How do I choose self-publishing platforms that honor craft?

Somewhere between finishing a manuscript and seeing it on a shelf, most authors realize the real work has only started. Choosing among self publishing platforms determines whether that manuscript reaches libraries, bookstores, and readers who never open Amazon at all, and that choice shapes the book’s life for years, not just its launch week. Nobody publishes an audited figure for how much of self-publishing runs through Amazon KDP. The numbers that circulate, usually “about 70%,” trace back to aggregators that name no source and no method. What is documented is simpler: authors who report the highest earnings tend to publish in more than one place. The Alliance of Independent Authors’ 2025 member survey puts median indie author income at $13,500 (self-reported, sample size undisclosed), against $6,000 to $8,000 for traditionally published authors (ALLi, 2025).

Self publishing platforms function as a layered system: Amazon KDP’s terms shape ebook royalties, IngramSpark’s print network reaches libraries, and Draft2Digital’s metadata fields decide whether a distributor’s database understands the book at all. This article covers how to evaluate platforms as craft infrastructure, how to build a distribution stack, how classification and metadata function as design work, and where the market is heading.

Layer these pieces correctly, the way ALLi’s distribution guidance recommends, and the book gets found in more places than any single channel could manage alone. Skip that layering, and you’re relying on Amazon’s algorithm alone to do work that used to belong to an entire distribution network. The sections that follow walk through what “craft-honoring” means in a platform, how to build a stack instead of a single choice, how classification and metadata function as design work, and where the market is heading next.

Key Takeaways

• Market reality: Amazon is the largest single channel and a baseline rather than a complete strategy. Its exact share is not independently documented, so don’t quote one.
• Print reach: IngramSpark is the main route into bookstores, libraries, and academic channels, a claim supported by Ingram’s own documented distribution network rather than by a share statistic.
• Earnings: ALLi’s 2025 member survey reports a $13,500 median for indie authors against $6,000 to $8,000 for traditionally published authors (self-reported, self-selected membership, sample size undisclosed).
• Classification is craft: BISAC and Thema codes serve industry systems, while Amazon categories and keywords serve reader search language.
• Backlist compounds: Full-time self-published authors publishing since 2018 saw mean incomes rise 76% by 2022 (ALLi, 2024).

What Makes Self-Publishing Platforms Honor Craft

Self publishing platforms honor craft when they let a book be classified accurately, produced well, and found by the readers it was written for. IngramSpark’s own onboarding documentation frames this same test as its core value proposition: protecting a title’s specifications rather than flattening them into whatever format is easiest to process at scale.

Maybe you’ve uploaded a book, picked the categories that seemed closest, and moved on, only to find it buried three pages deep in an irrelevant search result. That’s a common pattern, and it’s fixable once you understand that categories are read by machines and humans differently. Researcher findings published in Taylor & Francis journals show that platform design itself governs discoverability: categories, algorithms, and recommendation systems decide which readers ever see a listing, meaning craft has to extend past the manuscript into platform literacy itself (Taylor & Francis, 2023). A carefully written book with careless categories competes at a disadvantage against a competent book with precise ones.

Format matters here too. Ebook, print, and audiobook revenue split roughly across the self-publishing market, and no platform handles all three with equal strength. A service built for Kindle files may offer weak print specifications; a print-on-demand printer may have no audio distribution at all. When evaluating any platform, check its print trim and paper options, its royalty and pricing controls, how it handles territorial rights, how much detail its metadata fields actually accept, and what it costs you, in time or in stranded files, if you decide to leave.

Questions to Ask Before You Upload

Three checks separate a platform that serves the book from one that simply hosts it.

• Print fidelity: Can you specify trim size, paper stock, and cover finish that match your design intent?
• Metadata depth: Does the platform accept full BISAC and Thema codes, series data, and contributor roles?
• Reversibility: How easily can you withdraw the title and move distribution elsewhere?

Building a Distribution Stack Instead of Picking One Service

Self publishing platforms rarely work well alone, and the practical stack that shows up consistently in how experienced authors describe their own setups looks like this: KDP for Kindle and Amazon print, IngramSpark for bookstore and library print, and either direct retailer accounts or an aggregator such as Draft2Digital for Apple Books, Kobo, and Google Play. Each channel reaches a different reader, one who shops Amazon, one who browses a library shelf, one who reads exclusively on an iPad.

One common pattern looks like this: an author publishes exclusively on KDP, sees decent sales the first month, then watches them flatten by month three. When that author eventually adds wide distribution and a library-facing print edition, sales climb again, not because the book changed, but because more readers could finally find it. There’s no single correct stack for every book, and it’s fine to start narrow and expand once you notice which channels your readers actually use.

Per-retailer share of self-published ebook sales isn’t independently published. The figures that circulate come from aggregators with no stated method, so treat them as noise rather than evidence. What’s worth doing instead is tracking your own split from your own sales reports once you’re distributed wide. Authors who publish Amazon-exclusively leave an unmeasured, but real, portion of the wide ebook market unaddressed simply by never showing up there.

Publishing analyst Jane Friedman has documented Amazon’s continued dominance in the space and argues that authors need to understand “KDP’s ecosystem deeply, including its category systems, keyword fields, and merchandising quirks,” while still building distribution beyond it (Jane Friedman). Academic and ALLi sources push harder toward diversification away from single-platform dependency, which is a difference of emphasis rather than disagreement. Friedman is telling you to master the biggest room in the house; the diversification-minded sources are telling you not to live in only one room. Both are right. Treat any platform share figure you encounter elsewhere as directional at best, since it typically comes from industry modeling rather than transparent primary data. For more on where single-platform strategies go wrong, see how self-publishing platforms fail authors.

Treating Classification and Metadata as Design Work

Self-published titles with ISBNs rose 7.2% in 2023 alone, according to Publishers Weekly, surpassing 2.6 million titles and more than doubling over the prior decade (Publishers Weekly, 2024). At that volume, precision in classification stops being optional. Metadata is the only part of a book that distributors, librarians, and algorithms read before a single reader does.

BISAC codes (the standard categories North American bookstores and libraries use to shelve a book) and Thema codes (the international equivalent) govern how distributors, libraries, and bookstore databases understand a title. Amazon’s categories and keyword fields govern how consumers browsing the store actually find it. These serve two distinct audiences: industry codes need to describe the book accurately for professional systems that a reader never sees; Amazon fields need to match the actual words a reader types into a search box. Confusing the two produces a book that’s technically correct and practically invisible, or one that ranks well on Amazon but confuses every library acquisitions system that touches it.

Comparable titles used deliberately in descriptive copy situate a book within reader expectations, a practice borrowed directly from traditional publishing and one that works better than stacking keywords into a description. The most common mistake identified across sources is “upload and hope”: treating placement on a platform as equivalent to discoverability, rather than as the starting point for it (Taylor & Francis, 2023). Little peer-reviewed research quantifies exactly how one BISAC code or one keyword changes sales outcomes. Most of what works here is practitioner knowledge, tested by authors over repeated launches, not laboratory findings. For a closer look at how services can undercut this work rather than support it, see where self-publishing services miss craft.

A Backlist Maintenance Routine

Metadata decays as reader search language and category structures shift over time.

• Annual review: Re-check categories and keywords on every live title each year.
• Format expansion: Add audio or large-print editions to titles with tested sales.
• Description refresh: Update comp titles and opening lines as genre expectations change.

How the Self-Publishing Platforms Market Is Changing

The self-publishing services market, covering platforms, editing, design, and marketing together, continues to grow, and integrated upload-and-distribute services are becoming a more common way authors enter the space. That consolidation trend matters for platform choice: more tools now bundle distribution, formatting, and metadata entry into a single dashboard, which lowers the barrier to publishing wide but can also obscure the distinctions this article has been drawing between industry classification and consumer-facing keywords.

Expect AI-assisted category and keyword suggestions to show up inside more platform dashboards. These tools can be useful for surfacing options an author wouldn’t have considered, but they carry a real risk: steering classification toward whatever the algorithm currently favors rather than what accurately represents the book. Audiobooks continue gaining share of overall self-publishing revenue, which means platforms that can maintain one coherent metadata core across ebook, print, and audio editions, rather than treating each format as a separate upload with separate data, will likely become a real differentiator for authors managing a growing catalogue.

Earnings data helps calibrate expectations here. In the top-earning decile, traditionally published authors reported median incomes of $305,000 against $154,000 for self-published authors. Self-publishing infrastructure raises the income floor more reliably than it raises the ceiling, which is a different promise than the one often implied by platform marketing. Commentary from the University of Glasgow places 2018 to 2022 as the period when professionalized indie authors, using multi-platform distribution and metadata discipline, overtook traditionally published authors at the median (University of Glasgow, 2023). For more on protecting creative intent as platforms change, see how self-publishing services protect vision.

Why Self-Publishing Platform Choice Matters for Your Book

With roughly 2.6 million self-published titles now competing for attention (Publishers Weekly, 2024), the infrastructure surrounding a book determines whether careful writing ever reaches a reader. Platform decisions made once at launch keep shaping sales, library access, and international reach for the entire life of the title. An accurate, well-distributed catalogue compounds in value across years, while a single-channel launch tends to peak early and fade.

Conclusion

Choosing self publishing platforms that honor craft means building a layered stack rather than betting on one service. KDP supplies scale, IngramSpark supplies print legitimacy and bookstore access, and wide ebook distribution reaches the readers who buy outside Amazon entirely. Classification and metadata carry as much weight for a book’s life as line editing does for its prose; they are not an afterthought to the writing, but a continuation of it.

Consider auditing one existing title this month. Verify its BISAC codes, review its Amazon categories against current reader search language, and confirm whether its print edition is actually available to libraries and bookstores. Notice what patterns turn up, without judgment about what you find. Then decide which missing channel to add next.

Frequently Asked Questions

What are self publishing platforms?

Self publishing platforms are the services authors use to produce, distribute, and sell their own books directly to retailers and libraries without a traditional publisher. Examples include Amazon KDP for ebooks and print, IngramSpark for bookstore and library distribution, and aggregators like Draft2Digital for wide ebook reach.

How do I choose the best self publishing platform for my book?

Choose based on format needs rather than a single “best” option. Check whether a platform lets you specify print trim size and paper stock, accepts full BISAC or Thema metadata, and allows easy withdrawal if you switch distributors. Most authors combine KDP, IngramSpark, and an aggregator rather than relying on one service alone.

Is Amazon KDP enough on its own, or do I need other platforms too?

KDP alone leaves out library shelves, bookstores, and non-Amazon ebook readers. Jane Friedman notes authors should master KDP’s category and keyword systems, but ALLi’s distribution guidance and academic research both recommend adding IngramSpark and wide ebook distribution to avoid single-platform dependency.

What is the difference between BISAC codes and Amazon keywords?

BISAC codes (and their international equivalent, Thema) are the standard categories bookstores, libraries, and distributor databases use to shelve and classify a title. Amazon categories and keywords are consumer-facing search terms. Confusing the two can make a book technically correct but invisible to readers, or well-ranked but confusing to library systems.

Does IngramSpark reach libraries and bookstores that Amazon KDP doesn’t?

Yes. IngramSpark’s documented print network is the main route into bookstores, libraries, and academic channels, places KDP’s Amazon-centric print distribution doesn’t reach as effectively. IngramSpark’s onboarding materials frame protecting a title’s exact specifications as its core value.

Do self-published authors actually earn more than traditionally published authors?

At the median, yes: ALLi’s 2025 member survey reports $13,500 for self-published authors versus $6,000 to $8,000 for traditionally published authors. But at the top-earning decile, traditionally published authors report $305,000 versus $154,000 for self-published authors, meaning self-publishing raises the income floor more than it raises the ceiling.

Sources

• Alliance of Independent Authors (ALLi) – Self-reported member income survey, 2025. Self-selected membership, sample size undisclosed.
• Taylor & Francis – Peer-reviewed publishing research.
• Jane Friedman – Industry analysis.
• Publishers Weekly – Trade reporting.
• University of Glasgow – Academic research on publishing.