Convert from PDF guide
Convert PDF to Markdown for a Docs Site
Updated
Short answer: Drop the PDF into PDF to Markdown; larger text becomes headings and wrapped lines become paragraphs, ready to paste into your docs.
Docs sites built with tools like MkDocs, Docusaurus or a Git-based wiki want Markdown, while legacy manuals and specs live in PDFs. Converting gives you a first draft in the right format, which is far quicker than retyping.
Step by step
Step 1: If the PDF is a scan, run OCR first so there is text to convert. Try selecting a word to check.
Open OCR PDF(runs on our server)Step 2: Open PDF to Markdown and drop the file.
Open PDF to Markdown(runs on our server)Step 3: Open the .md file in your editor and check the heading levels match the document's structure.
Step 4: Rebuild tables and add images by hand; extract the images separately.
Open Extract images from PDF(runs on our server)Step 5: Split long manuals into one page per chapter, following your docs site's navigation.
Why this works, and what to watch for
PDFs don't store structure the way Markdown does. There's no “this is a heading” tag in most PDFs, just text placed at positions in fonts of certain sizes. PDF to Markdown infers structure: text noticeably larger than the body becomes a heading, with the biggest size as # and smaller ones lower down, and lines that wrapped in the PDF are joined back into paragraphs. Page headers and footers may repeat in the output and are worth deleting.
What doesn't come through: images aren't included, and tables and multi-column layouts arrive as plain lines of text, because a PDF stores table cells as separately positioned words. For a document with important tables, convert those pages with PDF to Excel or PDF to Word and rebuild the Markdown tables from there. Pull out diagrams and screenshots with Extract images, then reference them from your Markdown.
Code samples are the other thing to check. PDFs don't distinguish code from prose, so code may arrive as ordinary paragraphs with its indentation flattened. Wrap it in fenced code blocks and restore the indentation from the original.
If your target is HTML rather than Markdown, such as a single reflowing web page, PDF to HTML does the same structural reconstruction and outputs simple HTML. For feeding a document into a search index or an AI tool, Markdown is usually the better choice: it's compact, and headings give useful structure.
Each conversion is one server job toward your daily limit. The file is processed and returned straight away and not stored.