About the Extract PDF Pages by Keyword
In a long report, bank statement or manual, you often need only the pages that mention a certain thing: a name, an invoice number, a product. Searching is easy, but making a new file of just those pages is not. This tool reads the text of every page, finds the ones that match your words and builds a new PDF from them, in their original order. Type one or more words separated by commas, and choose whether a page must contain any of them, all of them, or the exact phrase. You can match case and whole words only. You see the list of matching pages before you create the file. Everything happens in your browser, so private documents stay private.
Updated October 2026
How to use the Extract PDF Pages by Keyword
- 1Pick your PDF.
- 2Type the words to look for, separated by commas or new lines.
- 3Choose any, all or exact phrase, and the case and whole-word options.
- 4Press Find pages and check the list.
- 5Press Make a PDF of these pages and download it.
Good ways to use it
Pick a distinct word to avoid too many matches. An invoice number or a surname finds exactly the right pages, while a common word like total finds almost every page. If you need both a name and a date on the same page, use All mode with the two terms.
Check the page list before you build the file. If the pages look wrong, try turning on Whole words only, or the exact phrase mode, to get a tighter match.
Think about spelling variants before you search. A name may appear as Smith, SMITH or Smith, J., and a product code may include a dash in some places and not in others. Using several terms in Any mode, with case matching off, catches most of these. If you only want pages that carry two facts at once, use All mode and pick words that rarely appear alone.
Examples
- Statements: pull out the pages with one account number.
- Contracts: collect every page that mentions termination.
- Manuals: extract the pages about one product code.
- Meeting packs: keep only the pages with your name.
- Research: gather the pages that cite one author.
Related tools
To simply search without creating a file, use Search in PDFs. To pick pages by number, use Split PDF, and to drop unwanted pages choose Delete PDF Pages. Make scanned files searchable with PDF OCR, and count the words in the result with the PDF Word Counter. Always keep the original PDF unchanged.
Features
- Several words at once
- Any, all or exact phrase matching
- Match case and whole words options
- Shows which pages match before you build
- Keeps original page order and quality
- Works on text PDFs, offline in your browser
Frequently asked questions
How does the keyword matching work?
The text of each page is read and compared with your words. In Any mode the page needs at least one of the words, in All mode every word, and in Exact phrase mode the words in that order.
Can it find words in a scanned PDF?
Only if the scan already has a text layer. Otherwise the tool says no text was found. Run PDF OCR first to add one.
Are the matched pages changed?
No. They are copied unchanged, with the same layout, fonts and images, into a new file.
What does whole words only do?
With it on, cat finds the word cat but not concatenate or category.
Can I use wildcards or regular expressions?
Not in this version. Words are matched as typed. Use several words in Any mode to cover variants.
Is there a limit on the number of pages?
No fixed limit. A very long PDF takes longer to read.
Will it find text split across two lines?
Line breaks are treated as spaces, so a phrase that wraps to the next line is still found.