Question 1

What exactly converts to Markdown — and what doesn't?

Accepted Answer

Text and structure: paragraphs (soft-wrapped lines joined back), headings ranked by font size (# for the biggest tier, up to ######), **bold** and *italic* runs, bullet lists (-) and numbered lists (1.) when the lines are visibly marked, > quotes when a passage is indented AND italic, and link annotations as [text](url). Not converted: tables (use PDF to Excel — it is built for them), images, colors, and fonts. A line that literally starts with *, #, _ or [ is backslash-escaped so it can never turn into fake Markdown.

Question 2

My PDF has two columns and tables — how do those come out?

Accepted Answer

Honestly approximate, and we say so. Two-column pages come out as a reading flow where a line from the left and a line from the right column that share a baseline may join into one line — fine for notes, not a faithful layout. Tables are NOT converted to Markdown tables: their cells arrive as plain text lines. When you need the table, run PDF to Excel on the same file — it reconstructs grids properly; when you need the prose, this tool is the right one.

Question 3

What do the switches do — one file per page, rules, headers?

Accepted Answer

Pages picks a subset (1-3, 5, 8- with an open end). "One file per page" writes a separate .md per selected page instead of one document — handy for notes per chapter; the result screen also offers everything as a ZIP. "Rules between pages" adds a --- thematic break where each PDF page starts, keeping page boundaries visible in one file. The provenance comment adds  at the top; YAML front matter adds a --- block with title, source and page count for static-site generators (both are off by default — the .md stays clean).

Question 4

What happens with scanned or encrypted PDFs?

Accepted Answer

A scanned page is an image with no selectable text, so the tool stops honestly with a "no selectable text" message instead of saving an empty .md — run the PDF OCR tool on the file first (English and Hindi are supported), then convert the result here. Encrypted files fail with a clear "this PDF needs a password" message; unlock the file first, then convert. And heading ranks come from the document's own font sizes, so a PDF that is all one size comes out as plain paragraphs — honestly, not as invented headings.

PDF to Markdown

How it works

Frequently asked

PDF to Markdown converter