Almost every self-publishing guide either says “editing is essential” or “you can manage without,” neither of which helps you spend your book budget well. So here’s what editing service to buy at each budget level, in order; what each one actually gets you; and what to do when you can’t afford the service you want.
Key facts
- Four stages, 80,000-word novel About $7,100 to $9,400 at 2026 medians
- Buy first, almost always The stage that revises the most
- Never skip on a book for sale A proofread of the laid-out pages
- Cheapest high-value purchase A proofread, at 1.2 to 2.0 cents a word
- Best free thing you can do Put the manuscript away for six weeks
- Worst common mistake One “everything” edit instead of staged passes
- Second worst Spending the whole budget on the wrong stage
- If the budget is zero Beta readers and time, in that order
Do you actually need an editor to self-publish?
Yes. Other guides say no, but if you intend to sell to strangers and compete in the marketplace, you need an editor in order to publish something that can stand next to a traditionally published book.
A reader who buys your book has no way of knowing it was your first, and they’ll judge it against the last book they read, which likely absorbed somewhere between 120 and 200 hours of professional attention. That’s the actual competition.
The most common failure I see is where money went to the wrong stage of editing: $2,000 of sentence polishing on a manuscript whose real problem was that the first 80 pages are backstory. The prose is now lovely, yet nobody is finishing the book.
What does the full sequence cost, and what do people actually spend?
Buying every stage once costs roughly $7,100 to $9,400 for an 80,000-word novel at 2026 medians. Nonfiction of the same length lands between $10,000 and $13,600. Most self-published authors spend a fraction of that, and the useful question is how to spend that fraction well on what the manuscript needs most.
Table 1. One pass of each stage, 80,000-word manuscript, 2026 Editorial Freelancers Association medians.
| Stage | Fiction | Nonfiction | What it changes |
|---|---|---|---|
| Manuscript evaluation | $1,200 to $1,440 | $1,200 to $1,440 | Nothing. It tells you what to change. |
| Structural edit | $2,400 to $2,800 | $3,200 to $4,000 | The shape of the book |
| Line edit | $2,160 to $2,800 | $2,800 to $4,000 | How the sentences read |
| Copy edit | $1,600 to $2,160 | $2,400 to $3,200 | Correctness and consistency |
| Proofread | $960 to $1,600 | $1,600 to $2,400 | Errors on the finished pages |
The median figures are from the Editorial Freelancers Association 2026 rate chart, drawn from a survey of more than 1,100 working editors about the rates they charged in 2025. (Note that these are the industry’s average editing rates. Senior editors with in-house publishing backgrounds and, better still, verifiable multiple bestseller and award-winner credits should charge above these bands.)
Two things to take from the table before we get to the tiers. Nonfiction costs more at every stage, so if you write nonfiction and you’ve been budgeting from figures about novels, add about a third. And the proofread, the cheapest line, is the one that protects everything above it, which is why it’s the last thing to cut rather than the first.
What does each budget realistically buy?
Here’s the direct answer, for an 80,000-word novel. Adjust up by about a third for nonfiction, and down proportionally for a shorter book, since almost everything is priced per word.
Table 2. What to buy at each budget, 80,000-word novel.
| Budget | Buy this | Do this yourself |
|---|---|---|
| Under $1,000 | Nothing yet. Save toward one stage. | Beta readers, six weeks’ rest, one structural self-edit |
| $1,000 to $1,700 | A proofread of the laid-out pages | Everything upstream, carefully, with software |
| $1,700 to $2,700 | A manuscript evaluation, then act on it | The revision, then trade a proofread with another writer |
| $2,700 to $4,000 | A copy edit and a proofread | Structure and style, with beta readers |
| $4,000 to $6,500 | A structural edit, a copy edit, and a proofread | The line-level work, after the structural revision |
| Above $7,000 | All four stages, in order | The author review round at each stage |
The tier that surprises people is $1,000 to $1,700, where my advice is to buy the last stage (the proofread) rather than the first (the manuscript evaluation). The reasoning is that a proofread is the only stage whose absence is visible to every reader on page 1. Structural weakness loses you readers slowly and ambiguously; nine typos in the opening chapter lose you the review. If you have exactly one purchase and your manuscript has been through beta readers and several honest revisions, the proofread is the one that shows.
Above that, the logic inverts. Once you can afford two stages, buy the earliest one you need plus the proofread, and do the middle yourself. Structure is the stage where an outside eye is least replaceable because it’s the one thing you genuinely can’t see in your own book.
If you can buy only one stage, which one?
Buy the stage that revises the most, and the answer depends on one question: do you know what’s wrong with your book?
If you don’t know: a manuscript evaluation. It costs about half the price of a structural edit and its whole purpose is telling you where to spend next. Buying it is how you avoid the expensive mistake, which is spending everything on a stage that wasn’t your problem.
If you know it’s structural: the structural edit, and accept you’ll be doing the rest yourself. An editor is least replaceable here. You can’t see the shape of your own book, and no amount of rereading fixes that because rereading is what caused it.
If the book is genuinely solid and you’re publishing: the proofread, on the laid-out pages. Cheapest, most visible, protects everything else.
What can you genuinely do yourself?
These are the actions that actually move a manuscript, in rough order of how much they return for the effort.
- Put it away for six weeks. Free, and the single most effective thing on this list. You can’t edit a manuscript you can still recite. Distance is the only thing that restores the ability to read your own work, and there’s no shortcut.
- Read it aloud, all of it. Slow, tedious, and it catches more than any software: flat rhythm, repetition, dialogue nobody would say, and a surprising share of real-word errors.
- Get beta readers, and ask them the right question. Not “Did you like it?” Ask where they stopped reading, where they got confused, and who they thought the book was about. People are unreliable at diagnosis and very reliable about their own attention.
- Run software before you hire anyone. It clears repetition, doubled words, inconsistent hyphenation, and most homophone slips. It costs nothing and it lowers your bill by removing noise.
- Build your own style sheet. A single page recording your choices: character spellings, place names, numbers, whether you use the serial comma. Doing this yourself catches inconsistencies and saves a copy editor real time.
- Do a structural outline on paper. Print it one line per scene or section on index cards or a list, and look at the sequence with the prose stripped out. Problems that are invisible in the text are obvious in the outline.
What you can’t do yourself, at any level of skill: read your own book as a stranger. Every writer I know has been surprised by an outside reader on their own manuscript. That’s not a failure of skill, it’s a property of having written the thing.
If you want the mechanics of the last stage, I’ve written a separate guide to reading your own work for errors, and one on working through tracked changes for when you do hire someone.
What should you never skip?
Two things, and both are inexpensive relative to what they protect.
A proofread of the laid-out pages. Not the manuscript, the finished pages, on the actual file you’re going to upload. Layout introduces errors that didn’t exist in your Word document: broken hyphenation, a running head naming the wrong chapter, a section break that lost its space so two scenes read as one. Nobody upstream could have caught them because they didn’t exist yet.
A cold read by someone who has never seen it. Free or nearly free, and it catches what familiarity has made invisible to you and to everyone who has been close to the project. Traditional publishers build this in deliberately, late, for exactly this reason. One competent reader, on the final pages, told to note anything that stops them.
Publishing in both formats means checking both. A clean print file says nothing about the ebook, which fails in its own ways: emphasis dropped during conversion, contents links pointing nowhere, scene divisions that simply disappear. Open it on a real phone, and open it again at a large accessibility text size, which is how a good share of your readers will read it.
What should you not spend money on?
Three things, and the money saved on these is what buys a stage that matters.
- A “free assessment” from a company that sells packages. It concludes that you need the package. It always concludes that you need the package. A read you pay for is one that’s allowed to end with nothing to sell you.
- A second opinion on a stage you’ve already bought. Two structural edits from different people produce contradictory advice, and you’ll paralyze yourself trying to choose. Buy one, act on it fully, and judge afterwards.
- Editing an unfinished manuscript. If you know the ending is wrong, finish it first, however roughly. An edit of half a book mostly tells you it’s half a book.
A fourth item, which isn’t really about money. Don’t commission an edit when the thing you actually want is encouragement. Arriving at a professional read hoping for praise guarantees a bad few weeks and an argument over every mark on the page. The wish is a normal one; a writing group and a few generous readers answer it better, and for nothing.
Does an edited book actually sell better?
Editing doesn’t sell books. It prevents a specific kind of failure, which is not the same claim and is the honest version.
What sells books is discoverability, cover, category, price, and whether the first pages hold. Editing touches the last of those directly and none of the others. An excellently edited book with a bad cover in the wrong category sells nothing.
What editing does is stop a reader from deciding, in the first few pages, that you didn’t take this seriously. Errors early are read as a signal about everything after them, which is unfair but completely reliable. Reviews mentioning typos suppress sales in a way that’s hard to recover from because it’s the one criticism no reader can be argued out of.
So the accurate framing is defensive, and I’d rather give you that than a promise I can’t keep. Editing is not a growth investment with a return you can calculate (even though, ultimately, a good editor’s feedback will make you a better writer). It’s the removal of a reason to stop reading, and it’s most valuable exactly where you’re least able to judge it yourself.
How do you know when it’s ready to publish?
When the changes you’re making have stopped improving it and started just being changes. Every writer recognizes that point and almost everyone pushes past it, in both directions.
Practical signals that it’s ready:
- Two consecutive readers report the same experience of the book, and it’s roughly the one you intended.
- Your last revision pass changed mostly words rather than paragraphs.
- You’ve stopped finding structural problems and started revising sentences you’ve already changed twice.
- The proofread came back and the corrections were genuinely small.
Signals it isn’t:
- You’re hoping nobody notices something. They will, and it will be in the reviews.
- You can’t summarize what the book is about in two sentences.
- Readers keep stalling in the same place and you’ve decided they’re wrong.
- You haven’t looked at the laid-out pages, only the manuscript.
The last one catches more people than any of the others. A book you’ve only ever seen as a manuscript is a book you haven’t actually read yet.
Frequently asked questions
What’s the minimum I should spend to publish something decent?
For an 80,000-word novel going on sale, most guides advise roughly $1,000 to $1,700 for a proofread of the laid-out pages as long as you do the free work first: a long rest, beta readers, a read-aloud pass, and a software check. Competition is stiff, though, and definitions of “decent” vary. I suggest you don’t settle for an “it’s better than nothing” approach and budget higher: $2,700 to $4,000 for a copy edit and a proofread. Why add to the Amazon wasteland?
Can I trade editing with another writer?
For beta reading and cold reads, absolutely; and it’s one of the best free resources available. For copy editing and proofreading, only if the other person actually has the training. Enthusiasm and being good at grammar aren’t the same as knowing how to proofread laid-out pages.
Is it cheaper to edit a shorter book?
Yes, almost exactly proportionally because nearly everything is priced per word. A 50,000-word novel costs about 60 percent less than an 80,000-word one at the same rates. Cutting your manuscript is a legitimate way to reduce the bill, and it usually improves the book.
What if I’ve already published and it has errors?
Correct it and replace the file. Re-uploading is something print-on-demand books and ebooks allow and a printed trade edition does not, so this is one of the few places where publishing yourself is straightforwardly easier. Have the laid-out pages read properly first, or you’ll be doing it again.
Should I hire the same editor for every stage?
It’s efficient and it’s common on independent projects, as long as the stages happen in the right order, with some time in between. Separate passes are the ideal, and combined passes (typically, a substantive and line edit) are frequently the right call if you’re working with a veteran editor who regularly performed this type of combined pass in-house. The argument for a different or an additional proofreader at the end is strong, though: the value of that stage is fresh eyes, and the person who has already edited the book no longer has them (unless, say, a month or so has passed).
Working out where a limited budget should go? That’s a real question with a real answer, and it depends on your manuscript rather than on a general rule. Show me what you’ve written, how long it is, what readers have said, and roughly what you can spend, and I’ll tell you which stage to buy and what to do yourself. If the honest answer is that you shouldn’t spend anything yet, you’ll get that answer too.