Skip to main content

Why can’t I copy text from this PDF?

You usually cannot copy text from a PDF because the page contains an image or outlined shapes instead of extractable characters, the document’s security settings prohibit copying, or its text encoding is damaged or incomplete. Start by checking whether you can select individual words, then inspect the document’s permissions and test a short pasted sample. If you have broader trouble changing the document, see the guide to editing text in a PDF.

REAL TEXT Quarterly results 2026 ✓ Selected PICTURE OF TEXT Quarterly results 2026 ✕ No text Same look — only real text can be selected. OCR adds it.
Two lines can look identical. A drag selects real text but finds nothing on a picture of text — there are no characters until OCR adds a text layer.

The short explainer below shows why two identical-looking PDFs can behave differently when you try to copy from them, and which failures optical character recognition can repair.

Short explainer: real characters, a picture of text and outlined letters look the same on screen, but only real characters can be copied. The video also separates copy failures caused by permissions and broken character mapping from the ones OCR can fix.

A PDF can show words without containing copyable text

COPY

Characters stored as text with a font. Selectable, copyable, searchable.

Inspector — text layer: yes

Controlled synthetic test · text layer measured

Three pages can look identical yet behave differently; only real selectable text can be copied or searched.

A PDF records instructions for displaying a page. Those instructions may include real text characters, a photograph of printed text, or vector shapes that merely look like letters. All three can appear identical on screen, but only the first reliably gives a PDF viewer actual characters to place on your clipboard.

Copying also requires more than visible letters. The viewer must be allowed to extract them, determine which Unicode characters they represent and reconstruct their reading order. A failure at any of those stages changes what happens when you select, copy and paste.

Scans and screenshots contain pixels instead of characters

A scanned PDF usually contains one image for each page. The scanner photographs the paper, but it does not automatically identify the words inside the photograph. You can zoom in, print the page and read it clearly while the PDF still contains no selectable text.

The same problem occurs when someone creates a PDF from:

  • a screenshot;
  • a photograph taken with a phone;
  • scanned pages without optical character recognition;
  • image files placed into a document and exported;
  • a fax or photocopy workflow that flattened each page into an image.

The clearest symptom is that dragging across a sentence does not highlight individual words. The viewer may select the whole page as one rectangular image, or it may not show a selection at all. Search also tends to fail because the application has no characters to search.

Optical character recognition, or OCR, analyzes the shapes in the image and creates machine-readable text. A searchable-image PDF normally retains the original page image while adding an invisible text layer. That layer lets you select, search and copy words without replacing the visible scan.

OCR is recognition rather than recovery of an original document, so its output needs checking. Small type, handwriting, low contrast, skewed pages, decorative fonts and blurred scans can cause substitutions such as 0 for O, 1 for l, or rn for m. Tables and multi-column layouts can also produce an incorrect copy order even when most characters are recognized correctly.

Outlined letters are shapes rather than live text

Design and publishing software can convert type into outlines. This changes each letter from a text character into one or more vector paths. The result can remain perfectly sharp at any zoom level, which makes an outlined page look more like normal digital text than a scan does.

However, the visible shape of an A does not by itself tell a PDF viewer that the shape represents the Unicode character A. It may simply be a collection of lines and curves. Selection may capture an entire graphic object, skip the letters or behave as though the page were blank.

Outlined text is common in logos, illustrations, technical drawings and print-production files. It can also affect a whole page if the creator intentionally flattened or outlined all type before exporting the PDF. Converting text to outlines can preserve a particular appearance or remove a font dependency, but it also discards the editable character data unless a separate text layer is retained.

OCR may recover usable words from outlined letters because it can analyze their visible shapes in much the same way as a scan. The recognition quality still depends on the letterforms, spacing and page layout.

Security settings or damaged text data can block extraction

A PDF may contain genuine selectable text while its author’s security settings disallow copying. In that case, you might be able to highlight words but find that the Copy command is unavailable, has no effect or produces an empty clipboard. The document’s restrictions may separately control copying, editing, printing and other operations.

A different failure occurs when the PDF contains text objects but does not provide enough information to map its stored character codes to standard Unicode characters. The page can display correctly because the embedded font knows which glyph to draw, yet another application may not know that a particular glyph means A, é or 你.

This mapping problem produces symptoms such as:

  • pasted text becoming boxes, symbols or unrelated letters;
  • correct letters appearing in the wrong order;
  • spaces disappearing or appearing inside words;
  • line breaks being inserted after every fragment;
  • columns being mixed together;
  • only certain fonts or sections copying incorrectly;
  • search finding some words but not others.

The PDF may also store text in the order it was drawn rather than the order in which a person reads it. A visually normal table, multi-column page or positioned graphic can therefore paste as a scrambled sequence. Document structure tags and a sound reading order can help applications interpret complex layouts, but not every PDF includes them.

Selection, permissions and a paste test reveal the cause

Do not judge copyability from appearance alone. Run a few small tests in order, using a short and distinctive sentence rather than selecting the entire page.

Test result Most likely explanation Best next step
Individual words highlight and paste correctly The PDF contains extractable text Copy normally or export the document if you need a large amount
The whole page highlights as one object The page is probably a scan or flattened image Run OCR
Nothing highlights over the visible letters The letters may be pixels, vector outlines or inaccessible page content Test search, then use OCR or obtain the source file
Text highlights, but Copy is unavailable Document permissions probably prohibit copying Check the security summary and ask the owner for an authorized copy
Text highlights, but pasted characters are wrong The character-to-Unicode mapping is missing or damaged Re-create, convert or OCR a copy
Pasted words are correct but arrive in the wrong order The content order or document structure is poor Export to an editable format, repair the reading order or copy smaller sections
Copying works in a desktop PDF app but not in the browser The browser viewer may be limiting the operation Download the file and open it in a full PDF application

First, switch to the viewer’s text-selection tool and drag through one line. A normal text selection usually follows letters or words. A rectangular selection around the page suggests that you selected an image rather than text.

Next, search for an unusual word that is clearly visible. If the viewer reports no results, the page probably lacks recognized text. Search is not a perfect test because bad encoding can also prevent matching, but a total failure across several visible words strongly points to a scan, outlines or a defective text layer.

If you can select words but cannot copy them, check the file’s properties or document information. In Adobe Acrobat Reader, the Security tab includes a Document Restrictions Summary. Look for the content-copying permission rather than assuming that an unavailable Copy command is a temporary application fault.

Then paste one sentence into a plain-text editor. Plain text removes most formatting variables and shows what characters actually reached the clipboard. Compare the pasted result with the PDF:

  • Random symbols or incorrect letters indicate a character-mapping problem.
  • Correct words in a strange sequence indicate a reading-order problem.
  • A few recognition mistakes on an image-based page indicate imperfect OCR.
  • A completely empty paste after a valid selection may indicate a restriction or viewer problem.

Also compare more than one part of the file. A PDF can contain mixed page types: its first pages may have live text, while later pages are scanned attachments. A signature page, inserted exhibit or old photocopy can require OCR even when the rest of the document copies normally. One paragraph may also use a damaged embedded font while other paragraphs use a font with valid character mappings.

Finally, try the downloaded file in a dedicated PDF application if you are currently viewing it in a browser, email preview or document-management portal. If the same text copies correctly elsewhere, the PDF itself probably contains usable text and the first viewer is the immediate problem. If every viewer produces the same failure, the file’s content or permissions are the more likely cause.

Match the repair to the type of copy failure

The right fix depends on what the tests revealed. OCR is appropriate for image-based or outlined pages, but it does not legitimately remove an author’s security restrictions, and it may not be the cleanest repair for a badly generated digital PDF.

Use OCR for scans, screenshots and outlined text

Apply OCR when words are visible but cannot be selected as individual characters. Choose the correct document language before recognition because language settings help the OCR engine distinguish similar shapes and apply the right spelling patterns.

A practical OCR workflow is:

  1. Save a backup of the original PDF.
  2. Select the pages that lack copyable text.
  3. Choose the correct recognition language.
  4. Run OCR to create searchable and selectable text.
  5. Search for several names, numbers and uncommon words.
  6. Copy sample passages into a plain-text editor.
  7. Correct important recognition errors before relying on the result.

Check headings, footnotes, tables, accented characters and small print rather than testing only a clean body paragraph. If the file contains several languages, recognition may need different settings for different page ranges.

OCR can also create a usable text layer over outlined vector letters. However, if you own the original design or word-processing file, exporting a new PDF with live text is usually better. Re-exporting preserves the actual characters and avoids recognition errors that OCR may introduce.

Obtain an authorized copy when permissions prohibit copying

If the document restrictions state that content copying is not allowed, ask the owner or sender for a version that permits extraction. The author may be able to change the security settings, provide the required owner password, send the original editable document or export the passage for you.

A password that opens a PDF is not necessarily the password that controls its permissions. PDF security can distinguish between opening the file and changing restrictions. If you are authorized to alter the document, use the appropriate owner credentials and save a permitted copy through software that supports those settings.

If copying is allowed according to the security summary but the command still fails, save the PDF locally and test it in another current viewer. This separates a document restriction from a browser, clipboard or application problem.

Re-create or convert PDFs with broken text mappings

When text selects but pastes as gibberish, OCR is one possible repair, but first look for a cleaner source. A new export from the original word-processing, spreadsheet, presentation or publishing file can restore proper character mappings while preserving better layout and accessibility than OCR.

If the source file is unavailable, try converting a copy of the PDF to an editable format. Conversion software may reconstruct paragraphs, columns and tables more successfully than direct clipboard copying. Inspect the converted result carefully because the same damaged mappings or content-order problems can carry into the new file.

For a PDF that you created, return to the source application and:

  • keep body text as live text rather than converting it to outlines;
  • embed or subset fonts using normal PDF export settings;
  • avoid printing through workflows that flatten pages into images;
  • export a tagged PDF when the application supports it;
  • verify the reading order of columns, tables and positioned text;
  • open the exported PDF and test search, selection and paste before distributing it.

If only one passage uses a defective font, replacing or re-entering that passage in the source document may be enough. If every page has unreliable mappings, rebuilding the PDF or applying OCR to a duplicate may be faster than repairing individual objects.

Preserve the original throughout the process. OCR, conversion and re-export can change compression, hyperlinks, form fields, signatures, bookmarks or layout. Work on a copy, compare the repaired file with the original and confirm that the text you need now selects, searches and pastes in the correct order.

Source: Scanned-document recognition and searchable text guidance from Recognize text in scanned documents, copying permissions and restriction checks from Copy content from PDFs, text extraction, Unicode mapping and reading-order guidance from Accessibility features in PDFs, character-mapping requirements from the PDF Reference, version 1.6, and outlined-text behavior from Create paths from text outlines, Adobe. Page last checked 2 October 2026.