Authors tend to view the question of traditional vs. self-publishing as aspirational, when it's better seen as practical. Your costs of production and profits per copy sold aren't just footnotes in a publishing contract... They're the foundation of your ability to turn writing into a living.
To make a sound decision, you'll want a nuanced enough understanding of the two paths to know exactly who does what, what it costs now, and what it costs later.
In this guide, we’ll walk you through exactly what you need to know, including:
- Who does what during each step of the process (and how much it costs to do it yourself)
- How the money works: advances, royalties, and what it takes to make a living as an author
- The main factor outside of publishing model that impacts your lifetime royalties
By the end, you'll be able to make an informed decision about whether you prefer to partner with a publisher or do it on your own.
The very, very short version
The best reason to self-publish is when you view your book as a cornerstone of your career and/or income, and you’re willing to accept the upfront risk and effort to make that happen.11Self-publishing tends to result in 4x higher earnings per copy sold, at the cost of taking responsibility for your book’s editing, layout, printing, and distribution. There's nothing overly complex about any of it, but it does involve a lot of moving parts, which takes an extra 1-3 months and is admittedly kind of annoying. If you can learn new software, hire freelancers, and oversee a moderately complex project, then you can self-publish at a professional level. A good checklist also helps.
The best reason(s) to traditionally publish are: you want to minimise the fiddly busywork of producing a book; you care more about reputation than royalties; and/or you really need placement in major brick-and-mortar bookstores.22Bookstore placement only matters for the "impulse buy" genres like pop psych, pop sci, and manifesto nonfiction. For useful nonfiction and genre fiction, bookstore sales are a rounding error compared to Amazon.
In either case, the vast majority of your effort will go into writing a good book.
Also in either case, you’ll probably have to do your own marketing.
The biggest misconception about self-pub vs. tradpub is about who handles the marketing. Authors tend to believe that a publisher will do it for them, which simply isn’t the case.33If you need proof, just look at a standard book proposal, which hinges on how you plan to promote and market your own book.
A publisher will help produce and distribute your book (including bookstore placement), but they won’t help market it—at least, not until after it’s already succeeding. Publishers are great at amplifying success, not creating it.
To become successful in the first place, a modern author is on the hook for selling the first 10,000 copies of their own book, regardless of how they have chosen to publish.
For the details, let's jump into a phase-by-phase breakdown of exactly who does what, from proposal through to publication:
Phase 1: Writing and revising:
The first phase is about making a great manuscript. A publisher will save you some money by providing your editor44If you're in a financial position that makes it impossible to hire a freelance copy editor, then the traditional option becomes significantly more appealing., but the experience of writing (and rewriting) a great book is otherwise largely the same.
Self-Publishing | Traditional publishing | Notes | |
---|---|---|---|
Pitching & proposals | N/A | Author | Some tradpub authors also pitch an agent before going to publishers, which adds an extra step (whereas selfpub authors just start writing) |
Drafting & early revisions | Author | Author | Don’t forget to do your Beta Reading! |
Editing & refinement | Author DIY ($0, but slow) and/or freelance editor ($2,000+) | Publisher-supplied editor | A good developmental editor costs $5,000+ and a copy editor costs $2,000+, so getting this for "free" is certainly a perk of tradpub. For selfpub authors, the lack of a developmental editor can be compensated for with extra Beta Reading, but I'd always recommend hiring a proper copy editor and proofreader, if at all possible. |
Phase 2: Preparing for publication:
This is the moment where a tradpub author relaxes55When I was earlier in my career, my time was worth so little that I was happy to do all of this myself. These days, I'm more inclined to outsource large chunks of it to good freelancers (or to a good hybrid publisher, which is essentially a team of freelancers with a bundled price).I do sometimes fantasize about sending my manuscript to a publisher and then slipping off somewhere to sip margaritas for three months, but the difference in royalties is too much for me to actually do so., and a selfpub author becomes a project manager for the next 1-3 months.
Self-Publishing | Traditional publishing | Notes | |
---|---|---|---|
Proofing | Author (with AI; $0) or freelance proofreader ($250-800) | Publisher | Proofreaders offer incredible value for money. If you only have a few hundred dollars to spend on your book, spend it on either a proofreader or a cover designer. |
Interior layout | Author (with a DIY tool) and/or freelance designer ($150-400) | Publisher | For DIY tools, I like Vellum ($240) and Atticus ($150). I prefer DIY here if possible, since otherwise you have to re-hire freelancers whenever you want to update. |
Cover design | Author (with a template; $0) and/or freelance designer ($200-500) | Publisher | Always begin the cover design process by finding at least six examples of book covers you love. |
ISBN and imprint | Author ($10 via Bowker) | Publisher | This step mystifies many authors due to bad information, but it’s really, really easy. Here’s a walkthrough. |
Distribution to Amazon etc. | Author (via KDP; $0) | Publisher | One major advantage of selfpub is global distribution within a week; tradpub takes months here. |
Distribution to physical bookstores | Not possible (or at least extremely situational) | Publisher | Physical bookstores are aspirational, but aren't especially meaningful in terms of sales (except in certain genres that target impulse buyers). |
Phase 3: Marketing, audiobook, events:
The big task here is Seed Marketing. I know every author hates to hear it, but that's on you either way.
Self-Publishing | Traditional publishing | Notes | |
---|---|---|---|
Seed marketing and early promotion | Author | Author | See our guide to selling the first 10,000 copies |
Audiobook recording | Author | Author | Typically takes 2-4 hours of recording time per 10,000 words; be sure to try mastering a small audio sample before recording the whole thing. |
Audiobook mastering | Author (with a tool like Descript; $30) or freelancer ($200-500) | Publisher | For a full walkthrough, see our video course on recording and mastering your own audiobook ($30) |
Audiobook distribution | Author (via ACX or Findaway Voices; $0) | Publisher | ACX is owned by Amazon and only available to authors in the US, UK, CA, and IE. Findaway Voices is owned by Spotify and available to everyone, but takes a higher percentage. Either will get your book available on all major platforms. |
Bulk printing for events and giveaways | Doable via IngramSpark or LuluDirect | Not possible (you have to buy your own books) | If you do lots of conferences, consulting, or workshops, this can be a big reason to self-publish. If you don’t, then it’s usually not relevant. Here's how I do bulk printing. |
Royalties, advances, and earning a living as an author
Royalties are annoyingly complex: each format is handled differently66For example, a publisher's paperback royalties are a percentage of retail price, which is easy to understand. Whereas ebook and audiobook royalties are a percentage of what the publisher receives from the sales platform, so that 25% ebook royalty is actually 25% of 70%, or 17.5%., contracts can be individually negotiated, the platforms do wacky stuff with deals and discounts, and new publishers keep popping up with different models.
So for the sake of sanity, I’ve attempted to “normalize” everything into the unified language of cover-price royalties. And although this simplification will create plenty of opportunities for nitpickers to nitpick, it should allow the rest of us to make a like-for-like comparison without going loopy.
Here's what that looks like:
"Normalized" royalties
Self-publishing "normalized" royalty rate | Traditional publishing "normalized" royalty rate | Self-pub earning multiple | Note | |
---|---|---|---|---|
Paperback royalty | 35-55% | 7.5-12.5% | 3-5x | Selfpub payouts change based on the book's cost of printing, which is why the range here is so wide |
Ebook royalty | 70% | 17.5% (25% of 70%) | 4x | Publishers pay out a quarter of what they receive from the sales platform, so 25% of 70% = 17.5% |
Audiobook royalty | * See note | * See note, but generally a quarter of what self-pub gets | 4x | Exact audiobook earnings are quite opaque since they’re sold via credits and subscriptions, but the ratio holds true |
Overall, with reasonable assumptions about pricing, printing, and deals, a self-published author tends to take home about 4x more per copy sold.77For a more detailed justification and explanation of some of these figures, see the sections on self-publishing paperback royalties.
Dollar earnings per copy sold
Royalty rates are all well and good, but what matters most to me, as an author who uses books to pay my mortgage and support my family, is dollar earnings per copy sold.
This tells me how many books I need to sell per week, which tells me whether I need to be hustling the hands-on marketing, or whether I can relax and rely on back catalog sales.
After accounting for printing costs and cover price88These numbers assume a $14 paperback with 140 pages, and a $9.99 ebook. For a detailed explainer and justification of these figures, see How it works: Self-publishing paperback royalties, your per-copy earnings might look like this:
Self-publishing | Tradpub | Note | |
---|---|---|---|
Paperback | $7 / copy ($14 price point, 50% royalty) | $1.4 / copy ($14 price point, 10% royalty) | To learn how to find the exact figures for your own book, see How it works: Self-publishing paperback royalties. |
Ebook | $7 / copy ($9.99 price point, 70% royalty) | $1.75 / copy ($9.99 price point, 17.5% royalty) | Due to Amazon’s ebook royalty structure, pretty much every ebook should be priced at $9.99. (Except serial fiction, where the first book in the series should be as cheap as possible to bring in new readers.) |
One takeaway here is that per-copy earnings for paperback and ebook99Audiobook is a whole other story and excluded for sanity's sake. are actually quite similar. This won't always be the case, but it's usually close enough to simplify your mental math.
Now that we know our average earnings per copy, we can estimate some income tiers:1010This table uses the same assumptions about pricing and printing as the previous one, and the numbers may vary for your own book.
Self-publishing | Tradpub | Note | |
---|---|---|---|
1 copy sold | $7 author earnings | $1.6 author earnings | I'm assuming a 50/50 balance of sales between ebook/paperback, and by ignoring audiobooks |
5,000 copies sold | $35,000 | $8,000 | As a self-published indie, 5,000 copies a year (or 416 copies per month) will earn you about $3k per month in extra income. |
10,000 copies sold | $70,000 | $16,000 | To earn a living from books, set your sights on self-publishing and selling 10k copies per year for $70k in income. This is non-trivial, but doable. |
50,000 copies sold | $350,000 | $80,000 | To earn a living as a tradpub author, you need to be moving 50k copies per year. This requires that your book be a big hit. |
100,000 copies sold | $700,000 | $160,000 | Six-figure book sales are financially meaningful regardless of which route you follow. |
Decide how much you want to earn from books, check the number of copies required to get there, and then follow whichever route feels more viable.
Back catalog or bust
Beyond the publishing path, there's one other critical factor that can significantly improve your expected outcome: whether your book continues selling after its launch window has ended.
“Book publishers make more than 90% of their profit from books they published more than six months ago. And yet they put 2% of their effort into promoting and selling those books.”
— Seth Godin, Understanding the Backlist
This is called entering the back catalog (or backlist), and if you can achieve it, it transforms the world of books from a scramble for survival into a paradise of plenty.1111The whole concept behind Write Useful Books is to write nonfiction in a way that can more reliably achieve back catalog success.

Screenshot of Amazon KDP royalties dashboard for Rob's books, showing 1.3 million in lifetime earnings and growing sales across more than 10 years
Imagine that you’ve spent a hard year writing your book and only sell 5,000 copies before the launch hype fades and it stops selling. In that case, your year of effort has returned only $35k if self-published (or $8k if traditionally published).
If, however, your book enters the back catalog and continues selling an additional 5,000 copies per year, then that same year of effort has created a potentially life-changing extra income.
Plus, aiming at the back catalog allows you to reframe your overall goal into a far more bite-sized weekly target:
Copies sold per week | Selfpub yearly income | Note | |
---|---|---|---|
5,000 copies/year | 100 copies per week | $35,000 | When I reached this milestone (~2016), I used the 3k/month to buy a small sailboat and take a 3-year sabbatical |
10,000 copies/year | 200 per week | $70,000 | When I hit this milestone (~2019), I used the royalties to buy an apartment in Barcelona and do passion projects |
50,000 copies/year | 1,000 per week | $350,000 | I'm still slightly shy of this milestone (I do about 800 copies/wk right now), but that still leaves plenty of surplus |
100,000 copies/year | 2,000 per week | $700,000 | Hoping to reach this someday ;) |
The big question: Selfpub or tradpub?
As you may have judged by now: it depends.
My personal preference is probably clear by now (i.e., self-publish if you can), but it certainly depends on the genre, since traditional publishers are better-suited to helping books in some genres than in others.
Here's how I would personally think about publishing each genre:
Genre | Self-Publish? | Traditionally Publish? | Notes |
---|---|---|---|
Useful nonfiction (how-to's, skill-builders, problem-solvers, goal-achievers) | 70% | 30% | Since readers buy these books via high-intention search (not as impulse buys), and rely on recommendations and reader reviews (not mainstream approval), a publisher has fewer ways to help. IMHO: self-pub for more income or tradpub for less hassle. For deeper advice, read Write Useful Books. |
Serial genre fiction | 1,000% | — | This is a place where self publishing is actually a real advantage, since an indie author can be much more aggressive with a range of fan-gathering tactics such as steep discounts, widespread giveaways, cross-promotional bundles, and more. (Plus, if you need fans to buy the next book in the series, then you really want to own the relationship with them.) |
Pop science, pop psych, and pop business | — | 100% | These books build momentum with a combination of traditional PR and prominent retail placement to drive impulse buys, both of which are firmly within the wheelhouse of tradpub. If you're self-publishing, you'll need to shell out $50k+ on a top-tier hybrid publisher and PR firm (and it'll still be a gamble since this genre is particularly all-or-nothing). |
Nonfiction manifestos (i.e., more inspiration than information) | 20% | 80% | Nobody actively searches for a book like this, so retail placement is essential, especially in places like train stations and airports filled with depressed commuters. I'd only self-publish if I either had a huge fanbase, or if using the book as a "business card" without caring about sales. |
Literary fiction | — | 100% | Success requires the approval of mainstream gatekeepers from the awards circuit and newspapers who only pay attention to traditionally published titles. But if you're writing this kind of book, you're doing it for the art, not the royalties, anyway. |
Literary nonfiction | 50% | 50% | Not the easiest genre for the last couple centuries. I'd probably self-publish if I had a platform, and traditionally publish if not. Seems to help your chances if you have a Nobel Prize. |
Academic nonfiction | 50% | 50% | This genre has some weird incentives, since the goal of publication isn't actually to sell any copies. In general, a Trade Publisher will make life easy and serve the purpose. If you want to sell some books, you might consider reframing your research as a Useful Book. (Alternatively, if it's a literal textbook, then 100% go with the big publishers for access to their academic sales teams.) |