Converting a manuscript to ebook format isn't as simple as hitting "Save As"
Most people who ask me to turn my book into an ebook have no idea how much actual work is involved once the manuscript leaves the page they wrote it on. The conversion itself takes maybe twenty minutes. The cleanup, the formatting checks, the metadata setup—that's where the real time goes. You upload your manuscript. A converter processes it. You get an epub or mobi file. That's the surface version of it. Here's what nobody tells you about the gap between those two points. First, your source file matters enormously. A clean .docx from Word with proper heading styles? That converts cleanly most of the time. A PDF scan, a Google Doc with every paragraph styled differently, or a manuscript pasted from a forum post? That's where things break. I once had someone send me a 400-page novel that was a single Microsoft Word document with no styles applied—just bold for chapter titles and a few line breaks. The automated converter spat out an ebook where every chapter title was formatted as body text, the section breaks created double page breaks everywhere, and the table of contents was completely broken because there were no heading tags for it to grab.
My workaround was to run the file through a stylesheet preprocessor first. I used a free tool called Calibre's "Edit Book" feature to strip the manual formatting, then re-applied basic heading levels using regex substitutions. Took about 45 minutes for a 400-page book. After that, the actual conversion ran clean on the first try.
The EPUB vs. Kindle Format Question
You need to decide early whether you're targeting Amazon Kindle, broad epub distribution, or both. They're different formats with different constraints. EPUB is the open standard. Kindle uses its own KF8 or AZW3 format, though Amazon now accepts EPUB uploads and converts them on their end. The counter-intuitive thing here is that uploading an EPUB to Kindle Direct Publishing often gives you a worse result than formatting directly for Kindle. Amazon's conversion pipeline is not gentle. It strips CSS it doesn't understand, reshuffles your layout, and occasionally drops entire sections of styling. If your book has complex formatting—poetry, code samples, sidebars, footnotes—you'll get better results by using Kindle Create or Sigil to build the Kindle file yourself.
Get the Full Details

Metadata Is Not Optional
This is the part everyone skips and then wonders why their book doesn't show up correctly in stores. Metadata includes the title, subtitle, author name, series number, genre categories, keywords, and description. Each store has different requirements. Amazon wants BISAC codes. Apple Books wants specific category paths. Google Play has its own keyword field that actually affects search ranking. I learned this the hard way when I submitted a technical book with proper BISAC codes in one system and forgot to include them in another. The book went live but appeared under the wrong category for three weeks before anyone noticed. Fixing it required a manual review request. Takes about ten minutes to do it right the first time across all platforms.
Common Pitfalls That Waste Hours
Broken hyperlinks are the most frequent problem. If your manuscript has footnotes that link to references at the back, the converter needs to know which are footnote markers and which are citations. Most tools handle basic footnote conversion, but custom link structures break silently—you won't know until a reader reports it. Font embedding is another trap. Some converters auto-embed fonts, some don't. If you use a specialty font for chapter headings and it doesn't embed, readers on certain devices see it fall back to a system font. Your carefully designed layout disappears on Kobo and Nook devices while looking fine on Kindle and iPad. Table of contents generation fails when heading levels are inconsistent. I've seen chapter headings marked as Heading 2 in one place and Heading 1 in another, which makes the TOC jump around unpredictably. The fix is to standardize all heading levels before conversion. Run it through a validator like the EPUBCheck tool—it's free and takes about three minutes to run. It will flag errors before you upload, saving you from having to re-submit.
What Automated Tools Can't Handle
Here's the blunt truth: automated "turn my book into an ebook" services work fine for straightforward novels with plain text and simple chapter breaks. They fail on anything with images, complex layouts, sidebars, tables, mathematical notation, or multi-column pages. If your book is a cookbook, a textbook, a graphic novel, or contains embedded media, you need manual editing in a tool like Calibre, Sigil, or Adobe InDesign. For image-heavy books, you also need to consider file size. High-resolution images in an EPUB can push the file to 100MB or more, which slows down loading on older devices. Downsample images to 150 DPI maximum for print-quality reproduction, or 72 DPI if they're decorative. This usually cuts file size by 60 to 80 percent with no visible quality loss on screen.

A Realistic Timeline
From finished manuscript to published ebook, budget between 2 and 8 hours depending on complexity. A simple novel with clean formatting: 2 hours. A nonfiction book with images and references: 4 to 6 hours. A textbook or heavily formatted work: 8 hours or more. Factor in testing on at least two different reading apps before you consider it done.