PDFJadoo › Blog › PDF to Markdown, Cleanly

PDF to Markdown: structure out, ready to paste anywhere

Markdown is the plain-text format the modern world drafts in — notes apps, documentation, GitHub, static sites, AI tools all read it natively. Getting a PDF's content into that world is PDF to Markdown's job: the document's structure is recovered — headings from font sizes, lists from their markers, paragraphs re-joined from their wrapped lines — and delivered as clean Markdown text to copy or download. The requirement is stated up front: the PDF must carry a real text layer (a scan needs OCR first), and multi-column layouts may need a human eye on the reading order. This article explains how the recovery works and how to judge the result.

Free, in your browserFiles never uploadedHeadings · lists · paragraphs — structured
PDFJadoo logoBy the PDFJadoo Team5 min read
Diagram: a PDF page transforming into Markdown text with hash headings and list dashes
Not a text dump: the document's skeleton, recovered and written in Markdown.

How structure is recovered from a flat page

A PDF does not store “headings” — it stores text runs at sizes and positions. The converter infers the skeleton from those signals: text set larger than the body size becomes a heading (the largest becoming the top level); lines beginning with bullets or numbers become list items; wrapped lines are re-joined into paragraphs so the Markdown flows as prose rather than as the page's line breaks. The result is Markdown that reads as the document was meant, not as the page happened to break it — which is precisely what pasting into a notes app or docs system needs, and what a raw text extraction (which keeps every line break and flattens every heading) cannot give.

The inference has honest limits, and the tool states the main one: multi-column layouts. A two-column page's text runs interleave in the file's storage order in ways no algorithm can always untangle — the output may read down one column and then the other (correct) or across (a jumble), depending on how the PDF was produced. The remedy is the human eye the tool asks for: scan the output's joins, especially at column breaks, before the text is built upon. Tables, similarly, come out as their text rather than as Markdown table syntax — structure recovery is confident about headings, lists and paragraphs, and modest about grids.

Where the Markdown goes next

The destinations explain the format's value: notes and knowledge bases (the report becomes a searchable, linkable note), documentation systems (the PDF's content joins the docs site), version control (text diffs meaningfully; PDFs do not), and republishing — Markdown typesets onward to web pages or back to PDF (Markdown to PDF) with fresh design. In each, the conversion is the bridge from the document's fixed past to a malleable future.

Convert a PDF to Markdown, step by step

Step 1 — confirm a text layer exists. Select a sentence in the PDF: selectable means a layer. A scan has none — OCR it first (Make a Scanned PDF Searchable).

Step 2 — drop the PDF onto PDF to Markdown. The reading and structuring happen in your browser.

Step 3 — copy the Markdown or download the .md file.

Step 4 — proof the joins. Read the output where the layout was complex: column breaks, list boundaries, heading levels. Fix the few jumbles by hand — minutes, against the hours of retyping the document.

Step 5 — paste it into its new home. Notes, docs, repo, site — the text is now a citizen of the text world.

Markdown out, Markdown in

The round trip is worth knowing: PDF → Markdown liberates content; Markdown → PDF publishes it. Teams that keep their masters in Markdown and treat PDFs as exports never need the liberation step at all — the converter exists for the world's far larger stock of documents whose only surviving form is the PDF.

Set the words free — skeleton included.

Headings, lists and paragraphs recovered from the page's signals — clean Markdown to copy or download.

Open PDF to Markdown

A worked example: the policy manual joins the wiki

A company's procedures live in a 60-page PDF, updated by emailing a new PDF — unsearchable, unlinkable, un-diffable. Converted to Markdown, the manual moves into the company wiki: headings become the wiki's structure, each procedure gains a URL, edits become tracked changes, and search finds the clause instead of the file. The conversion took minutes; the proofing pass — two column-jumbles in an appendix, fixed by hand — took twenty. The PDF remains as the printable export, regenerated when the wiki changes: the master moved to text, where masters belong.

Edge cases

Heading levels flatten. Font-size inference distinguishes big from small; a document with four heading sizes may map imperfectly. Adjust levels in the pasted result — a two-minute normalisation.

Headers and footers repeat. Page furniture (running titles, folios) is text too, and may appear in the flow. A quick find-and-remove pass cleans the seams.

Scans. No text layer, no conversion — OCR first, then convert the searchable result.

Before you finish: the 60-second check

Fair questions

How do I convert a PDF to Markdown?

Confirm the PDF has a text layer (its text is selectable), drop it onto PDF to Markdown, and copy the result or download the .md file. Headings are recovered from font sizes, lists from their markers, paragraphs re-joined.

Why did my scanned PDF produce nothing useful?

Scans are pictures of pages with no text layer to recover. OCR the scan first — Make a Scanned PDF Searchable — then convert the searchable result to Markdown.

How is this different from extracting text?

Plain extraction dumps the text with the page's line breaks and no structure. Markdown conversion recovers the skeleton — headings, lists, paragraphs — so the output pastes into notes and docs as a structured document.

What about tables and multi-column pages?

Tables come out as their text rather than Markdown table syntax, and multi-column reading order can jumble depending on how the PDF was produced. Proof the output at those places and fix the few joins by hand.

Where can I use the Markdown?

Notes and knowledge bases, documentation systems, GitHub and version control, static sites — and back into a typeset PDF via Markdown to PDF when a document export is needed.

Is my PDF uploaded?

No — the conversion runs in your browser; the file never leaves your device.

Tool behaviour — structure recovery from font sizes and list markers, the text-layer requirement, the multi-column reading-order caution, and copy-or-download delivery — is from PDFJadoo's own PDF to Markdown page. Checked 18 November 2026.

Keep reading