How to self-publish a book: a real person's step-by-step guide.

Photo by Pexels

I am a developmental editor. That means I get paid to tell people their books need work. Writing a guide on how to self-publish is, professionally speaking, like a dentist writing a guide on flossing. Conflict of interest noted. I am also entirely correct. Let us proceed.

Self-publishing means you own the whole pipeline. Writing, editing, cover design, formatting, distribution, marketing, pricing. What a traditional publisher does for you, you now do yourself. Or hire out. Or do a bit of both and occasionally ring me at 9pm asking why your ISBN is showing the wrong edition. (It is always the wrong edition.)

The quick answer: Finish the manuscript properly. Have it edited in three passes. Commission a cover that looks like it belongs next to the genre's current bestsellers, not alongside clipart from 2003. Format into two separate files (print and ebook are different things, and yes, that matters). Write a description that sells rather than summarises. Choose a distributor. Set your price between $2.99 and $4.99 for the ebook. Push publish. Four to eight months from manuscript to live book. Between $1,000 and $5,800, depending on the state of the draft.

The rest of this guide walks each step in the order it needs to happen. My apprentice Jake finds the distribution section "genuinely tedious." He said this to an author whose book was not showing up in the library system. He has since revised his opinion somewhat.

Should you self-publish?

Worth asking before you spend money: is this actually the right path, or are you here because a few rejection letters have made traditional publishing feel unreachable?

Both are valid reasons to self-publish, for the record. But they lead to different expectations, so it helps to know which one applies to you.

Self-publishing probably makes sense if:

• You want complete creative control over your cover, content, price, and timeline
• You are writing in a genre where indie publishing is well-established and well-respected: romance, thrillers, fantasy, science fiction, cozy mysteries, most narrative non-fiction
• You have an existing platform or audience, meaning a readership that does not have to be built entirely from zero
• You want 35 to 70% ebook royalties instead of 10 to 15%
• You want to publish in months, not years
• You are building a backlist across multiple titles over time

Traditional publishing might suit you better if your primary goal is a major-imprint spine in an airport bookshop, or if you are writing literary fiction where imprint prestige matters for career progression. The marketing support from traditional publishers varies a lot by deal size. "The publisher will handle marketing" is not a certainty at most deal levels. It is a hope, and sometimes it is justified.

One thing that is not a good reason to self-publish: the idea that you can skip the editorial process. Nine out of ten self-published books that earn bad reviews share the same problem. Not badly formatted. Not mis-priced. Under-edited. I am a developmental editor saying this. The conflict of interest is obvious. I am also comfortable making the claim after watching the pattern repeat across three years of manuscripts. The Alliance of Independent Authors publishes annual data on what separates successful indie authors from unsuccessful ones. Editorial investment is consistently near the top of the list.

Self-publishing vs. traditional publishing

Let us do this comparison properly, because "indie is always better" and "traditional is always better" are both wrong and both serve somebody's talking points.

The royalty maths is where most authors focus, and it is worth being precise: 70% of $3.99 on a self-published ebook is $2.79 per copy sold. A traditional ebook royalty of 15% of net on a $9.99 ebook is roughly $1.05 per copy. The gap is real.

For print, the maths is tighter. A traditionally published author with a decent advance and strong bookstore distribution can move more physical copies than most self-published authors. Where self-publishing compounds over time is in ebook backlist: each new title markets all the previous ones, and the higher royalty rate means backlist income builds faster than it does on a traditional royalty statement.

Neither path guarantees a living. The variable that matters most in either case is whether the book is genuinely good and whether people can find it. Self-publishing does not change those two requirements. It just changes who is responsible for meeting them.

How much does it cost to self-publish a book?

The honest answer: as much as you let it. Cutting costs in the wrong places is significantly more expensive in the long run than doing it properly the first time.

Priority order matters more than the total. Editing comes first. Always. A beautiful cover on an under-edited manuscript is, as I keep telling Jake, like putting a bow tie on a golden retriever. Adorable. Ineffective. The golden retriever has not become more qualified for the job.

You can legitimately trim on formatting (free tools exist and are genuinely good) and on ISBNs (using a platform's free ISBN is fine). You cannot trim on editing and expect the book to absorb the shortcut invisibly. Readers notice. Reviewers notice. The algorithm that influences whether your book gets recommended notices, because a four-star average and a two-star average behave very differently in terms of what the platform shows to whom.

Photo by Pexels

How to self-publish a book: the 9 steps

Here they are in the order they need to happen. Skipping steps is allowed. Reversing them costs money. In my experience, every author learns this exactly once.

1 Know your reader and your genre

Most self-publishing guides put this step last, under "marketing." That is exactly backwards. Knowing who your reader is and what they expect from your genre is not a marketing decision. It is a writing decision, a cover decision, a pricing decision, and a distribution decision all at once. You cannot discover these things after the fact and retrofit them.

Spend an afternoon on Amazon in your specific genre and subgenre. Not the literary classics. The books selling well right now. Look at:

• What the covers have in common, not what makes each one individual
• What the descriptions are doing in their first two sentences
• Which categories the current bestsellers are listed in
• What the one- and two-star reviews say. Those tell you exactly what this readership does not forgive.

None of this means writing to a formula. It means your book needs to signal, at a glance, that it belongs in the conversation the genre is already having. Readers recognise genre cues visually and in prose style. When a book breaks those cues without earning trust first, readers experience it as a broken promise. Not an artistic statement. A broken promise.

Know your genre the way a musician knows their genre. Then make your choices inside it deliberately, with full awareness of what you are playing toward and what you are departing from.

2 Finish your manuscript

Not "mostly done" finish. Not "I'll fix it in editing" finish. Actually done.

"I can fix it in editing" is the most expensive sentence in self-publishing. Professional editing is priced on word count and on the amount of work required. A manuscript needing significant structural revision costs more and takes longer. That delays the cover designer, the formatter, and the launch. Rushing an unfinished draft into the production pipeline is how a $1,500 editorial invoice becomes $3,000, with four extra months added to the timeline and a somewhat strained working relationship with your editor. I am speaking from experience. Both ends of it.

Before you hire an editor, get the manuscript in front of beta readers. Not your spouse. Not your best friend, unless they are the rare kind who will tell you, with genuine warmth, that chapter twelve does not work and your protagonist goes passive for sixty pages in the middle of act two. (If you have a friend like that, keep them. They are doing more useful work than they probably realise.)

Beta readers who are actual readers in your genre are the most useful. They tell you whether the book is working at the reader experience level, which is the only level that matters. Your editor then does the professional deep dive. The goal is to walk into developmental editing with a draft you genuinely cannot improve on your own. Not a draft you are afraid to look at. One you are proud of, that still has a professional editing pass left to give it.

3 Edit in three passes

This is the step I have the most to say about. Partly because editing is my job. Partly because it gets skipped more than any other step. Both of those things are probably related.

Three layers. They do not substitute for each other.

Developmental editing works at the big picture. Structure, pacing, character arc, point-of-view consistency, plot logic, and the ending. Does the story work as a whole? Does the reader experience it the way you intended? This is the layer that catches "the middle drags," "your protagonist is passive for sixty pages," and "the twist lands but does not pay off what you set up in act one." It happens first, before any sentence-level work, because there is no point polishing sentences in a chapter that is about to change significantly.

Copy editing works at sentence and paragraph level. Repetition, clarity, word choice consistency (your character's eyes cannot be brown in chapter three and green in chapter fourteen, unless that is the twist, in which case: well played), grammar, style, and readability. Copy editing is not "fixing typos." It is a full manuscript review from someone paying close attention to how your prose actually reads, line by line.

Proofreading catches what everything else missed. The typo your brain autocorrects every single time. The chapter header in the wrong font. The "teh" that survived twelve rounds of revision because your eye never sees it. This is the last pass before the file goes to the formatter.

Do them in this order. Reverse them and you are paying to refine a draft that is about to change substantially. That is not a good use of anyone's time or money.

For a full-length novel: all three. For a shorter non-fiction book with clean logical structure: developmental editing and a proofread at minimum, copy editing if the prose is doing significant work. For a novella: copy editing and proofread, plus a developmental consult if the structure has any real complexity.

I am a developmental editor recommending developmental editing. That conflict of interest is on the table, in writing. I am also comfortable making the recommendation after watching, for three years, what happens to books that skipped it. The pattern is consistent enough to be worth saying plainly. If you want to see what a story edit from BlackPen actually covers, the pricing page has the full breakdown.

Photo by Pexels

4 Design your cover

Your cover has three jobs: signal genre, communicate quality, and give a browser a reason to click. Notice what is not on that list: represent every subplot, capture the full emotional arc of the novel, or look exactly the way you imagined it at 11pm during chapter four.

The cover is marketing. It does its work before the reader has read a single word of your prose. And readers are genuinely good at using covers as signals. They have been trained by years of browsing. When your cover breaks genre cues, the brain registers "not the book I was looking for" before the conscious mind gets involved. That is the threshold you are competing at. A fraction of a second, thumbnail size, in a list of forty results.

• Spend an afternoon looking at top-selling covers in your specific subgenre. Look for what they share, not what makes each unique. Design toward those conventions from the inside.
• Hire a designer whose portfolio shows your genre. "Good designer" and "right designer for your book" are not always the same person. A literary fiction specialist may not know what contemporary romance is supposed to signal.
• Test the cover at thumbnail size. In most online storefronts your cover displays at roughly 90 pixels wide. If the title is unreadable at that size, it is not doing its job.
• Budget $200 to $800. Premade covers from genre-savvy services are a legitimate option for debut authors on tighter budgets. Custom work gives you more differentiation once you have an audience worth differentiating for.

Save the artistic departure from genre conventions for when your readership already trusts you enough to follow you somewhere unfamiliar. Before that trust is built, departing from conventions reads as a cover that does not know what it is.

5 Format your interior

Three solid tools cover most situations. Vellum (Mac only, one-time purchase, elegant output). Atticus (cross-platform, subscription, handles both ebook and print well). Reedsy's free formatter (genuinely capable for a free tool, handles both formats). The goal is readable, well-structured pages that do not look like a Word document that got exported and hoped nobody would notice.

A few details that trip people up:

Print and ebook are separate files. Your print interior is a PDF with margins calculated from page count and binding type. Your ebook is a reflowable EPUB where the reader controls font size, line spacing, and screen brightness. Do not try to use one file for both. Distributors will accept it. Readers will notice what happened. That is considerably worse.

Front matter and back matter matter more than most people think. Front: title page, copyright page, dedication, table of contents. Back: author bio, acknowledgments, and, critically, a preview of your next book and a link to your newsletter. Your back matter is where a reader who just finished your book and loved it gets directed toward everything else you have written. That conversion opportunity disappears if you end on the final chapter and close the file. Do not leave it blank.

Scene breaks and chapter headings should be consistent throughout. Run a final check before you hand the file to your formatter. They will make it look beautiful. They cannot fix inconsistent source material, and they will cheerfully make every inconsistency beautiful in exactly the wrong way.

6 Write your book description

The step that most self-publishing guides skip or treat as an afterthought, which is interesting because it is one of the most important pieces of writing in the whole process. Most authors who can write 80,000 words of excellent fiction find 150 words of blurb completely defeating. There are good reasons for this.

You spent a year or more with this story. You know every layer of it. Writing a description means choosing three things to emphasise and leaving everything else out. That selection is genuinely hard when you care about all of it equally. It is also not optional.

A framework that works across most genres:

Open with the hook. One sentence. Who is at the centre of the story and what situation they are in, written with enough specificity to create immediate tension. Not a rhetorical question. Questions as opening lines have been somewhat overdone and tend to land as soft starts. A declarative statement with stakes already built in.

Establish what they stand to lose. The concrete thing your protagonist risks by engaging with the central conflict. "Everything she holds dear" is not a stake. "Her daughter's custody and the one remaining client who might save her law practice" is a stake. Specific always beats vague.

Close with tone. One line that communicates how the book feels. Thrillers end with urgency. Romance ends with anticipation. Literary fiction ends with a question that lingers. This line does not need to be clever. It needs to be accurate.

Do not summarise the plot. Do not give away the resolution. Do not write the description the way you would explain the book at dinner, with "so basically what happens is" followed by three paragraphs of context. Write the way the best back-cover copy feels: compelling, specific, and just incomplete enough that the only way to resolve the tension is to buy the book.

If you cannot make it work after three or four attempts, hire a blurb writer. It is a genuine specialisation. It is also not expensive relative to how important the task is.

Photo by Pexels

7 Get your ISBN and choose your distributor

An ISBN is your book's identity number in the retail ecosystem. It tells a bookshop, a library, and a distributor's system that a specific edition of your specific book exists and where to find it. You need one per format (print is a different ISBN from ebook) and new editions count separately.

Three options:

Amazon KDP's free ISBN works and costs nothing. The catch: it ties that ISBN to Amazon's system. If you later want to use IngramSpark for print distribution, you will need a new ISBN for that edition. Fine if you are publishing exclusively through KDP and plan to stay there.

IngramSpark's free ISBN is the same situation in reverse. Their ISBN, their platform. Useful if IngramSpark is your primary print distributor.

Your own ISBN through Bowker: $125 for one, $295 for ten. Yours permanently, independent of any platform. If you plan to publish more than one book, ten for $295 is obvious arithmetic. You can move between distributors, update editions, and expand formats without any platform holding the identity number.

For distribution itself, the main options:

• Amazon KDP is the largest single market. Up to 70% ebook royalties in most territories. Print-on-demand built in. KDP Select enrollment (90-day exclusive windows) gives access to Kindle Unlimited, which pays per page read and suits high-volume genre fiction very well. The trade-off: KDP Select requires ebook exclusivity to Amazon during each window.
• IngramSpark is how you get print books into physical bookstores and libraries. Most independent bookshops and libraries order through Ingram's distribution system. If you want your print edition on shelves beyond Amazon, this is the path.
• Draft2Digital distributes ebooks to Apple Books, Kobo, Barnes and Noble, and about a dozen other storefronts without requiring you to manage fifteen separate accounts. They take a percentage, but the time they save is worth it for most authors.
• Smashwords does similar work to Draft2Digital, with its own direct storefront and an established community of indie readers who buy there specifically.

Most authors serious about building a readership use a combination: KDP for Amazon ebooks and print, IngramSpark for bookstore print distribution, Draft2Digital for wide ebook distribution. That covers most of the market without giving any single platform control over your whole catalogue.

Jake called the platform research "genuinely tedious." He is currently explaining to an author why her book is not available in the library system. We are working through the lesson together. Slowly.

8 Set your price and publish

Pricing generates strong opinions and a surprising amount of folklore. Here is the practical version.

For ebooks, $2.99 to $4.99 is the effective range for debut and early-catalogue fiction. Below $2.99, Amazon's royalty rate drops from 70% to 35%. That is a significant difference. $0.99 signals "basically free" and tends to attract readers who collect rather than read. $7.99 to $9.99 is defensible once you have an audience and a backlist doing the legitimising work alongside it.

For print, let your distributor calculate the minimum viable price from page count and production costs. A 300-page paperback through KDP typically costs $3.50 to $4.50 to print. Add $4 to $7 margin and you are at a competitive retail price while earning something meaningful per copy sold. A $0.00 royalty is technically a published book. It is not a business.

One tactic worth knowing: price the ebook at $0.99 for the first seven to ten days to drive volume and improve the chance of early reviews, then move to the permanent price. Does not work for every book in every genre. Works often enough that it is worth understanding before you decide against it.

Fill in your metadata carefully when you upload: title, author name, categories, keywords, and your book description. Your categories and keywords determine which search results your book appears in and which Amazon recommendation emails it gets included in. Under-optimised metadata is invisible money. Tools like Publisher Rocket, or a few hours of manual genre research browsing your category's top sellers, give you better results than guessing.

Preview the book before you push publish. Print books go live within 24 to 72 hours. Ebooks on Amazon can be live within 24. Changes after publication are possible but disruptive. Get it right before the button.

Photo by Pexels

9 Market and launch your book

The step most new self-publishers under-budget for, in both money and time, then wonder why the book is not finding readers. Publishing the book is the beginning of the sales process, not the end. That is not intuitive after two years of writing when "published" looked like the finish line. It is not.

Build your email list before launch day. An email list is the only direct communication channel you own outright. Social media platforms change their algorithms, reduce organic reach, and occasionally become considerably less useful without warning. Your list does not. A simple landing page with a short story, a deleted chapter, or a first-chapter preview as the sign-up incentive is enough to start. Build it before you need it.

Get advance reader copies out early. Reviews matter at launch, not because a single review moves the needle, but because a book with zero reviews on its first day in the store is genuinely hard to sell. Readers use review count and average as a quality signal. Send ARCs to beta readers, to ARC services like BookSirens and NetGalley Indie, and to your reader community, four to six weeks before you publish. Ask for honest reviews, not positive ones. Honest reviews from real readers carry more credibility and more weight.

Research your categories before you upload. Your Amazon categories determine which charts you can rank on. Getting into a relevant but less competitive category can put you on a bestseller list, which affects visibility in a meaningful way. This is not gaming the system. It is understanding how the system works.

Consider a pre-order period. Two to four weeks of pre-order before release concentrates your launch-day sales into a single day rather than spreading them across a week. That concentration tends to help with chart placement and visibility on Amazon.

Write the next book. I know that is not what anyone wants to hear after finishing this one. Jake made a face when I said it. But every survey of indie author income tells the same story: authors with three or more titles earn five to ten times more per book than single-title authors. Each new title markets the previous ones. A reader who loved book one has somewhere to go. The long game in self-publishing is a catalogue game, not a single-launch game. The sooner you start the next one, the sooner everything else compounds.

How to self-publish a novella

Novellas, roughly 20,000 to 40,000 words, sit in an interesting middle zone. Too short to feel like a "full book" to some readers, too long to give away for free without real editorial investment behind it, and genuinely excellent for certain publishing strategies that full-length novels cannot replicate.

The nine steps above apply. Two practical differences worth knowing:

Pricing is different. Novellas typically sell for $0.99 to $2.99 as ebooks. A print novella under 40,000 words is hard to price competitively because printing costs consume too much of the available margin. Many novella authors go ebook-only, or bundle the novella into a collection. Both are entirely legitimate. Neither is a workaround you should feel embarrassed about.

Series strategy is where novellas work best. Write three interconnected novellas, price the first at free or $0.99, and let readers self-select into the series. This is one of the most effective reader-acquisition strategies in indie publishing, and it works precisely because a novella is short enough to read in a single sitting. A reader who commits to one and finishes it in an evening has very low friction to buy the next one. That low-friction path through a series is harder to build with full-length novels.

Editing is not optional for a novella either. Shorter does not mean simpler. A 25,000-word novella that does not work structurally is a 25,000-word disappointment. A well-edited one is a satisfying complete experience. The ratio of editorial investment to word count is higher for novellas, not lower. The story still has to work.

Frequently asked questions