When the manuscript is finally done — revisions complete, line edit incorporated, copyediting reviewed — there’s a reasonable temptation to believe the heavy work is behind you. The text exists. It just needs to be put on the page.
The page count, it turns out, is a number you produce, not one you’re given.
And that number locks everything downstream.
Here’s the dependency chain: text finalized → typeset → page count locked → spine width calculated → dust jacket file finalized → press files released → printer order placed → print run delivered. Each step requires the previous one to be complete. The typesetter doesn’t just produce pages; they produce the measurement every other production step depends on. The spine width on your dust jacket — the dimension your cover designer needs before they can finalize the wrap — is calculated from your page count multiplied by the paper stock thickness your printer specifies. You cannot lock the jacket without locking the pages. You cannot lock the pages without a typeset.
Which is why typesetting is not, in any meaningful sense, a formatting job you can do in Word over a weekend. And why who does it — and how — shapes every physical artifact the book becomes.
What Typesetting Actually Is
Typesetting is the translation of a manuscript — a word-processing document with editorial markup — into a composed page layout that meets the production specifications of a printer. This means:
- Type is set in a professional layout application. Adobe InDesign is the industry standard; professional typesetters don’t produce final print files in Word.
- Page geometry is calculated to the printer’s spec: trim size, margins, gutter allowance for binding, bleed settings for print-ready PDFs.
- Typography decisions are applied: leading (line spacing), tracking (letter spacing), hyphenation and justification settings, the face and point size for body text, chapter headers, running heads, and folios.
- Special treatments are executed: drop caps, ornamental dividers, epigraphs, part title pages, interior photographs or illustrations if any.
- The PDF is output to the printer’s format. For offset, this is typically PDF/X-1a. For POD platforms, each platform has its own spec.
For ebooks, typesetting also produces a validated EPUB3 file — not a PDF, not a converted document, but a proper EPUB3 file that passes the W3C’s epubcheck validator. More on that distinction below, because the difference between a real EPUB3 and something that merely resembles one matters considerably for distribution.
This is the baseline. A literary novel with structural complexity — narrative breaks, chapter epigraphs, interior dialogue with unusual typographic treatment, recurring formal elements — adds to it. A debut literary hardcover adding sewn binding, French flaps, or a specific paper stock adds more.
POD vs. Offset: Two Different Technical Jobs
The distinction between print-on-demand and offset printing isn’t just economic — it’s technical, and it determines what your typesetter needs to produce.
POD typesetting works to standard platform specifications. KDP Print, IngramSpark, BookBaby — each publishes spec documents for trim size, interior PDF requirements, and spine width calculation formulas. A POD typesetter is working to those specs, which are standardized by platform and don’t require custom calibration per printer. IngramSpark’s print PDF specification page is the clearest public reference for understanding what POD technical requirements actually look like.
Offset typesetting works to the spec of the specific printer. A book printed by Thomson-Shore has different bleed and margin requirements than one printed by McNaughton & Gunn or Four Colour Imports. International offset printers — including Ingram Lightning, C&C Offset, and specialty literary printers — each maintain their own technical spec sheets. The typesetter needs the exact printer template before opening the file. A layout composed to IngramSpark specs dropped into a Thomson-Shore template will have wrong margins, possibly wrong spine placement, and will require a full reflow to correct.
This is why the printer decision upstream affects the typesetting job downstream. When evaluating typesetters for offset work, one of the first questions to ask is which printer specs they have established relationships with and current templates for. A typesetter fluent in IngramSpark is not automatically fluent in Thomson-Shore or Four Colour. That’s not a criticism — it’s a scope question. Match the typesetter to the printer.
The Four Qualifying Questions
A research-grade typesetter evaluation runs four questions before getting to portfolio or pricing:
1. Do you deliver offset-ready PDFs to custom printer templates? This distinguishes typesetters with offset production experience from those who work only in POD-platform specs. If the answer is “we need the printer’s spec sheet to get started” — correct answer. “Our files work with any printer” is the answer that needs a follow-up.
2. Do you have a literary fiction portfolio? The typographic sensibility for literary fiction differs from genre, business, or non-fiction work. Literary fiction tends toward more considered treatment — white space handled with intention, chapter architecture that serves the reader’s experience rather than announcing itself. A typesetter whose portfolio is business books will set a literary novel to look like a business book, not because they’re doing bad work but because their calibration point is different.
3. Do you deliver packaged InDesign source files — INDD and IDML — with the project? This is the source file question, and it matters more than it appears to. If your book goes back into editing — a corrected second printing, a new edition, an error caught post-press — the ability to make those corrections in the original layout file is the difference between a minor invoice and a full re-typeset. A typesetter who delivers only press-ready PDFs leaves you owning a static file you cannot modify. Own the source.
4. Do you produce a validated EPUB3 file? The specific follow-up: “Do you validate against epubcheck?” The answers you want to hear: “yes” and “yes.” The answers that signal a different conversation: “we convert your Word file to EPUB” and “we run it through Kindle Create.”
The EPUB3 Question (and Why the Conversion Workflows Fail)
EPUB3 is the current international standard for reflowable digital books. A file that passes epubcheck validation is a proper EPUB3 — it renders correctly on Apple Books, Kobo, Nook, Scribd, and Kindle (where it’s converted to Kindle format, but starting from a correct and complete source structure).
A file produced by converting a PDF to Word and running it through Kindle Create is not EPUB3. It is a Kindle file, produced via a conversion chain that introduces structural errors, loses typographic intent, and is not distributable to non-Amazon retailers in compliant form.
The distinction matters for distribution. IngramSpark, Draft2Digital, and Smashwords distribute to Apple Books, Kobo, Nook, and other retailers alongside Amazon. If your ebook file isn’t proper EPUB3, it may pass platform upload validation and still render incorrectly on specific devices — dropped italics, reflow errors on unusual screen widths, chapter navigation that doesn’t function. These aren’t hypothetical edge cases; they’re documented failures of PDF-to-conversion workflows on professional production projects.
When vetting typesetters, the question to ask is specifically: do you produce EPUB3 from native InDesign export, or do you use a conversion workflow? And do you validate the output file against epubcheck before delivery? “We validate against epubcheck” is the answer you’re looking for. “We submit it and it gets approved by the platform” is a different answer that doesn’t mean the same thing.
An otherwise-strong typesetter candidate can be eliminated from a shortlist specifically on this basis: an ebook production workflow that runs PDF-to-Word-to-Kindle Create may produce excellent print work but won’t deliver an ebook distributable to Apple Books or Kobo without additional remediation. For a book going to wide distribution, that remediation cost belongs in the original project scope — not as a surprise after delivery.
The Market, Named and Priced
The typesetting market runs from template-based services at a few hundred dollars to full-service literary interior design at $5,000 and above. Here’s the full landscape, priced from what scoped quote-collection consistently surfaces:
Template-based / low cost
Damonza offers print interior formatting starting around $395 for print and ebook combined, with a turnaround of approximately fourteen days and unlimited layout revisions. “Unlimited layout revisions” in a template-based service means adjustments within the template — spacing, margins, font choices within the available options — not custom chapter architecture. At this price point, you’re getting professional execution of a standardized template. That’s the right answer for digital-first novels, POD releases, and books where the interior aesthetic is less load-bearing. It’s the wrong answer for a debut literary hardcover where the interior needs to stand next to Graywolf or Coffee House on a bookstore shelf.
Mid-market / distribution-focused
Bookmobile occupies the interesting middle ground. Pricing ranges from roughly $1,000 for simpler projects to $4,000 and above for complex literary interiors, with offset production experience and a client roster that includes Graywolf Press and Coffee House Press. If the production standard you’re aiming for is peer-quality to small-press literary fiction — because that’s the shelf and the reviewer you’re targeting — Bookmobile’s portfolio is the right reference point. Their familiarity with the technical requirements of literary-fiction distribution, from custom offset specs to independent bookstore returns programs, is built into how they approach the work.
Reedsy’s interior designers range from mid-market to high-end, with the platform’s vetting process providing baseline credential verification. For ebook-only or POD projects where thoughtful interior design matters, Reedsy’s marketplace is a reasonable shortlisting resource.
Full-service literary design
1106 Design operates in the full-service tier. Their standard offering for a literary novel includes two original interior design concepts, revision rounds, a press-ready PDF delivered to your printer’s spec, and validated EPUB3 produced from native InDesign export — not conversion. Proposal-stage pricing for a full literary novel interior runs approximately $4,800–$5,000 before add-ons. The premium reflects what you’re actually buying: original interior design rather than template application; InDesign source files (INDD and IDML) delivered with the project; and an ebook that distributes correctly to every major retailer. When the book is going to a custom offset printer and to wide distribution, the source-file ownership and validated ebook are worth pricing into the original comparison.
INABI prices by word count: their Diamond tier (60,000+ words) lists at $2,500 for print and ebook, covering layout, one revision round, and PDF/EPUB delivery. Add-ons — rush fees, additional revision rounds, complex interior elements — stack on top of that base at rates that can push a complex project considerably higher than the listed tier suggests. Request a scoped quote rather than comparing menu prices; the base rate is rarely the delivered price for a literary novel with structural complexity.
Red Flags Worth Naming
Three patterns to look for and walk away from:
The PDF-to-EPUB pipeline. Any service producing ebooks by converting your PDF to Word and running it through Kindle Create or a similar tool is not delivering EPUB3. The test is simple: ask for the epubcheck validation report. If they can’t produce one, the file wasn’t validated against the standard. Fine if you’re Amazon-only and intend to stay there. A real problem if you’re distributing widely.
“Our files work with any printer.” Offset printing requires printer-specific technical specs. A typesetter who hasn’t actually worked through a custom offset printer’s template process — calibrating bleed, verifying spine calculation, matching paper-thickness adjustments to the printer’s formula — doesn’t know what they don’t know. Ask for a recent project delivered to a named custom offset printer and what their workflow was for matching that spec.
Source files as an add-on. Get the InDesign source files (INDD and IDML) included in the contract, explicitly. Some services deliver press-ready PDFs only and position the source files as an additional charge. The source files are not optional for a book you intend to maintain, correct, or reprint. They are the production asset. Own them from the beginning.
The Timeline and Lock Sequence
Here’s the dependency chain in full:
Content lock (manuscript final, copyedit reviewed, style sheet current) → Typesetting begins (four to eight weeks for a full literary novel, depending on complexity) → First-pass pages reviewed (author review round, queries flagged and answered) → Final pages approved → Page count locked → Spine width calculated (using the printer’s formula: page count × paper stock thickness) → Dust jacket or cover file finalized (back to cover designer with the locked spine measurement) → Press files released → Printer order placed.
Each step requires the prior one. This is the chain that makes typesetting the gate everything else waits on — which is also why it’s not a step you accelerate by running it in parallel with editorial. It requires the locked content that developmental editing, line editing, and copyediting are designed to produce. Building those passes before typesetting begins isn’t procedural caution; it’s the only order that works.
Working backward from a desired release date: add four to eight weeks for typesetting from content lock, plus printer lead time (eight to fourteen weeks for custom offset; three to five weeks for POD). The earlier you engage a typesetter and understand their availability calendar, the more margin you have at the production back end when something — and something always does — takes longer than projected.
Finding the Right Match
The standard vetting sequence: assemble a shortlist of three to five candidates based on literary fiction portfolio, offset production experience (or POD fit, if that’s your path), EPUB3 credentials, and source file delivery policy. Request scoped proposals rather than comparing menu rates — the menu rate for a literary novel with complex interior treatment is usually not the delivered price. Get the source file delivery terms in the contract before you sign, not after.
Jane Friedman’s self-publishing resources include a thorough breakdown of the production timeline and what each service provider is responsible for. The Book Designer maintains a regularly updated overview of interior design services and interior layout software — a useful supplementary reference for understanding what the deliverables should look like before you start vetting against them.
The decision that matters most is matching the typesetter to your specific production path. A template service at $395 is the right answer for a digital-first novel going to Kindle and IngramSpark. A full-service literary designer at $4,800 is the right answer for a debut hardcover going to independent bookstores, with sewn binding, going to press at 2,500 units. The mistake isn’t choosing the wrong price tier — it’s applying the wrong tool to the production path you’ve already committed to.
The typeset pages are the moment the manuscript becomes the book. Get the right person to set them — and get the source files when they do.
e.
Cross-links
- The Real Cost of a Book Cover — on the cover design pipeline that depends on your locked page count for spine width
- Offset vs. POD: Which One Your Novel Actually Needs — on the printing decision that determines your typesetting technical spec
- Four Edits Every Novel Needs (and What Happens If You Skip One) — on the editorial passes that lock the content before typesetting begins
- Style Sheets: The Unsexy Document That Saves Your Novel — on the typographic decisions the typesetter needs from the editorial record