Quick answer
If you can't select text in a PDF, the page is most likely a scanned image, not real text. Run OCR (optical character recognition) to add a hidden text layer, and the page will look the same but become searchable and copyable. If text selects but copying is blocked, the PDF has permission restrictions that need the owner's password.
You open a PDF, try to copy a paragraph, and nothing happens. Either the whole page turns blue like a picture, or the cursor won't select anything, or the text pastes as gibberish. This is one of the most common PDF frustrations, and it almost always has one of three causes. Each has a different fix.
Three reasons you can't copy text from a PDF
| What you see | Most likely cause | Fix |
|---|---|---|
| Clicking selects the whole page, or nothing at all | The page is a scanned image with no text layer | Run OCR |
| Text highlights, but Copy is grayed out or pastes nothing | The PDF has permission restrictions | Remove them with the owner password, or ask the sender |
| Text copies, but pastes as random symbols | The PDF uses fonts without a proper character map | Run OCR on the affected pages, or convert to Word |
The first case is by far the most common. Anything that went through a scanner, a phone scanning app or a fax machine is, by default, a photograph of paper. The PDF reader shows you the picture, but there are no letters inside the file to select.
What OCR does
OCR, short for optical character recognition, looks at the image of each page, finds the shapes of letters and words, and turns them into real text. A good OCR tool then places that text in an invisible layer exactly over the matching words in the image.
The result looks identical to your original scan. But when you drag the cursor across a line, it selects the words; when you press Ctrl+F, it finds them; and when you copy, you get text you can paste into an email or a document. Screen readers can also read the page aloud, which makes the document accessible.
Step by step: make a scanned PDF copyable with MyPDFDone
- 01Confirm it is a scanTry to drag-select a single word. If the whole page highlights as one block, or nothing highlights at all, the page is an image and needs OCR.
- 02Upload the file to OCR PDFOpen MyPDFDone's OCR PDF tool and drop the scanned PDF, JPG or PNG on the upload box. Files up to 100 MB and 200 pages are accepted.
- 03Pick the document's languageChoose the language the document is written in. The right language noticeably improves accuracy, especially for accents and special letters.
- 04Test the previewThe preview is free. Check that the pages look unchanged and that the recognized text matches the original on a few lines.
- 05Download and copyDownload the searchable PDF for $4.99, paid once, open it in any PDF reader and select, search or copy the text as usual.
Processing and the preview are free, so you can see whether the recognition works for your document before paying anything. The download is a one-time $4.99 for that file. There's no subscription hidden behind it, and files are deleted automatically after 2 hours.
Tips for better OCR results
- Straighten and crop first. Recognition is best on pages that are straight and evenly lit. If you're scanning with a phone, use the app's scan mode rather than the regular camera; it flattens and crops the page for you.
- Scan at 300 dpi. Lower resolutions make small letters ambiguous; higher ones don't help and make the file larger.
- Use the right language. Accented letters, Cyrillic and other non-English characters need the matching language selected.
- Split very long scans. OCR PDF handles up to 200 pages per file. For a longer scan, cut it into parts with Split PDF and run OCR on each part.
Free ways to copy text from a scan
You may already have a tool that can read text from images. These are good for grabbing a sentence or a phone number; for whole documents, a proper OCR pass is more reliable.
Windows
On Windows 11, the Snipping Tool has a "Text actions" button. Take a snip of the part of the page you need, choose Text actions, and copy the recognized text. It works on anything on your screen, including a scanned PDF. Microsoft Word can also open many PDFs (File > Open) and convert them into an editable document, sometimes including scanned pages, although complex layouts may shift.
Mac
Recent versions of macOS include Live Text, which recognizes text in images. In Preview, you can often select text directly in a scanned page or image as if it were typed. If it doesn't work, take a screenshot of the area (Cmd+Shift+4), open it in Preview, and select the text there.
iPhone and iPad
Live Text works in Photos and in the camera. Point the camera at the paper or open a screenshot of the PDF page, tap the text icon, then select and copy. It's excellent for a few lines and hard to use for many pages.
Android
Google Lens, built into the camera or Photos app on most Android phones, can select text in a photo or screenshot. Choose the Text option, select what you need and tap Copy.
Any browser
Your browser's built-in PDF viewer shows the same thing your desktop reader does: if the page is an image, there's nothing to select. A browser alone can't add a text layer; you need an OCR tool.
When copying is blocked on purpose
Some PDFs contain real text but have permissions set by the author that disable copying, printing or editing. Your reader may show "(SECURED)" in the title bar or a lock icon, and the Copy command is grayed out.
These restrictions are set with an owner password. If the document is yours, or the author gave you the password, Unlock PDF removes the protection when you enter it. MyPDFDone does not crack or bypass passwords: if you don't know it, ask the sender for an unrestricted copy. Our guide on how to remove a password from a PDF explains the difference between open passwords and permission passwords in more detail.
When pasted text looks like gibberish
Sometimes the text selects and copies fine, but pastes as a string of odd symbols. That happens when a PDF uses embedded fonts without a map from shapes back to letters, which is common in some older or design-tool exports. The PDF looks right because the shapes are drawn correctly, but the computer doesn't know which letter each shape is. OCR fixes this too, because it reads the shapes on the page rather than trusting the broken font data.
After OCR: check before you rely on it
OCR is very good, but it isn't perfect. Before you paste recognized text into anything important, check:
- Numbers, amounts and dates match the original exactly (5 vs S, 0 vs O, 1 vs l are classic mix-ups).
- Names and addresses are spelled correctly.
- Tables copied in the right order; OCR reads columns as best it can, but complex tables may need tidying.
- Search finds a few words you can see on different pages.
If what you really want is a document you can rewrite, read how to convert PDF to Word and keep the formatting. And if your scan started as phone photos, how to turn photos into a PDF covers how to get clean pages that OCR reads well.
Frequently asked questions
Use your PDF reader's search (Ctrl+F or Cmd+F) for a word you can see on the page. If it finds nothing, or clicking selects the whole page like a picture, the page is an image. Slightly crooked lines and paper texture are other signs.
No. OCR PDF keeps the original page images and places the recognized text invisibly on top. You see exactly the same pages, but you can now search, select and copy.
On a clean, straight scan of printed text at 200–300 dpi, accuracy is usually very high. Handwriting, faint photocopies, tiny print, stamps over text and pages photographed at an angle cause errors. Always proofread numbers, names and amounts you copy.
English, Spanish, Portuguese, German, French, Italian, Russian and Ukrainian. By default it uses the site's language plus English, and you can pick a different one before starting.
No, that is a permission setting, not a scan problem. The author restricted copying. If you have the owner password, you can remove the restriction with Unlock PDF. If you don't, ask the sender for an unrestricted copy.
Yes. PDF to Word detects scanned pages and runs OCR automatically, giving you real, editable text in a Word file. Use it when you need to change the wording; use OCR PDF when you need the original look with searchable text.