Why Can't I Copy Text From a PDF? Two Causes, Two Fixes

By PDF Editing Tools Editorial Team ·August 27, 2026 · Updated August 27, 2026

You can read the paragraph perfectly well on screen. You drag across it and nothing highlights — or it highlights, you press copy, and what lands in your clipboard is nothing at all. Both are common, they have different causes, and they need different fixes.

Two causes, and only two

Almost every case comes down to one of these:

1. There is no text to copy. The page is a picture of a document. What looks like a sentence is an arrangement of pixels, and there are no characters underneath it.
2. There is text, but copying is restricted. The characters exist and your reader can see them, but the file carries a flag asking readers not to permit extraction.

Everything else — a reader bug, a stubborn preview pane, a page still loading — resolves itself by opening the file properly in a different application. If the behaviour follows the file from one reader to another, it is one of the two above.

The five-second test that tells you which

Try to select the text with your cursor.

Nothing highlights at all. Your cursor behaves like it is dragging across a photograph, because it is. This is cause 1 — a scanned or image-based page.

Text highlights, but copy produces nothing. The selection working proves characters exist. Something is refusing the extraction, which is cause 2.

That single test decides which half of this guide applies to you. It is worth doing before anything else, because the two fixes have nothing in common.

A useful second signal: try searching the document for a word you can clearly see. If search finds nothing, there is no text layer — cause 1 again. This is the same underlying issue as why a PDF is not searchable.

Cause 1: the page is an image

This is by far the more common of the two.

A PDF made by scanning, photographing, or faxing a document contains one picture per page. Your eyes read it fine, and to a computer it is a grid of coloured dots with no more linguistic content than a photograph of a beach. There is nothing to select because nothing textual was ever stored.

You can often recognise these files before testing. They tend to be much larger than a text document of the same length, because photographs of paper are heavy while characters are cheap. They may show slightly grey or uneven backgrounds, faint scanner edges, or a barely perceptible tilt. And zooming in far enough makes the letters blur into pixels rather than staying crisp — real text stays sharp at any magnification, because it is drawn from a font rather than stored as a picture.

The fix: create a text layer with OCR

Optical character recognition examines the shapes on each page, decides which characters they represent, and writes an invisible text layer positioned over the image. The page still looks exactly as it did — the same scan, unchanged — but there are now characters behind it to select, copy, and search.

OCR PDF does this in your browser. The recognition engine runs locally in a background worker, so the pages are never uploaded, which matters when the document you are trying to copy from is a contract or a medical record.

The sequence is:

1. Open OCR PDF and select the scanned document.
2. Choose the document language if more than one is offered.
3. Run recognition and wait — this is the slow step, and how slow depends on your own hardware rather than a queue.
4. Export a searchable copy, then copy text from that file.

What OCR will and will not manage

Recognition works from the pixels it is given, so the quality of the scan sets the ceiling.

A straight, well-lit scan of printed text at a reasonable resolution produces excellent results. Accuracy drops — sometimes sharply — with low-resolution scans, handwriting, pages that are rotated or tilted, decorative fonts, complex tables, and documents mixing several languages on one page.

Two things measurably improve the outcome and cost almost nothing. Rotate the pages upright first, because a sideways page gives the engine nothing to work with. And crop away the scanner edges, since black borders and speckle are noise the engine will try to interpret.

Always read back anything that matters. No OCR engine is perfect, and the errors it makes are the quiet kind — a 5 read as an S, a 0 as an O, a decimal point lost. For a phone number or a figure in a contract, check the copied text against the page before relying on it.

Cause 2: the file restricts copying

If text highlights but will not copy, the document carries permission settings.

PDFs support two different passwords, and confusing them is the source of most of the frustration here.

A user password is required to open the file at all. Without it nothing displays — you would not have got as far as trying to select text.

An owner password leaves the file freely openable while marking certain actions as restricted: printing, copying, extracting, modifying. The document opens instantly for anyone, and only reveals the restriction when you try to do the marked thing.

So a file you can read but not copy from is almost always carrying an owner password. Both mechanisms are defined in the PDF specification, published as ISO 32000-2, and the PDF Association maintains reference material on how the format handles them.

These permissions are honoured, not enforced

This is the part that surprises people, and it is worth understanding because it explains the inconsistent behaviour you may already have noticed.

Permission flags are a request recorded inside the file. A well-behaved reader reads them and disables the corresponding menu item. A reader that ignores them will happily let you copy, because nothing about the text is actually encrypted — the characters are sitting there in plain form.

That is why the same document sometimes copies in one application and refuses in another, and why advice threads are full of people reporting contradictory results on the same file. They are not describing different files; they are describing different readers making different choices about a flag.

What we do, and what we deliberately do not

Unlock PDF exports an unprotected copy of an encrypted document — when you supply the correct password. It validates the password locally and, if it is right, produces a copy that no longer prompts. That is genuinely useful when you own a protected file and need a working copy for a process that cannot handle encryption.

It does not guess passwords. There is no dictionary attack, no brute force, and no bypass of encryption you do not hold the key to. That is a deliberate design decision, not a missing feature — see how to unlock a PDF when you know the password for what the tool does step by step.

Which leads to the honest answer for the most common version of this problem. If someone sent you a document with copy restrictions and you do not have the owner password, this site does not offer a way around that, and no amount of clicking will change it.

What to do instead

The practical routes, in the order worth trying:

Ask the sender. By far the fastest. Restrictions are frequently applied by a default setting in whatever produced the file rather than by anyone’s deliberate decision, and a request for an unrestricted copy is usually met with one.

Ask for the source. If you need the text, the original document — the word processor file, the spreadsheet, the email it was written in — is better than any extraction from a PDF, because it carries the structure a PDF discarded.

Retype the part you need. Unglamorous, but for a paragraph or a reference number it is faster than the alternatives and it always works.

Consider whether you should. Restrictions are sometimes there for a reason — licensing on a report you paid to read, confidentiality on a document circulated in confidence. The flag is weak, but the intention behind it is information, and it is worth a moment’s thought before working around it.

When both causes are present at once

A scanned document can also carry permission flags, and this combination produces the most confusing symptom of all: nothing selects, and even if it did the file would refuse to copy.

Work in that order. Recognition has to come first, because there is no text to restrict until OCR creates some — and the text layer OCR produces is new content in a new file, so the original restriction has nothing to attach to. If the file is also encrypted with a user password so it will not open at all, that has to be resolved first, and only with the correct password.

A quick reference

Once the text is there

With a searchable copy in hand, Search PDF will find a phrase across a long document and step you through every match — useful when you needed one clause out of ninety pages rather than the whole thing.

Everything described here runs in your browser. The recognition, the search, and the export all happen on your own machine, so a document you were not able to copy from does not have to be handed to a third party in order to become one you can.

Open OCR PDF