Everything You Owe Before You Publish a Book

Most checklists list tasks; this one counts deliverables, codes, categories, keyword fields, and copy at fixed lengths, each of which has to agree with the book and with every other artefact. Change one classification decision and a surprising share of the rest stops fitting.

What follows counts the inventory plainly, shows how much of it depends on two or three classification decisions, works through two published examples, Agatha Christie’s The Murder at the Vicarage and J.C. Ryle’s Expository Thoughts on Matthew, and stays honest about which parts of the system retailers actually document. Where a mechanism is undocumented, this article says so and offers a way to test it rather than repeating folklore.

A book publishing checklist works because classification decisions made early lock in the vocabulary available to everything written later. Choose a primary BISAC code, and the categories, keyword fields, and description copy all have to work within that subject’s language. Skip a step, or write each platform’s copy from memory instead of a shared source, and the same book starts reading as several slightly different books. The sections that follow count the full inventory, trace why classification sits upstream of the rest, work two published examples in fiction and non-fiction, and lay out a sequence for making forty artefacts agree.

Key Takeaways

• Volume is the problem: a single title owes roughly forty separate written artefacts before publication, and most authors meet them one form field at a time.
• Classification comes first: BISAC and Thema codes, three of each recommended, sit upstream of categories, keywords, and description copy.
• Amazon documents its fields, not its mechanics: KDP publishes what to enter but says very little about how keyword boxes and categories are weighted.
• Consistency outranks cleverness: the same book described three different ways across three platforms reads as three weaker books.
• Baseline before enhancement: a documented minimum metadata set clears the launch floor before richer fields get added.

What a Finished Manuscript Still Owes: The Full Book Publishing Checklist

A working book publishing checklist for a single title, distributed to both Amazon and IngramSpark, runs to roughly forty artefacts. Grouped by function, the list stops feeling arbitrary and starts feeling like a sequence.

Classification accounts for about eight items: one primary BISAC code plus two supporting codes (BISAC being the code system U.S. bookstores use to shelve print titles by subject), up to three Thema codes for international markets (the parallel system many overseas retailers use instead), one or more retailer browse categories, and an age or grade range where applicable. According to Ingram Content Group, this group is chosen first because everything downstream depends on it.

Search fields add about nine items: seven Amazon keyword fields, one IngramSpark keyword string, and the short list of search phrases an author is actually targeting. Maybe you’ve stared at those seven blank keyword boxes on a KDP form with no idea what goes where. That blank-page moment is common, and it usually means the classification work above wasn’t finished first.

The copy group is the largest, around fifteen items: title, subtitle, series name, long description, short description, back-cover blurb, one-line logline, one-paragraph summary, extended summary, author bio at two lengths, and promo copy cut to fixed character counts for retailer, newsletter, and social fields. Production and identity fields add roughly eight more: ISBN per format, contributor records, publication date, territory rights, format, trim size, page count, and price by market, per Amazon KDP’s help documentation. A positioning group closes the list with two or three comparable titles carrying ISBNs, plus one or two written reader personas.

One manuscript is a single deliverable; the listing is about forty.

The Artefacts Nobody Warns You About

Several items appear only when a specific form demands them, usually late in the process.

• Character-capped promo copy: the same pitch rewritten to fit fields of different maximum lengths
• Comp titles with ISBNs: two or three genuinely comparable, reasonably recent books
• Reader personas: undocumented by any retailer, but the thing every keyword and blurb decision silently assumes

Why Classification Decides the Rest of the Checklist

Subject codes get chosen first, and that choice fixes the vocabulary available to everything written after it. Once a primary BISAC code is set, categories have to nest inside it, keyword phrases have to reinforce it rather than contradict it, and description copy has to use language a reader searching that subject would recognize. Change the primary code partway through the process, and categories, keywords, and blurb copy all stop matching the record they were built for. That’s the mechanism: one classification decision constrains roughly thirty other artefacts written afterward.

BISAC serves U.S. retail and library systems. Thema serves international markets that BISAC does not travel into cleanly, which is why IngramSpark’s checklist asks for a parallel set of codes rather than treating BISAC as sufficient everywhere. Current guidance from Self-Publishing Pro favors a small number of well-chosen categories, often one broad and one narrower, supported by keyword phrases and description copy, over increasing category count.

Here’s the honest hedge. Amazon KDP’s help documentation publishes which fields exist and what belongs in them, but it documents almost nothing about how the seven keyword boxes are weighted, whether phrase order matters, or how category placement interacts with search ranking. Widely repeated advice on these points is inference, not published mechanism. There is no public controlled study isolating how much a BISAC change, a keyword rewrite, or a category swap moves sales, and pretending otherwise does authors no favors.

Six subject codes, chosen in an afternoon, quietly set the vocabulary for thirty pieces of copy written over the following month.

How to Test Amazon Claims Yourself

Treat retailer behavior as testable rather than known.

• Check what displays: compare the categories shown in the live product page’s Best Sellers Rank against what you submitted
• Search your own strings: run each keyword field verbatim in a logged-out browser and record whether the book appears and at what position
• Change one variable: alter a single field, wait a full indexing cycle, and re-run the same searches before concluding anything

Two Worked Examples: Fiction and Non-Fiction

The checklist behaves differently depending on whether a book’s identity is generic or subject-driven. Two published sample reports, covering one novel and one theological work, show the complete artefact set assembled for each. As the Book Industry Study Group’s BISAC subject list confirms, crime fiction and religious commentary sit under entirely different codes, which is why the two reports diverge from the first field onward.

Take Agatha Christie’s The Murder at the Vicarage. The dependency chain starts with a primary BISAC in the crime and detective area, paired with Thema codes for international listing, then one broad browse category and one narrower rankable one. Seven keyword fields carry what the subject codes cannot express: period, setting, series position, tone. Comp titles and a reader persona built around the classic-mystery reader shape the blurb and the short promo copy that follows.

J.C. Ryle’s Expository Thoughts on Matthew asks classification to do heavier work. The subject codes have to distinguish commentary from devotional from general Christian living, so the subtitle carries a load a novel’s subtitle rarely does. Keyword fields absorb author name variants, tradition, and reading use case. The persona narrows so, and the description names a specific use rather than a general appeal.

The same forty-artefact checklist produces two very different documents depending on whether the subject codes or the keyword fields are doing the identifying. Both reports are published in full at perivane.com and can be read end to end rather than summarized here.

Making Forty Artefacts Agree: Working the Book Publishing Checklist

The hard part of a book publishing checklist is agreement, since every artefact makes a claim about what the book is, and contradictions are visible to readers even when they can’t name what’s wrong.

A sequence drawn from published guidance holds up in practice: settle title, subtitle, and description first; assign BISAC and Thema codes; select categories; write keyword fields and the distributor keyword string; verify ISBN, territory, format, and price; then submit the same record everywhere. The Book Designer recommends keeping one master file as the single source of truth, reused across every platform form and updated as market response accumulates.

Joshua Tallent of Firebrand Technologies, cited by NetGalley Insights, identifies a minimum set to clear before anything else: title, ISBN, author, format, publication status, BISAC subject, description, publication date, territory, page count, and physical specs. Table of contents entries, excerpts, reviews, additional keywords, title relationships, and merchandising themes come after that floor is met, not before. There’s no requirement to get every field perfect on the first submission. Metadata can be revised after launch, and most experienced authors do exactly that.

Common failures are worth naming directly: assigning BISAC and skipping Thema entirely, writing each platform’s form fresh from memory so descriptions diverge, choosing comp titles that are famous rather than comparable, and setting everything once at launch with no plan to revisit it. A pattern that shows up often looks like this: an author picks a punchy comp title because it’s well known, not because it’s genuinely similar, then wonders why ads built on that comparison underperform. Swapping in a less glamorous but more accurate comp usually fixes the mismatch within a few weeks. Authors weighing where this work fits into a broader budget may find a strategic view of self-publishing costs useful context, and those still assembling a launch plan around this checklist can cross-reference common self-publishing mistakes before submitting anything.

Forty artefacts written in one voice from one decision set will outperform forty written independently across four evenings.

Why the Inventory Matters

This work stays invisible until it’s missing: an incomplete BISAC field is what actually stops an IngramSpark upload, not an abstract sense that something is wrong. It arrives in fragments across several forms that never show the full picture at once. An author who counts the artefacts in advance writes them as one coherent document instead of improvising under deadline. The record that results follows the book through every catalogue that ingests it, long after launch week ends. Getting the classification right once saves a rewrite of thirty other fields later.

Conclusion

A book publishing checklist is an inventory of around forty written artefacts, six of which are classification codes that quietly govern the rest. The difficulty is coherence across all of them, and the only reliable method is deciding classification first, writing everything from one master file, and testing retailer behavior instead of assuming it works a certain way.

Retailers document their fields far better than their mechanisms, so treat confident claims about keyword weighting as hypotheses worth testing rather than settled fact. For a next step, the full sample reports for The Murder at the Vicarage and Expository Thoughts on Matthew at perivane.com show the complete artefact set assembled for one novel and one commentary. Building the same document for your own title, cost considerations included, is worth weighing beyond the sticker price alone.