Skip to main content

Workflow guide

Import documents into Obsidian as Markdown

An Obsidian vault stores notes as local Markdown files, so document conversion can remove a large part of the manual retyping. It does not reproduce a Word page, a slide layout, or an image inside the vault. The reliable workflow is to extract useful structure, review it, then add Obsidian-specific links, properties, and attachments yourself.

Last updated
August 12, 2026
Test environment
To-Markdown web converter; Firecrawl AnyDoc Word and PowerPoint examples; Zhipu vision OCR image example

1. Prepare the source before import

For Word files, accept or reject tracked changes, keep meaningful heading styles, and save any embedded images you need as separate files. For PowerPoint, decide whether the slide sequence is a useful note outline and identify charts or spatial relationships that need a written explanation. For screenshots or scans, crop and rotate the image, use the highest practical resolution, and remove unrelated UI.

Convert one source at a time. The web converter accepts one file up to 10 MB. Download or copy the generated Markdown, but keep the source document until the imported note has been checked.

  • Use a descriptive destination file name such as migration-checklist.md.
  • Choose a vault folder based on the note's subject, not its original file type.
  • Save required images in the vault's configured attachment folder.

2. Review the Markdown before copying it into the vault

Compare headings, nested lists, links, tables, names, and dates with the original. A simple Word table can become a pipe table, while merged cells or layout tables may need to be rewritten. Slide content is flattened into reading order. OCR can infer headings and tables, but punctuation, symbols, handwriting, and narrow columns are not guaranteed.

The Word converter's published sample exposes a real edge case: the generated table can include a blank header row before the source's Item and Owner row. Remove that row if it does not belong in the note. Publishing observed quirks is more useful than assuming every conversion is clean.

  • Open the file in Obsidian's source mode when debugging Markdown syntax.
  • Use normal Markdown links for web URLs and add wiki links only after note names are stable.
  • Add image embeds after copying attachment files; conversion does not export Word images.
  • Check whether a plugin expects properties in YAML front matter.

3. Add Obsidian-specific structure deliberately

Add aliases, tags, or properties only when they support an actual navigation or query workflow. Avoid generating a large tag list from every source word. Link the imported note to an existing index note, and add links to related notes where the relationship is meaningful.

If the document will be updated repeatedly, keep a stable source field and define whether future imports replace the note or create dated snapshots. Manual edits and automated overwrites otherwise become difficult to reconcile.

  • Record source, imported date, and owner in properties if your vault uses them.
  • Add one link from an index or project note so the imported file is not orphaned.
  • Review attachment paths on another device if the vault is synchronized.
  • Search for unresolved placeholder text before deleting the original document.

Observed Word outline ready for an Obsidian note

The Word converter page publishes a reproducible DOCX sample. Its tested output keeps the heading hierarchy, list, and hyperlink shown here.

Source elements

Heading 1 and Heading 2, two bullets, one external link, and an Item / Owner table

Observed Markdown

# Migration checklist

## Before you publish

- Review the generated file
- Check tables and images

Read the [Markdown guide](https://www.markdownguide.org/) before publishing.

Next step: After review, save the result as a .md file in the vault, add any exported images, and create the Obsidian links or properties your vault actually uses.

Review checklist

Complete these checks against the original source and the real destination before treating the Markdown as finished.

Check Acceptance criterion
Headings One clear note title and a logical section hierarchy.
Tables No blank, merged, or shifted cells that change the meaning.
Attachments Required images were exported and linked from the vault.
Links External URLs still work; wiki links point to intended notes.
Properties Front matter is valid for the plugins and queries in this vault.
Navigation An index, project, or related note links to the imported file.

Use a tested source converter

Word and DOCX to Markdown Start with the published DOCX example and support matrix. PowerPoint to Markdown Flatten slide text into a reviewable note outline. Screenshot and image OCR Extract visible content, then correct OCR mistakes.

Related workflow guides

Convert documents to Markdown for RAG ingestion A practical document-to-Markdown workflow for RAG ingestion: extract source content, review structure, preserve provenance, then chunk and index it downstream. Prepare converted Markdown for GitHub documentation Turn Word or PowerPoint content into reviewed Markdown for GitHub, MkDocs, or Docusaurus, including links, assets, front matter, navigation, and build checks.