Manuscript Export Formats, and What Each One Is Actually For
A finished manuscript has to take a shape someone else expects: an agent, a formatter, an ebook store, a print-on-demand service. None of them expect the same shape, which is why a single "export" button hides four genuinely different jobs. Knowing which format does which job saves a re-export at the worst possible moment, usually the night before something is due.
Markdown: the plain-text version with just enough structure
Markdown is ordinary text with a handful of light marks added: a pound sign for a heading, asterisks for italics, nothing else. It opens in almost anything, forever, because nothing about it depends on one program's internal format, and the marks themselves are still readable even if nothing on earth knows what Markdown is anymore. That makes it the right choice for archiving a manuscript, moving it between different tools without losing the paragraph breaks and chapter headings, or reading it in five years on whatever you happen to be using by then. It's a poor choice for anything that needs precise page layout, because Markdown doesn't really have an opinion about pages at all — margins, page numbers and fixed fonts sit outside what it describes, which is the words and their basic shape.
DOCX: the format a human editor expects to open
DOCX is a word processor document, and it's the format most agents, editors, and formatting services still ask for by default, because it's what opens directly in the tools they already use for tracked changes and comments. Many submission guidelines still describe a specific look for this stage: a plain, readable font, generous spacing, page numbers in a header, nothing decorative. Use DOCX when a real person on the other end needs to mark up your pages rather than simply read them. The trade-off is that it carries a lot of formatting information along with the words, which is exactly what an editor wants and exactly what makes it a clumsy format for archiving or converting later, since all that extra structure can shift when it passes through a different program.
EPUB: the shape an e-reader actually understands
EPUB is built for reflowing text: it lets a reader change the font size, the screen width, or the device entirely, and the pages rearrange themselves to fit, the way a paperback can't. This is what an ebook store expects, and it's structured internally as a small set of linked files rather than one continuous document, which is also why a manuscript exported straight to EPUB with no adjustment can come out looking stranger than the same file opened as a DOCX. Chapter breaks, in particular, are worth checking after export: EPUB treats each one as its own small file, so a missing break can quietly merge two chapters into one on the reading screen. It's built for reading on a device that moves, not for a printer.
Plain text: the version that survives everything else
A .txt file has no formatting at all, just the words, and that's the entire point. It's the format most likely to still open correctly decades from now, on a machine that doesn't exist yet, because there's nothing in it that can go out of date, no styling to render wrong and no proprietary structure to fail silently. You wouldn't send it to an agent or load it onto an e-reader, but it's exactly what a backup should be: a copy stripped down to nothing but the thing that actually matters, sitting quietly on a disk in case every other format, and every app that produced them, ever stops opening the way it used to.
What gets lost when a format converts to another
Every conversion between these four loses something, and knowing what in advance beats discovering it in a finished file. Going from DOCX to Markdown usually drops tracked changes and comments, since Markdown has no concept of either. Going from Markdown to DOCX gains nothing extra, only the plain structure you already had, dressed in a template. Going from DOCX to EPUB can scramble footnotes and tables, because reflowing text has to decide where a table goes when the page itself no longer has a fixed width. This isn't a flaw in any one format. It's what happens whenever information built for one job gets pushed into a shape meant for another, which is exactly why checking the result after a conversion matters more than trusting the conversion itself.
Matching the format to the moment
The pattern underneath all four is simple once it's named: Markdown and plain text are for keeping and moving your own words safely, DOCX is for the moment another person needs to work on them, and EPUB is for the moment a reader picks the book up on a screen. A single manuscript will likely need all four at different points: a plain-text backup running quietly in the background from the first page, a DOCX the day an editor asks for one, and an EPUB much later, once there's a book to read rather than a draft to fix. Choosing where to write shouldn't mean choosing which of these four you'll be stuck without later.
How Scrin does it
Exports leave as Markdown, HTML, DOCX or EPUB, with PDF available through native printing. A plain-text copy is also written to your own disk every day, automatically.