implied by the title or subtitle missing?
  • Coverage gap: is the reader asked to act without a required decision, resource, or safeguard?
  • Overlap: do multiple chapters repeat the same explanation?
  • Scope creep: does a section attract a different reader or promise a second book?
  • Evidence debt: are important claims present without a credible sourcing plan?

05 THE TOOL

The Galley chapter card

Create one card for every proposed chapter. The card is deliberately demanding: if a chapter cannot fill it clearly, the chapter is not ready to draft, and the empty fields tell you exactly what is missing.

WORKING CHAPTER TITLE

A plain statement of the stage. Clarity now; polish belongs to the title pass, much later.

CHAPTER JOB

One sentence beginning: by the end of this chapter, the reader will be able to. Use an observable verb.

READER ENTERS WITH · READER LEAVES WITH

The entry and exit condition. Together they form the chapter's own small Point A and Point B.

ESSENTIAL QUESTIONS

The three questions this chapter must answer. More than five usually means two chapters are hiding in one.

EVIDENCE PLAN

Claims requiring sources, original experience or data, places where expert review is required, and facts likely to change.

READER ACTION

The decision, exercise, or artifact that turns reading into progress.

NOT INCLUDED

What this chapter deliberately excludes, and where that material lives instead.

TRANSITION

The sentence that makes the next chapter feel necessary.

Once the cards exist, run the keep-merge-cut test: one verdict for every proposed section, no abstentions.

VERDICT
WHAT IT MEANS
Keep
Required to deliver the chapter job.
Merge
Repeats or depends heavily on another section.
Cut
Interesting, but it does not serve this reader promise.

Cut does not mean bad. Keep a parking-lot document for material that could become a bonus, an article, an email, or a future book. The manuscript is not the only place an idea can live.


06 DECISION AID

The outline validation scorecard

Score each dimension from 0 to 2. The score is a decision aid, not a sales forecast. Write one sentence of evidence beside every score in your working notes.

Promise alignment

0 · WEAK

Several chapters do not support the central promise

1 · MIXED

Most chapters relate, but some are optional or indirect

2 · STRONG

Every chapter is necessary to deliver the promise

Reader progression

0 · WEAK

Chapters are a list of topics

1 · MIXED

Some movement exists, but key dependencies are unclear

2 · STRONG

The reader moves through a deliberate sequence

Chapter distinctness

0 · WEAK

Roles overlap or repeat

1 · MIXED

Most chapters differ, with some duplication

2 · STRONG

Every chapter has one clear, unique job

Sequence logic

0 · WEAK

Order feels arbitrary

1 · MIXED

Order mostly works but has weak transitions

2 · STRONG

Each chapter prepares the reader for what follows

Scope control

0 · WEAK

The book serves multiple readers or outcomes

1 · MIXED

Boundaries exist but scope still expands in places

2 · STRONG

Reader, outcome, and exclusions are explicit

Evidence readiness

0 · WEAK

Major claims have no sourcing plan

1 · MIXED

Some sources and examples are identified

2 · STRONG

Consequential claims, examples, and freshness risks are mapped

Actionability

0 · WEAK

The reader mainly receives information

1 · MIXED

Some chapters end in decisions or actions

2 · STRONG

Every major stage produces a decision, artifact, or behavior

Navigation clarity

0 · WEAK

Titles are vague or inconsistent

1 · MIXED

Most titles are understandable

2 · STRONG

The contents page alone communicates the journey

RUNNING TOTAL
/16

13 to 16 means BUILD: the outline is ready for a chapter-by-chapter drafting plan. 9 to 12 means REVISE: keep the promise, but repair weak chapters, sequence, evidence, or scope. 0 to 8 means REBUILD: return to the reader, Point A, Point B, and the central promise. A serious accuracy, rights, safety, or feasibility problem overrides the number.

THE PRINTED EDITION

Take the chapter cards with you.

This page has the method. The designed field guide has it in the form you actually use: twenty-five pages built to be printed, with a blank chapter card for every chapter, the full outlining prompt, the scorecard, and the pre-draft gate as a working checklist. Same content, better working copy.

The download opens straight away and a copy arrives by email. We send a note when a new field guide lands, and you can leave any time.

07 IN PRACTICE

A worked outline example

TOPIC

Productivity

AUDIENCE

First-time managers of distributed teams.

PROMISE

Build a calm, repeatable weekly operating system for coordinating a remote team without filling the calendar with meetings.

POINT A · BEFORE

Newly responsible for a remote team. Updates are scattered across chat, meetings multiply, expectations stay implicit, and the manager cannot tell whether work is progressing without repeatedly asking.

POINT B · AFTER

The reader leaves with documented team agreements, a clear asynchronous update rhythm, purposeful meetings and one-to-ones, a lightweight feedback loop, and a 30-day installation plan.

The weak first outline was a subject inventory: what is remote work, leadership mindset, communication, productivity tools, meetings, company culture. Relevant subjects, every one, but the sequence never shows what the reader will build. Productivity tools could expand forever, and leadership mindset has no defined job. Rebuilt around the transformation, every chapter changes something observable.

CHAPTER
THE JOB
Introduction
Stop managing by interruption. Recognize the cost of reactive coordination, see the system to be built, and set realistic boundaries.
Chapter 1
Diagnose where coordination is breaking. Output: a one-page diagnosis and baseline.
Chapter 2
Establish the team's working agreements. Output: a draft team agreement.
Chapter 3
Build the asynchronous information rhythm. Output: a weekly update template and channel map.
Chapter 4
Redesign meetings and one-to-ones. Output: a revised meeting map and one-to-one agenda.
Chapter 5
Create feedback and accountability loops. Output: a lightweight review rhythm.
Chapter 6
Install the system over 30 days. Output: a 30-day implementation plan.
Conclusion
Protect the operating system. Reinforce the few habits that preserve clarity as the team and workload change.

Why it works: every chapter changes something observable and produces an artifact. The sequence moves from diagnosis to agreements, communication, meetings, accountability, and implementation. Tools appear only when a job requires them, never as a chapter of their own.


08 AI

Using AI without surrendering judgment

AI can produce many plausible chapter lists quickly. Plausibility is not the same as market evidence, expertise, accurate sequencing, or a promise your book can fulfill. Use it for divergence and critique, and keep every decision.

USE AI TO

Generate alternative structures. Cluster an unorganized question list. Identify possible prerequisites and overlap. Suggest reader objections you have missed. Convert a finished chapter plan into a drafting checklist.

DO NOT USE AI TO

Decide the book's reader or promise. Establish your expertise. Prove demand or factual truth. Invent sources, findings, or quotations. Every one of those jobs is yours.

Whatever the model proposes, run a human review pass before anything enters the outline. Delete what does not serve the validated promise. Replace generic material with your evidence and point of view. Verify the sequence against real reader prerequisites. Mark every factual claim that needs independent research. Rewrite chapter jobs in language you would genuinely use.

On disclosure, Amazon distinguishes AI-generated content from AI-assisted work. Brainstorming structure with AI while you create the final material yourself is AI-assisted. If AI creates actual text, images, or translations used in the book, KDP requires disclosure during publishing, even when the output is substantially edited. The publisher stays responsible for accuracy, rights, and customer experience.


09 WATCH OUT

Common outlining mistakes

Treating the contents page as a subject inventory

Relevance is not enough. Each chapter needs a unique contribution to the promised result.

Beginning with chapter titles instead of the reader

Attractive titles cannot repair an undefined Point A, Point B, or audience.

Copying the market leader's structure

This reproduces the existing product and may bring you close to protected expression or a proprietary framework. Study the market for expectations and gaps, then build your own route.

Asking AI for an outline from one vague sentence

The model fills missing context with common patterns. Supply the reader, promise, constraints, and research first.

Letting the introduction become a memoir

Use personal experience to establish relevance, not to delay the reader's progress.

Hiding the difficult part in one late chapter

If the promise depends on a major constraint, cost, or prerequisite, surface it early enough to inform decisions.

Creating an everything-else chapter

Titles like Other Tips collect material that lacks a job. Reassign, convert, or cut it.

Researching only after the prose is written

This creates evidence debt. Mark claims, sources, permissions, and freshness risks in the outline itself.


10 FAST START

The 45-minute outline sprint

This produces a decision-ready first outline, not a final manuscript plan. Use it after the niche and reader promise have passed validation, and keep the clock honest.

0 TO 5 MIN · WRITE THE CONTRACT

Name the specific reader. Write Point A and Point B in one sentence each. Write the central promise. List three scope exclusions.

5 TO 15 MIN · COLLECT REQUIRED QUESTIONS

List decisions, prerequisites, obstacles, and likely failures. Add repeated needs from competitor reviews or customer conversations. Mark factual and changeable claims.

15 TO 25 MIN · CLUSTER AND SEQUENCE

Group related questions by the job they perform. Order the groups by dependency. Merge duplicate stages, and move reference material out of the main journey.

25 TO 35 MIN · CREATE CHAPTER CARDS

Write one chapter job for every stage. Add the reader entry and exit condition, three essential questions, and one practical output.

35 TO 40 MIN · RUN THE GAP SCAN

Compare the outline with the title and promise. Look for missing prerequisites and repeated jobs. Identify evidence, permissions, or expert review you will need.

40 TO 45 MIN · SCORE THE OUTLINE

Complete the eight-part scorecard. Choose BUILD, REVISE, or REBUILD, and write the next three actions before drafting.

If the result is REVISE, repair the weakest dimension first. Adding more chapters is rarely the answer.


11 CARRY IT FORWARD

From outline to KDP-ready navigation

The editorial hierarchy you create now becomes part of the reader's navigation later. Amazon states that Kindle eBooks with chapters need a working logical table of contents accessible from the navigation menu, and strongly recommends a visible HTML contents page. Proper chapter heading styles make that navigation nearly automatic, and a reflowable eBook cannot rely on fixed print page numbers, so the structure has to carry the reader. Before calling the outline done, close the pre-draft gate.

  • The intended reader can recognize themselves in one sentence.
  • Point A and Point B describe observable conditions.
  • The central promise is useful, bounded, and credible.
  • Every chapter has one unique job.
  • The sequence follows reader dependencies.
  • Every promised benefit has a home.
  • Overlapping and out-of-scope sections have been removed.
  • Consequential claims have an evidence plan.
  • Each major stage produces a decision, action, or useful artifact.
  • The contents page alone communicates a coherent journey.
  • Chapter titles are ready to become real navigation headings.
  • The outline scores at least 13 of 16 with no overriding risk.

A blank page is not the real enemy. An undefined chapter is.

The sprint takes 45 minutes, and the full method perhaps an evening. That is the honest cost of structure, and it is worth paying, because it assumes the idea underneath has already been validated. Structure built on an untested idea is still a gamble, just a tidier one.

Galley runs the page-one checks and returns a verdict in about a minute. ENTER, CAUTION, or AVOID, with the numbers behind it, and a validated idea carries its reader, gap, and promise straight into a structured Book Blueprint, where every chapter inherits a job and every important claim gets an evidence plan.

Free. No card.

12 REFERENCE

Frequently asked questions

How many chapters should a nonfiction book have?
There is no universal correct number. Use as many chapters as the reader needs to move through the promised transformation without padding, repetition, or missing prerequisites. A tightly focused guide may need only a few stages; a complex subject may require more depth and safeguards.
How detailed should a nonfiction book outline be?
Detailed enough that each drafting session has a clear assignment. At minimum, define every chapter's job, entry and exit condition, essential questions, evidence needs, practical output, and transition. It should reduce blank-page decisions without preventing discoveries during writing.
Should I outline the introduction first?
Plan the opening's job early, but consider finalizing it after the body is stable. The finished introduction must accurately describe the journey the manuscript actually delivers, and that journey often shifts during drafting.
Can I change my outline while writing the book?
Yes. An outline is a controlled hypothesis about the best reader journey. Revise it when research reveals a missing prerequisite, a better sequence, or unnecessary scope, then update the master outline before continuing so the book stays one coherent plan.
Can AI create my nonfiction book outline?
AI can generate alternatives, cluster questions, and critique structure. It cannot validate demand, guarantee accuracy, establish your expertise, or decide what promise you can responsibly fulfill. Treat its output as proposals requiring human judgment, and disclose AI-generated content where KDP requires it.
Is it OK to study competing books when outlining?
Yes, when the purpose is understanding reader expectations, common coverage, and unmet needs. Do not copy wording, examples, chapter names, distinctive structures, or proprietary frameworks. Your outline should emerge from your own promise, research, and contribution.
Does a Kindle eBook need a table of contents?
Amazon requires a working logical table of contents for Kindle eBooks with chapters or sections and strongly recommends a visible HTML contents page. Proper chapter heading styles in the manuscript make reliable navigation easier to create and test.
Should I set a word count before outlining?
Use a target as a production constraint, not a reason to pad. First establish the required journey, then assign a word budget by the complexity of each stage. If required coverage exceeds a useful scope, narrow the promise or plan a series.

Every chapter gets a job before it gets a word count.

A defined chapter tells you what to write next, every session. The pre-draft gate tells you when you have earned the right to start.

SOURCES AND OFFICIAL REFERENCES
  • Amazon KDP Content Guidelines

    AI-generated content disclosure, rights, accuracy, and customer-experience guidance.

  • Amazon KDP: Table of Contents with a Navigation Document

    Current logical and HTML table of contents requirements and recommendations.

  • Amazon KDP eBook Manuscript Formatting Guide

    Using Heading 1 for chapter titles and preparing a contents page.

  • Amazon KDP: Prepare Books with Kindle Create

    Chapter detection, navigation, front matter, and book structure.

  • Amazon KDP: Create a Table of Contents

    How chapter styles and Kindle Create support interactive navigation.

Recheck platform guidance at publication and on a maintenance schedule, because KDP policies, tools, and interfaces can change.