How to Copy Text from a PDF That Won't Let You Select It

Not all PDFs contain a text layer. Scanned documents are images, the PDF is just a wrapper around a photo of a page, and there is no underlying text to select. PDFs with permission restrictions may also have text copying disabled by the creator. And some PDFs use non-standard font encoding that makes copy-pasted text appear as garbled characters.

Some PDFs use Type 3 fonts, custom bitmap fonts where each character is a small embedded image. Text in Type 3 fonts is technically present as text objects in the PDF, but the glyphs are images rather than standard Unicode characters. Extraction from these PDFs produces garbled or empty output because there is no Unicode mapping for the image-based glyphs. OCR on a rendered version of the page is the only reliable extraction method for these documents.

For research workflows where extracted text will be fed into analysis tools, run the extracted output through a basic cleaning step to remove common PDF artifacts before processing. Header and footer text that repeats on every page, page number tokens, and hyphenation breaks at line ends all appear in the extracted text and interfere with word frequency analysis, similarity scoring, and other text-based operations.

Right-to-left languages such as Arabic and Hebrew may extract in the wrong order when copied from a PDF, because the logical order in the file may not match the visual rendering order. PDF extraction tools that handle bidirectional text correctly will reverse the order automatically, but tools designed primarily for Latin script may require manual correction of extracted right-to-left passages.

Text extraction pulls the embedded text layer from a PDF that was digitally created, reports, exported Word documents, generated invoices. It is fast and accurate because the text is already there in machine-readable form. OCR is needed for scanned PDFs and image-based PDFs where no text layer exists. Use the Extract Text tool for digital PDFs and the Image to Text tool for scanned documents.

Scanned PDFs that have had OCR applied by a scanner or document management system contain a hidden text layer beneath the visible page image. Text extraction from these PDFs works and produces the OCR result, but the accuracy of that result depends on how well the original OCR was applied. Poor-quality scanner OCR produces poor-quality extracted text, re-running OCR on the page image with a better engine may give more accurate results.

Extracted text from financial PDFs, such as bank statements or invoices, often contains numbers split across lines or columns that appear contiguous in the visual layout but are separated by whitespace in the text layer. Before using extracted numbers for calculations or imports, verify that multi-digit amounts have not been split or merged incorrectly. Manual spot-checking of extracted financial data is worth the time.

Upload your PDF to the Extract PDF Text tool. The tool uses the same PDF rendering engine that powers Firefox to read the text layer from each page and compile it into a single output. The extracted text appears in a text area. Use the Copy button to copy everything to your clipboard at once, then paste wherever you need it.

If you need to extract text from only specific pages, not the entire document, split those pages out using the Split PDF tool first, then run text extraction on the resulting smaller document. This produces a focused extracted text output without the noise of unrelated pages, which is particularly useful for long documents with a mix of text and image-only pages.

Extracted text is plain text, no formatting, no columns, no fonts. Paragraphs appear as blocks of text separated by line breaks. Multi-column layouts may have their columns interleaved in reading order, which can require manual cleanup. Tables lose their grid structure and become rows of text. For purposes like research quotation, keyword searching, or feeding text into another tool, the plain output is usually sufficient.

For documents with multiple columns, the reading order of extracted text may not match the visual layout. PDF text objects are stored in the order they were placed on the page, which in a two-column layout is often column 1 followed by column 2 across the full page, rather than reading left-to-right within each line. Manual cleanup of column order is sometimes necessary for two-column academic papers and magazine-style layouts.

If the extracted text is empty or garbled, the PDF is likely scanned or uses non-standard encoding. Try the Image to Text (OCR) tool instead for scanned content. For garbled encoding, open the PDF in a browser, select and copy text manually to see if the browser renderer can decode it correctly. If browser copy-paste also produces garbage, the font encoding is non-standard and OCR is the only reliable option.

If you need to monitor whether a PDF's text content has changed between versions, for legal review, version control, or content auditing, extracting the text from each version and running a diff comparison is a reliable approach. Plain text comparison ignores formatting changes and highlights only content-level differences, making it easier to identify substantive changes in lengthy contract or policy documents.