Cause one: a permission flag
The page has real text. The document carries permission bits that switch copying off, and the reader grays out the copy action.
Signs of it: the file opens without asking for a password, search often still works, and Print and Edit are grayed out at the same time.
- Fix it on a copy you own, since the unlock tool clears the flags. It reports the scheme and which permissions were switched off before changing anything.
Cause two: there’s no text layer
The page is an image. A scan from a copier, a photo of a receipt, a PDF exported from a picture, or a fax gateway’s output. Nothing about it is text, so nothing about it can be selected, and no amount of clicking will change that.
- Tell-tale: the file is enormous for its page count. A text page costs a few kilobytes, and a 300 dpi scan costs a few hundred.
- Tell-tale: zoom in far and the letters go soft and blocky. Real text stays crisp at any zoom, because it’s drawn rather than sampled.
- Tell-tale: the cursor never changes as it crosses a word.
- Tell-tale: search finds nothing, including words you can plainly read on screen.
Tell them apart in five seconds
Open the file in the extract tool here and let it run. If the preview comes back empty, there was no text layer. If it comes back readable, the text was there all along and something was forbidding you from selecting it.
That’s a diagnosis you can run on a payslip or a medical letter without sending it anywhere, which matters, because those are exactly the documents people are tempted to paste into a converter they found in a hurry.
- Opens with no password, copy grayed out
- Permission flags, reversible on a copy you own
- Cursor never becomes a text cursor
- Image page, no text layer
- Search finds nothing, zoom goes blurry
- Image page, no text layer
- Huge file, few pages
- Image page, almost certainly a scan
What your options actually are
- Permission flag: clear it on your own copy. The text layer comes back intact, because it was never gone.
- Image page, and you need the words: tick the extract tool’s OCR option. It reads pixels, so the pages have to be in the tab, and it handles printed English well and handwriting badly. For three lines, retyping is still the faster route.
- Image page, and you need the page: screenshot it, print it, read it. There’s nothing hidden underneath. The picture is the document.