✂️ Split PDF by Text Online Free — No Upload
Split a PDF by text in your browser: keep only the pages containing a word, phrase or invoice number — perfect for long reports, contracts and manuals in the EU and US — free, no signup, nothing uploaded.
🔒 Your data never leaves this page
- The PDF is searched and rebuilt 100% in your browser — nothing is uploaded to any server.
- No account, no email capture, no tracking of your documents.
- Contracts, reports and client files stay on your device by design. Why that matters →
1. Your search phrase
2. Choose your PDF
Split a PDF by text in 3 steps
- Type your phrase. A client name, invoice number, chapter title or keyword — and choose whether to keep matching pages or everything except them.
- Choose the PDF. Every page's text layer is searched in your browser; matched page numbers are reported before download.
- Download the new PDF. Only the selected pages, in original order, as a separate file. Your original is untouched.
Why searching beats scrolling
PDF became an ISO standard in 2008 (ISO 32000-1:2008), and the format's text layer is what makes this tool possible: every character in a text-based PDF carries its position, so a browser can search hundreds of pages in seconds. Manually hunting a single clause through a 200-page contract, or one invoice inside a year of statements, is exactly the chore this tool removes — and it does it without sending the document anywhere.
Where text-splitting saves the day
🧾 One invoice from a year's statements
Search the invoice number and extract just those pages for your accountant — instead of emailing the whole year. Pair with PDF to CSV for accounting when the figures also need to land in a spreadsheet.
📜 One clause from a long contract
Search the clause heading or a distinctive term to pull the relevant pages for a lawyer or client, without forwarding the entire agreement.
📚 One chapter from a manual
Training handouts, exam excerpts and onboarding packs all start as "just chapter 4" — search the chapter title and the handout builds itself.
🧹 Strip the boilerplate
Use "pages NOT containing the phrase" to drop repeated cover sheets, disclaimers or terms pages from a bundle before sharing it — then tidy the order with reorder pages if needed.
Straight talk: limits
🖨️ Scanned PDFs have no text to search
Image-only PDFs cannot be searched by text, and the tool reports that instead of returning an empty file. Use a text-based export or an OCR step first.
🔤 Reading order matters
Phrases match when the words are consecutive in the PDF's reading order. Multi-column layouts can interleave text, so a phrase split across columns may not match — try a shorter, single-column phrase in that case.
🔒 Password-protected PDFs are not supported
Encrypted PDFs cannot be read in-browser here. Use an unlocked export with the document owner's permission.
Split PDF by Text FAQ
How is splitting by text different from splitting by page range?
Page-range splitting needs you to already know which pages to keep. Text splitting finds them for you: type a phrase like an invoice number, a chapter title or a client name, and every page containing it lands in the new PDF — even when the matches are scattered through a 200-page file.
Does this work on scanned PDFs?
No — scanned image-only PDFs have no text layer to search, and the tool says so plainly instead of returning an empty file. It works on text-based PDFs: exports, reports, contracts and e-books with selectable text.
Can I keep pages that do NOT contain the phrase?
Yes — switch the mode to 'pages NOT containing the phrase'. That is the fast way to strip cover sheets, disclaimers or a repeated appendix from a document.
Is the search case-sensitive?
By default it is not: 'Invoice', 'invoice' and 'INVOICE' all match. Tick the case-sensitive box only when you need exact-case matching, for example for codes where case carries meaning.
Are my documents uploaded anywhere?
No. The PDF is read with pdf.js and rebuilt with pdf-lib entirely in your browser. Your document never touches a server, which matters for contracts, reports and client files.
What happens to the original PDF?
Nothing — the original stays untouched on your device. The tool builds a separate new PDF containing only the matched pages, so you can always go back to the full document.
Can I search for a phrase spanning two lines?
Phrases are matched against the page's text with line breaks treated as spaces, so a phrase broken across two lines in the layout still matches as long as the words are consecutive in reading order.
