How to convert a PDF to Markdown (free, in your browser)

Markdown is plain text with a few simple marks for structure: # for headings, - for list items, | for table columns. It's what note apps like Obsidian use, what GitHub shows, and what AI assistants read best. PDFs are the opposite: they're built to look the same everywhere, not to be edited or reused.

This guide shows how to get from one to the other cleanly.

The quick way

1. Open the Docuflat converter and drop your PDF on it.
2. Wait a few seconds. The conversion runs in your browser, so the file never leaves your device.
3. Check the preview, then copy the Markdown or download it as a .md file.

The rest of this guide covers what the result looks like, how to tidy it, and what to do with scanned pages.

What you get

Docuflat reads the text layer of the PDF and writes it out page by page:

• The file name becomes the top heading, and each page gets its own ## Page 1, ## Page 2 heading, so you can always find your place in the original.
• Line breaks follow the lines in the PDF, so short lines such as addresses, list items and table rows stay on their own lines.
• Pages with no text layer are recognised as scans and run through OCR, marked (OCR) in their heading.

What it leaves behind is the fixed page layout: fonts, columns, images and exact positions. That's the part that makes a PDF a PDF, and it gets in the way once you want to reuse the text.

Tidy up in two minutes

PDFs don't store "this is a heading" or "this is a table". They store pieces of text at positions on a page. So it's worth a quick pass over the result:

Add headings. Put ## in front of the section titles you care about. A document with a real outline is easier to scan, and AI assistants use headings to find their way around.

Fix lists. Add - in front of bullet points if you want them to show as a list.

Rebuild important tables. If a table matters, turn its rows into a Markdown table (| a | b |), or convert the original spreadsheet instead if you have it.

Check the reading order. Two-column layouts, sidebars and footnotes can come out in an odd order. Read the first page to make sure the text flows the way the original does.

Remove repeated headers and footers such as a company name or "Confidential" on every page.

If you have the document in its original Word format, convert the .docx instead: Word files keep their real headings, lists and tables. See Convert Word documents to Markdown.

Scanned PDFs

If the PDF is a scan or a photo of paper, there's no text inside it to extract, only pictures of text. Docuflat notices this and runs OCR (optical character recognition) on those pages. Choose the document's language in the OCR language picker before converting; OCR is much more accurate when it knows which alphabet to expect. Our guide to converting a scanned PDF to text covers this in detail.

Why convert to Markdown at all?

• For AI assistants. Pasting Markdown into ChatGPT, Claude or Gemini uses far fewer tokens than attaching the PDF, and the model sees the structure clearly. See PDF to Markdown for AI.
• For notes and wikis. Obsidian, Logseq, GitHub and most documentation tools store pages as Markdown.
• For editing. Markdown opens in any text editor and converts easily to Word, HTML or a new PDF later.
• For search and version control. Plain text can be searched, compared and tracked in Git, line by line.

Privacy

Many online converters upload your file to their server. Docuflat converts in your browser instead, so contracts, statements and other private documents stay on your device. You can read more in Is it safe to upload documents to online converters?

Summary

Drop the PDF on the converter, pick the right OCR language if it's a scan, and spend a couple of minutes adding headings and fixing any tables. You'll have clean, reusable Markdown.