What does "extracting text from a PDF" mean?

Last reviewed: October 22, 2026 · Markdown version

It means taking the characters a PDF has stored inside it and writing them out as ordinary text you can paste, edit or feed to another program, without the page layout around them. It only works on a PDF that holds its words as text; a scanned page has nothing to extract until OCR has run. The confusing part is that PDF tools use "extract" for at least two other jobs, pulling out whole pages and converting to Word, and a document's security settings treat those separately. Here is what each one means, in the words of the tools that use it.

When a tool says	What you get	What it needs
Extract text	Plain text, in reading order, no formatting	A PDF with a text layer
Extract pages	A new PDF holding some of the original's pages	Page extraction not restricted
Convert or export to Word	An editable document that tries to keep the layout	A converter, and patience with tables

Text extraction: the words without the page

The plainest definition comes from pdftotext, a free command-line tool whose manual says it "converts Portable Document Format (PDF) files to plain text". The same manual shows what gets lost on the way. By default the tool undoes the physical layout, meaning columns and hyphenation, and will "output the text in reading order"; its -layout option is there to "Maintain (as best as possible) the original physical layout of the text" (pdftotext(1) manual page, Debian Manpages, read 22 October 2026). That "as best as possible" is the honest summary of the whole job. A PDF records where each piece of text sits on the page; it does not always record which column a line belongs to or where a table cell ends, so extracted text can come out in a surprising order.

Selecting a paragraph and tapping Copy is text extraction too, just by hand and one piece at a time. If nothing on the page will select, the file is probably a picture of text, and checking for a text layer settles it before you try another tool.

Page extraction: a new PDF, not text

In Adobe Acrobat, "extract" usually means pages. Adobe's help says that "Extracting pages allows you to reuse selected content from one PDF in a new document", and that "The extracted pages include all content, form fields, comments, and links from the original." You choose the pages under All tools, Organize pages, then Extract pages, and you can keep or remove them from the source; Adobe notes that "Any bookmarks or article threads associated with pages aren’t extracted." Its first step is a check: "Go to File > Document Properties > Security tab to view any restrictions" (Extract pages from PDFs in Acrobat on desktop, Adobe Acrobat Help, read 22 October 2026). Nothing in this job turns the words into text. A scanned page extracted this way is still a scanned page.

Why a PDF can allow one and refuse the other

The author of a protected PDF sets the two permissions in different places. On Adobe's page for restricting a document, the Changes Allowed menu includes an option called "Any except extracting pages", while copying sits in a separate checkbox, "Enable copying of text, images, and other content" (Restrict printing, editing, and copying PDFs, Adobe Acrobat Help, read 22 October 2026). So a file can let you copy its text and still refuse to give up its pages, or the other way round. Adobe Reader's help gives the reader's side: "You can copy content from a PDF unless the author has applied security settings that disallow copying", and you confirm it by right-clicking, choosing Document Properties, then the Security tab, where the Document Restrictions Summary is listed (Copy content from PDFs, Adobe Acrobat Reader Help, read 22 October 2026). What to do when copying is blocked is on why you cannot copy text from a PDF.

The same Reader page has a fourth meaning hiding in it. Its Snapshot tool "copies an area as an image that you can paste into other applications", which takes a picture of the words rather than the words themselves (the Copy content from PDFs page, read 22 October 2026).

Conversion: extraction with the layout attempted

Turning a PDF into a Word or Google Docs file is a step beyond extraction, because the converter tries to rebuild the formatting as well. Google's help for its own converter sets expectations by listing what survives: "Bold, italics, font size, font type, and line breaks are likely to be retained", while some structure does not, and it asks for files "2 MB or smaller" (Convert PDF and photo files to text, Google Drive Help, read 22 October 2026). If you only need the words, plain extraction is the simpler and more predictable choice; the steps for each route are on extracting text from a PDF.

When what you want is the pages that mention something

Sometimes the real goal is neither all the text nor chosen page numbers, but the pages where a word appears. DocFind does that on iPhone and Android: it searches inside many PDFs at once, and it can export results as a CSV spreadsheet, a Markdown file, a PDF report, or a PDF of just the pages that matched. Every export records what was searched and anything that was skipped. Scanned PDFs are read on your device first, for printed text, not handwriting. It does not edit, annotate or sign PDFs; it never changes the files you search, so an export is a new file beside the original.