A PDF fixes text in place so it prints the same everywhere, which is exactly why it is hard to edit. This converter rebuilds the PDF as a Word document (.docx) you can change in Microsoft Word, Google Docs, Apple Pages or LibreOffice.
It reads the PDF's text layer in your browser with pdf.js: every piece of text with its position, size and font. Pieces on the same baseline become lines, lines with the same size, spacing and left edge become paragraphs (with line-end hyphens removed), and larger text becomes Heading 1, 2 or 3, so Word's navigation pane works. Bold and italic come from the fonts, centred and indented lines keep their alignment, and every PDF page becomes its own Word section with the same page size and margins, so page breaks line up with the original.
Text set out in rows and columns becomes a real Word table; a bold header row repeats at the top of each page. Honest limits: merged cells, borders and shading aren't copied, and the PDF's exact font names aren't readable in the browser, so text is set in Calibri, Times New Roman or Courier New to match its style. Images, charts and drawings aren't extracted as separate objects; turn on "Add a picture of each page" to keep them as an image after the page's text. Pages in two or more columns are read across the page. Scanned PDFs have no text layer, so their pages are added as pictures, and their text isn't editable without OCR.