How long does the whole thing take?

A realistic first-book timeline: three to six months of drafting at 500 to 1,000 words a day, one to two months of revision with beta readers, a month of production and store setup, then the store's review. Six months to a year if nothing goes wrong; closer to two if life interferes. What stretches it is not the writing but the gaps — a draft touched once a month takes years, because every session starts with re-reading.

How many words should my book be?

Match the band readers of your category already buy: adult fiction runs roughly 70,000 to 100,000 words, romance and cosy crime sit in the lower half of that, epic fantasy comfortably past 100,000, general nonfiction at 50,000 to 80,000. The full table is in the length guide.

Should I fix things while I draft?

No. The draft's job is to exist. Fixing chapter one until it shines is how books die in chapter three — the polish never ends, so the draft never does. Write to the end, then fix. A finished draft with problems is a normal intermediate state; an unfinished draft with beautiful prose is a hobby.

Do I need to hire an editor?

Hire a copyeditor if you can — it is the pass almost every self-published book benefits from, because you cannot proofread your own book reliably. A developmental editor (story-level) is worth it if your beta readers found structural problems you could not fix yourself. Proofreading, the final pass, can follow the copyedit or be bundled with it.

Can I publish my first draft?

No. A first draft is the story you told yourself, and it contains things only you can forgive: chapters that exist to find out what happens, characters who change names halfway, an ending that arrived before the middle was settled. Readers forgive almost anything except the sense that nobody reread the book.

How to write a book (and self-publish it)

Writing a book and self-publishing it is one long process with six phases, and the phases happen in order: decide, draft, revise, produce, publish, then live with the book. Each phase below is a set of steps you can actually follow. The publishing steps link out to guides that go deep; the writing steps are all here.

Two things to know before the first step. First: the order matters more than any single step — most books that fail were written out of order, not badly. Second: the times given assume evenings-and-weekends pace, roughly 500 to 1,000 words a day when drafting. Scale to your life.

Phase 1: Decide what you are writing

Twenty minutes of decisions here saves months later, because these are the choices that are cheap now and expensive to reverse.

Pick a category you actually read
Not “a genre that sells” — a category whose books you have read dozens of, because you will be held to expectations you can only know from the inside. Romance readers expect the central relationship to carry the book to its end; mystery readers expect to be able to solve it. Where the book will sit in the store is decided by the same choice — the mechanics are in categories and keywords, and they are worth reading now, before the draft, because they describe your reader.
Set a word target for the finished book
Pick the middle of your category’s usual band and write it down. Adult fiction: about 85,000. Epic fantasy: 110,000 or more. General nonfiction: 65,000. The full table is in how long should a book be. The number is a steering target, not a cage — but without one, the middle of the book grows to fill whatever time you give it, and length sets printing cost and platform limits later (see the page count estimate).
Decide: one book, or the start of a series
You can change your mind later, but decide now, because it changes what book one must set up and how its ending is allowed to land. A standalone can close every question. A series opener has to close this book’s question while opening a bigger one.
Choose a point of view and a tense, and write them down
Third person past is the default of English fiction and a fine choice. First person present is a real choice too. What is not a choice is drifting between them, which readers notice on every page it happens. One viewpoint per book, or one per chapter as a deliberate system — put the decision on page one of your notebook.
Open a story bible
A text file is enough. From day one, record: every named character with a one-line description, every place, every date or distance the plot depends on, and who knows what secret at what point. This file is what saves you in revision, when you cannot remember whether the innkeeper was called Marta in chapter four.

Nonfiction substitutes its own list: define the promise the book makes to the reader, list the steps a reader needs in the order they need them, and make each chapter one step. Everything else below applies unchanged.

Phase 2: Write the first draft

One rule governs this whole phase: finish. The draft is allowed to be bad. It is not allowed to be abandoned. Everything below serves that.

Set a session target and count words, not hours
500 words a day finishes a novel in about five months; 1,000 in under three. Pick the number you can hit on a bad day, not a good one, and track it. A calendar with a word count in each box is the cheapest motivation system that works.
Start each session by rereading the last few paragraphs — then stop
Rereading finds the voice; it is not an invitation to fix yesterday. Do not open chapter one until the draft is done. Every hour spent polishing an unfinished draft is borrowed against the ending, which does not exist yet.
Write in scenes
A scene is one unit: a character wants something concrete on this afternoon, meets resistance, and leaves worse off or changed. If a scene ends with everything as it began, it is a candidate to cut — or to make cost something. Between scenes, give the character a beat to decide what to do next; that decision is what turns events into a story.
When you hit a hole, write a bracket and keep moving
[inn name], [the law in 1890], [how long the crossing takes]. Research is a scheduled job, not a mid-scene detour, and the bracket is a promise to your future self that the hole is known. You will collect twenty of these in a novel and clear them in an afternoon.
Mark scene changes with a blank line
In a manuscript, that is all the mark needed. What a printed book does with that line — and why a blank line alone fails at the foot of a page — is a typesetting question, covered in the scene-break guide. It is not your problem yet.
Draft in .docx or Markdown, with real heading styles
Chapter titles get Heading 1 in Word, or # in Markdown — not big bold text. This one habit is the difference between a manuscript that imports cleanly into every tool downstream and one that has to be reconstructed. The two-hour version is in preparing a Word manuscript. In Nueel the Writing screen is the manuscript, so this step is already done.

When the draft ends — and it should end imperfectly — stop. Do not send it to anyone. Sit on it for at least two weeks, ideally four. You cannot see the structure of a book you are still inside; the cooling gap is what lets the next phase work.

Phase 3: Revise it — three passes, in order

Revision is three different jobs, and the order matters because doing them out of sequence multiplies the work: line-editing a scene you are about to cut is paying for words twice.

Pass one: read for story only
Read fast, on paper or a different screen, and write in the margin: slow, confusing, don’t believe it. Then fix at the scene level — reorder, cut, merge, and rewrite the scenes that don’t work. Check five things: the opening makes a promise; the middle escalates instead of repeating; the turning points turn; the ending answers the question the opening asked; and every scene changed something. Sentences do not exist in this pass.
Send it to beta readers
Three to five people who read your category, after pass one and before the line pass. Ask specific questions: where did you stop reading, whose motives were unclear, which chapter dragged, what surprised you that shouldn’t have. Do not ask “did you like it” — the answer is useless. Expect to make another structural pass from what they say; that is normal, not failure.
Pass two: sentences
Now the line work: clarity first, then rhythm, then word choice. Read the whole book aloud — your ear catches what your eye forgives. Kill repeated constructions, the words you lean on without noticing, and any sentence doing work the paragraph does not need. Still no proofreading.
Pass three: proof
Typos, punctuation, doubled words, the timeline, the eye colours, the names. The story bible from phase 1 is now worth its weight. Do this last, on a printed or exported copy, because every earlier pass invalidates it.
Hire an editor, if the budget allows
Editing is three trades. A copyeditor works on the sentences and catches what you cannot see in your own book; this is the one pass worth paying for on almost every self-published book. A developmental editor works on the story — worth it if the beta round found problems you could not fix alone. Proofreading is the final surface pass, after the layout is settled. The advance reader copies for your wider list come after all of this, before launch.

Phase 4: Turn the manuscript into books

The manuscript becomes two products — an EPUB for the ebook and a print-ready PDF for the paperback. They are not the same file in different wrappers; ebook, print, or both? explains what each one controls and what the reader controls.

If you are still choosing the tool for this phase rather than following along inside one, the whole market is judged on the same criteria in the best book formatting software in 2026.

Assemble the front matter
Title page, copyright page, dedication, contents — in the conventional order, which the front matter guide lays out page by page. This is done once and inherited by every edition.
Choose the trim size
The physical size of the page, and the first decision everything print depends on. 6 × 9 is the most common paperback trim in the US; the genre conventions and the trade-offs are in choosing a trim size.
Export the print interior and check it against the store's rules
Fonts embedded, single pages, margins by page count — the checklist is in what a print-ready PDF is, and KDP’s numbers specifically in formatting for KDP. A store checks these at upload; you want to check them first.
Export the EPUB and test it
Open it in a real reading system — a phone, a tablet — not just on your computer. The testing pass is in testing your ebook. If a store refuses the file rather than rendering it oddly, why your EPUB was rejected covers the usual causes one by one.
Build the covers — last
An image for the ebook listing; a full wrap (back, spine, front) for print. The wrap needs the final page count, because the spine width comes from it — which is why the cover is the last thing made, not the first. The arithmetic is in setting up a cover file, and the spine width calculator does it for you.
Enter the book's details
Title, series, description, ISBN. The ISBN is a decision, not a field: a free platform ISBN locks the edition to that platform, and your own — from your national agency — keeps every channel open. The whole arithmetic is in do you need an ISBN?. The description is sales copy for the product page, written for the buyer; the book description guide covers the shape of it.

Phase 5: Publish

Upload to your first store — usually Amazon KDP
Account, tax interview, metadata, files, price. The whole path, including the decisions that are hard to reverse — ISBN, exclusivity, pricing band — is walked end to end in publishing on Amazon.
Order a proof of the print edition before you publish
It costs print price, arrives in days, and is the only version of the check that involves your hands: the cover finish, the ink, the binding, the margins at actual size. Almost every print complaint that reaches a store could have been caught here.
Go wide, after the launch settles
Apple Books, Kobo, Google Play and the library channels take the same EPUB; the direct-versus- aggregator question is covered in beyond Amazon. There is no rush — a week after launch day is fine, and the book’s page on each store is a form, not a project.

Phase 6: Market the book

Publishing puts the book on the shelf. It brings no readers to the shelf. For a first book, marketing is not advertising — it is reviews, a page that converts, and thirty days of deliberate noise. Budget little money here; budget attention.

Send ARCs early, chase reviews after launch
Two to three months before launch, send advance reader copies to bloggers, reviewers and newsletter writers in your category — they schedule weeks ahead. After launch, ask every early reader for an honest review where they bought the book, one by one. Ten genuine reviews in the first weeks matter more than a hundred social posts: they are what the next browser reads before the description.
Make the product page convert before you send traffic
No promotion survives a weak page. The cover earns the click, the description closes it, and categories and keywords decide who ever sees it. Check the page on a phone, where most readers arrive: first two sentences of the description, price, cover thumbnail. Fix those before telling anyone the book exists.
Plan the first thirty days, not just the launch day
Launch day is one spike; stores reward sustained velocity. Line up, in order: your own list (friends, newsletter, socials) in week one; category newsletters and promo sites in week two; a launch price you can raise later, or a Select free run if the ebook is enrolled in KDP Select. Tell readers exactly where to buy and what to do — “out today, honest reviews welcome” beats a link alone.
Play the long tail
Revisit categories and keywords once the sales pattern settles — the first guess is rarely the last, and they stay editable after launch. Expect the money to lag: royalties settle about 60 days after the month they were earned, so day-one sales are a signal, not a verdict. Then start phase 1 again: nothing markets book one like book two, and the series page is the cheapest discovery engine self-publishing has.

The whole process is long, but no single step in it is hard, and the sequence — decide, draft, revise, produce, publish, market — is what separates books that ship from books that stall. The expensive failures are always sequencing failures: the cover before the page count, the polish before the ending, the store account after the launch day.

Nueel exists for phases 4 and 5: it takes the finished manuscript and produces the print-ready PDF and the EPUB the stores ask for, checked against each store’s rules on the way out — so the writing keeps your attention, which is where it does the most good.

Common questions

How long does the whole thing take?: A realistic first-book timeline: three to six months of drafting at 500 to 1,000 words a day, one to two months of revision with beta readers, a month of production and store setup, then the store's review. Six months to a year if nothing goes wrong; closer to two if life interferes. What stretches it is not the writing but the gaps — a draft touched once a month takes years, because every session starts with re-reading.
How many words should my book be?: Match the band readers of your category already buy: adult fiction runs roughly 70,000 to 100,000 words, romance and cosy crime sit in the lower half of that, epic fantasy comfortably past 100,000, general nonfiction at 50,000 to 80,000. The full table is in the length guide.
Should I fix things while I draft?: No. The draft's job is to exist. Fixing chapter one until it shines is how books die in chapter three — the polish never ends, so the draft never does. Write to the end, then fix. A finished draft with problems is a normal intermediate state; an unfinished draft with beautiful prose is a hobby.
Do I need to hire an editor?: Hire a copyeditor if you can — it is the pass almost every self-published book benefits from, because you cannot proofread your own book reliably. A developmental editor (story-level) is worth it if your beta readers found structural problems you could not fix yourself. Proofreading, the final pass, can follow the copyedit or be bundled with it.
Can I publish my first draft?: No. A first draft is the story you told yourself, and it contains things only you can forgive: chapters that exist to find out what happens, characters who change names halfway, an ending that arrived before the middle was settled. Readers forgive almost anything except the sense that nobody reread the book.