Basic vs. Enhanced — and why neither is "perfect"
A Word document is made of paragraphs, tables, and styles. A PDF isn't: it stores drawing instructions, roughly "put the letter A at this exact spot on the page." When you open a PDF you see a perfectly laid-out document, but underneath there's no concept of "this is a table" or "this is a paragraph" — just text placed with pinpoint precision.
Basic takes the fastest, safest path: every line pdf.js can read becomes its own paragraph. It's instant and reliable, but reads a little choppy if you plan to edit it, since sentences that wrapped across lines in the PDF stay broken into separate paragraphs.
Enhanced does more work: it reflows lines back into natural paragraphs, spots headings by comparing font sizes across the document, and uses tab stops to keep simple side-by-side columns roughly aligned. It's a genuinely better starting point for editing — but it's still a best-effort heuristic, not a layout engine, so it doesn't preserve bold or italic styling, and multi-column pages and real tables may not line up perfectly.
Rebuilding a PDF's exact layout — tables, columns, images in their exact place — is a much harder problem that normally needs a heavy conversion engine running on a server, which is usually a paid product. Enhanced is our free, in-browser attempt to close some of that gap; it's unlocked after a short ad because that's what lets us keep it free with no accounts and no subscriptions.
Note: if your PDF is a scan (a photo of a paper document), neither version can extract text from it, because there's no real text inside the file — only an image. That would require optical character recognition (OCR), which this tool doesn't include.