Split PDF by Content – Keyword & Pattern PDF Extractor

Extract specific pages from a PDF by searching for keywords, regex patterns, or text markers. Get only the pages that match your content rules perfect for pulling invoices, sections, or clauses out of large documents. 100% free, no upload, browser-based.

Keyword Search Regex Matching Pattern Markers 100% Private

This is a CONTENT-BASED PDF splitter, not a page-range splitter

Our tool splits PDFs by searching for keywords, regex patterns, or text markers. It does not split by page number, page range, or "every N pages." You define content rules, and we extract the pages that match.

If you need page-range splitting (e.g., split a 100-page PDF into 100 individual files, or extract pages 5-10), use iLovePDF, Smallpdf, or Adobe Acrobat instead. See the "Page Range Splitting" section below for alternatives.

Content-Based Splitting vs Page-Range Splitting

There are two very different ways to split a PDF. Understanding the difference is key to picking the right tool:

Content-Based Splitting (This Tool)

"Find all pages containing 'Invoice #' and save them as a new PDF."

Best for: Extracting invoices from a batch PDF, pulling specific contract clauses, separating report sections.

Not ideal for: Splitting a 100-page PDF into 100 individual files, extracting pages by number.

Page-Range Splitting (Other Tools)

"Split pages 10-20 into a new PDF" or "Split every page into a separate file."

Best for: Dividing a book into chapters, breaking up scan batches, extracting specific page numbers.

Not ideal for: Finding pages by content. You need to know the page numbers in advance.

Both approaches are useful they just solve different problems. This page is about content-based splitting. If you need page-range splitting, see the alternatives section below.

Key Features

Keyword-Based Splitting

Add multiple keywords and find pages containing any of them. Each matched keyword is shown as a badge so you know exactly which pages triggered.

Regex Pattern Matching

Use JavaScript regular expressions for advanced matching emails, phone numbers, dates, invoice numbers, or custom formats. Live regex preview.

Pattern-Based Extraction

Define start/end or start/middle/end text markers to extract pages between specific text boundaries. Perfect for structured documents.

Multi-PDF Batch Search

Upload up to 5 PDFs and search across all of them simultaneously. Find matching pages across your entire document collection.

Visual Page Preview

See exactly which pages matched with keyword-specific badges. Full transparency over what content triggered each match.

100% Private

All OCR and PDF processing happens locally in your browser. Files are never uploaded to any server. Complete privacy guaranteed.

How to Split a PDF by Content

Upload PDFs

Upload up to 5 PDF files. Each file can be up to 25MB. Scanned PDFs are supported via automatic OCR.

Extract Text (OCR)

Click "Extract All" to run OCR on every page. Text is extracted and cached locally for fast searching.

Define Split Rules

Choose Keywords, Regex, or Pattern mode. Add your search terms the live preview updates in real time.

Split & Download

Click "Find Matching Pages" to search all PDFs. Review the matches, then download split PDFs.

Three Split Modes Explained

Keywords Mode

Use when: You know simple words that identify the pages you want.

Examples: "Invoice", "Summary", "Chapter 5", "Contract", "Patient Name".

Add multiple keywords a page matches if any keyword is present. Each matched keyword is shown as a colored badge.

Regex Mode

Use when: You need to match a pattern, not just a word.

Examples: /\d{3}-\d{2}-\d{4}/ for SSNs, /[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}/i for emails.

JavaScript regex syntax with optional flags (i, g, m). Live preview shows your regex as you build it.

Pattern Mode

Use when: The content you want sits between two text markers.

Examples: "Between Invoice # and Date:" or "Starts with Section, contains Conclusion, ends with References".

Best for structured documents with consistent formatting contracts, forms, reports.

Need Page-Range Splitting Instead?

If you want to split a PDF by page number or range (not content), here are your options:

iLovePDF Split PDF

Upload your PDF, choose "Split PDF," and select a page range, or split every page into a separate file. Free with limits. Files uploaded to iLovePDF servers.

Smallpdf Split PDF

Same feature set as iLovePDF split by range, split every page, split into two. Free with limits. Files uploaded to Smallpdf servers.

Adobe Acrobat Pro (Paid)

Tools → Organize Pages → Split. Adobe Acrobat Pro DC has robust page-range splitting. Requires a paid subscription ($12.99–$24.99/month).

macOS Preview (Free on Mac)

Open the PDF in Preview, drag pages from the sidebar into a new document, save. Free on any Mac, no upload needed. Good for extracting a few pages manually.

Why our tool doesn't do page-range splitting: Our splitter is designed to solve a different problem finding pages by what they contain, not by their position in the file. If you want standard page-range splitting, the tools above are better suited to that task.

What You Can Do With Content-Based Splitting

Extract All Invoice Pages

A PDF contains 50 invoices mixed together. Search for "Invoice #" and extract every invoice page into a new PDF. Set up once, reuse forever.

Pull Report Chapters

Extract only the "Executive Summary" and "Financial Results" sections from a large annual report. Use pattern mode with start/end markers.

Find Confidential Pages

Search for "Confidential", "NDA", or "Privileged" to extract every sensitive page for separate review or redaction.

Academic Research Extraction

Extract "Abstract" and "Conclusion" sections from a folder of research papers. Perfect for literature reviews and meta-analyses.

Extract Contract Clauses

Pull "Termination", "Liability", or "Indemnification" clauses from a large contract PDF. Regex search for specific dates, names, or amounts.

Separate Bank Statements

Search for "Statement Period" or "Account #" to extract each monthly statement from a combined PDF into individual files.

Why Choose Our Content-Based Splitter?

No Registration Required - Start splitting immediately
3 Split Modes - Keywords, Regex, and Pattern-based splitting
Live Regex Preview - See your regex build in real-time
Offline Caching - Processed PDFs are cached for instant re-searching
100% Free - No hidden fees, no watermarks, no page limits

Keyword-Specific Badges - See exactly which keywords matched each page
Multi-PDF Search - Search across all uploaded PDFs at once
100% Browser-Based - Files never leave your device
Smart File Naming - Split PDFs include original name + page count
Works on Any Device - Desktop, tablet, and mobile supported

How Content-Based PDF Splitting Works

Traditional PDF splitters only extract pages by number. Our tool reads the content inside your PDFs and extracts pages based on what they actually say. Here's the technology:

OCR Text Extraction

We use Tesseract.js and PDF.js to extract text from every page including scanned PDFs and image-based documents. Text is cached locally in IndexedDB for fast re-searching.

Intelligent Pattern Matching

Your keywords, regex, or pattern rules are applied against the extracted text. Our engine returns exact page-level matches with keyword badges showing what triggered each result.

Page-Level Extraction

Using pdf-lib, we rebuild a fresh PDF containing only the matched pages. Original formatting, fonts, and images are preserved.

Zero Server Uploads

Everything runs in your browser. Your PDFs are never sent to any server. This makes our tool safe for confidential documents like contracts, medical records, and financial statements.

Adobe Acrobat Split PDF vs OnlineOCR.io

Adobe Acrobat Pro DC has a "Split Document" feature (Tools → Organize Pages → Split) that splits by page number, page count, or file size. It's excellent for page-range splitting but has no content-based search.

Our tool does the opposite: it's designed for content-based splitting. You define keywords, regex, or patterns, and we extract matching pages. Both approaches are useful they just solve different problems.

Adobe Reader cannot split PDFs it's a viewer only. To split PDFs by page range, you need Adobe Acrobat Pro (paid) or a free online tool like iLovePDF or Smallpdf. To split by content, use our tool above.

Frequently Asked Questions

No. This tool splits PDFs by content keywords, regex patterns, or text markers. It does not split by page number, page range, or "every N pages." If you need page-range splitting, use iLovePDF, Smallpdf, or Adobe Acrobat Pro. See the "Page Range Splitting Alternatives" section above.

First, OCR extracts text from all PDF pages. Then you define split rules using keywords, regex, or patterns. The tool searches all pages and shows which pages match your rules. You can then download matching pages as a new PDF with one click.

Keywords Mode: Add multiple keywords and find pages containing any of them. Each matched keyword is shown as a badge.

Regex Mode: Use regular expressions for advanced pattern matching like emails, phone numbers, or custom formats.

Pattern Mode: Define start/end markers to find pages between specific text boundaries.

Yes! All PDF processing and splitting happens locally in your browser using PDF.js, Tesseract.js, and pdf-lib. Your files are never uploaded to any server. The OCR text is cached locally in IndexedDB for performance. Google Drive sync is optional and requires your explicit permission.

Yes! Upload up to 5 PDFs (up to 10 for registered users) and the splitter searches all documents simultaneously. Results show which PDFs and which pages matched your rules, with individual download buttons for each file.

The results panel shows detailed information for each matched PDF. You'll see colored badges showing exactly which keywords matched, and on which specific pages. This gives you complete transparency over the splitting process.

You can use standard JavaScript regular expressions with optional flags (i, g, m). Examples: emails (/\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}\b/gi), phone numbers, dates, SSNs, invoice numbers, or any custom pattern. The live preview shows your regex as you build it.

Split PDFs are named using the format: onlineocr.io-splitPdf-[page count]pages-[original name].pdf. This makes it easy to identify the source document and how many pages were extracted.

Yes! Our PDF Splitter is fully responsive and works on all devices desktop, tablet, and mobile. The interface adapts to your screen size. OCR processing is CPU-intensive, so desktop is recommended for large files.

Yes! Our tool runs OCR on every uploaded PDF, including scanned image-based documents. Even if your PDF has no selectable text layer, we'll extract the text so you can search and split by content. OCR supports 100+ languages.

Password-protected PDFs must be unlocked before uploading, since our browser-based OCR cannot process encrypted files. Once you remove the password, you can upload and split normally. All processing remains local and secure.

Our splitter is content-based it extracts pages by what they contain, not by page number. For page-range splitting, use iLovePDF, Smallpdf, or Adobe Acrobat Pro. Each supports split by page range, page count, or every N pages.

Yes, 100% free. No hidden fees, no page limits, no watermarks on output, and no credit card required. You don't even need to create an account.

Yes! Our tool works in Safari, Chrome, Firefox, and Edge on any macOS version (Intel or Apple Silicon). For page-range splitting on Mac, use the built-in Preview app (drag pages into a new document) or a free online tool. For content-based splitting, use our tool above.

Our splitter extracts matching pages into a new PDF. To merge PDFs, use our Merge PDF tool. Workflow: split first (extract the pages you want), then merge those results with other PDFs if needed.

Related PDF Tools

Ratings & Reviews

See what our users say about our tools. Your feedback helps us improve.

Drag & Drop PDF Files Here

Create Extraction Rule

Select OCR Language

Select Pages to Extract

Settings

Extracted Text —