How to Move Your Drafts Into Ulysses Without Losing Formatting
Exporting out of a writing app is easy. Getting 40 old drafts into a new one with their headings, italics and bullets intact is the part nobody writes about, and it is why a lot of people pay for Ulysses and then keep writing in Google Docs anyway.
Direct answer: convert everything to Markdown (or DOCX) first, import into one group, then fix heading levels manually. Plain text and Markdown imports survive almost perfectly; pasted rich text does not, and PDF imports are a last resort.
Which Import Formats Actually Work?
| Source | Best route | What survives |
|---|---|---|
| Plain .txt / .md | Drag into group | Everything, if the markdown is valid |
| Word .docx | File → Import | Headings, bold/italic, lists |
| Google Docs | Download as .docx, then import | Most formatting, but no comments |
| Scrivener | Compile to .docx | Structure lost unless you use folders as chapters |
| Avoid | Paragraph breaks only, sometimes broken mid-word |
The DOCX route is the workhorse. Google Docs → File → Download → Microsoft Word, then import the .docx into Ulysses. Footnotes convert cleanly; Google’s suggestion mode edits do not.
How Do You Convert a Mess of Files to Markdown First?
If you have 40 files in mixed formats, batch-convert with pandoc rather than doing them one at a time:
pandoc notes.docx -t markdown -o notes.md
Run it in a loop over the folder, then import the .md files. Pandoc’s markdown output is conservative and imports cleanly. Watch out for smart quotes — curly apostrophes in old documents sometimes translate into odd characters depending on the source encoding.
Where Does Formatting Still Break?
Three predictable spots:
- Nested lists. Three levels deep usually survive. Four and the indentation collapses.
- Tables. Markdown tables import as text rows unless the source has the pipe syntax. Word tables become paragraph runs.
- Heading levels. Word’s “Heading 1” often lands as a sheet title rather than an H1 inside the sheet. Fix this before the export step, not after.
Budget roughly one hour of cleanup per 20 documents. That is realistic, not pessimistic.
How Should You Organise the Imported Library?
Do not dump everything in one group. Build the target structure before importing, because dragging 200 sheets afterwards is misery.
- Create the top-level groups: In Progress, Published, Archive.
- Inside Archive, split by year, not by topic. Topics drift, years do not.
- Import in batches of one group at a time so you can check output per batch.
- After each batch, spot-check the three documents you care most about.
What About the Stuff That Never Converts?
Comments and tracked changes disappear entirely. If you have editorial feedback threaded in a document, export a PDF of the commented version and store it alongside — Ulysses has no comment system.
Custom styles (a house CSS look, for example) also do not come along. Ulysses has its own style presets, and you rebuild them under Preferences → Styles. For a two-person shop, one preset is enough.
Is the Migration Worth It for Occasional Writers?
No. If you publish less than once a month, the hour of cleanup plus $49.99/year buys you very little. The migration pays off for people with a body of work they search and reuse — the library-wide keyword search across 300 sheets is the real long-term value.
Key Takeaways
- Convert to DOCX or Markdown before importing; paste rich text and you will redo the formatting.
- Google Docs → .docx → Ulysses is the most reliable multi-format path.
- Use pandoc for batch .docx → .md conversion instead of importing one by one.
- Nested lists past three levels, tables and Word heading styles are the usual breakages.
- Comments and custom styles never transfer; archive a PDF if the feedback matters.
