How to Extract Pages From a PDF — Save Specific Pages as a New File

You rarely need all one hundred pages of a PDF. You need the purchase agreement on pages twelve through twenty. You need the figure on page seven. You need the appendix that starts at page sixty. Sending the whole document makes the reader do unnecessary work, and in many cases exposes information that was never meant to reach them.

Extracting pages is the right answer. Done well, it produces a small, focused PDF that answers the reader's question and nothing else. Done badly, it introduces broken references, cut-off sentences, and filenames that nobody understands. This guide walks through how to do it well.

When extracting is the right move

Extracting is better than splitting or forwarding the whole file when:

• Only one section is relevant. A contract with twenty appendices where the reader needs one.
• Privacy matters. A report where only the non-confidential sections should be shared.
• Email size is a constraint. A small, extracted PDF slips under attachment limits where the full document would bounce.
• You want to reuse a section. A chart from one report can become a standalone handout.
• The reader only has a few minutes. A three-page summary is more likely to be read than a forty-page deep-dive.

If the document is already short, extraction just adds steps. Send the whole file.

The difference between extract and split

The two operations overlap but are not the same.

Splitting divides a PDF into several new files at boundaries you define. You usually end up with multiple outputs.

Extracting pulls specific pages out and saves them as a single new file. You usually end up with one output.

PDFWhirl's Split PDF tool supports both patterns. You can specify ranges like 10-20 to extract, or you can split every N pages to create multiple outputs. Pick the pattern that matches the job.

Before you extract: plan the page list

Spend thirty seconds planning.

Open the PDF and identify the pages you want. Note the actual PDF page numbers, not the printed page numbers. Some documents have a cover page or table of contents that offsets the numbering.

Check for continuity. If you want pages 12 through 20, does page 20 end at a natural stopping point? Or is there a paragraph that continues to page 21? If it does, extend the range.

Check for context. Does the extracted section rely on terminology or references introduced earlier? If yes, include a page or two of context or add a short cover page after extraction.

Decide whether you want one file or several. If you are pulling out three non-adjacent sections, should they live together in one file, or do they serve different purposes and belong in separate files?

Step-by-step: extracting pages

1. Open the Split PDF tool.
2. Upload the source PDF.
3. Wait for the page thumbnails to appear.
4. Choose range mode and enter the page numbers you want — for example, 12-20. For multiple non-continuous ranges, enter them separated by commas: 1, 5-7, 15.
5. Start the extraction. The tool produces one new PDF containing exactly those pages in the order you specified.
6. Download the new file.
7. Rename it with a descriptive filename: "Contract-Appendix-B.pdf" reads better than "split-output-1.pdf".
8. Open the extracted file and scroll through it to confirm every page is there, in the right order, with no cut-offs.

The source file is never modified. You always have the original to extract from again.

Non-continuous extraction

Sometimes the pages you want are scattered. Page 2 is the summary. Page 7 is the chart. Page 24 is the conclusion. You want all three in one file, with nothing in between.

Enter the pages as a comma-separated list: 2, 7, 24. The tool pulls those three pages and combines them in the order you specified. The resulting PDF has three pages, each meaningful on its own, without the twenty-one pages of filler.

This is the fastest way to turn a long report into a briefing document.

Sanity checks on the extracted file

Open the extracted PDF and check four things.

First page makes sense on its own. If the reader would need more context, go back and include an earlier page, or write a short cover page that frames the section.

Last page ends cleanly. No cut-off sentences, no hanging references, no signatures separated from their clauses.

Page numbers. The extracted file starts at page 1 internally, but the visible page numbers printed on each page will still show the original numbers (for example, 12, 13, 14). This can confuse readers. If that matters, consider regenerating pagination in the new file.

Cross-references. A sentence that says "see Section 4 on page 28" will look broken if page 28 is not in your extract. Either include the referenced page or add a footnote explaining the reference.

When extraction is the wrong tool

There are cases where extraction does not help.

Editable content. If you want to change text in the extracted pages, you need to convert to Word first. Extraction produces a PDF, not an editable document. The PDF to Word guide covers how to move to an editable format without losing layout.

Legal documents with clauses that reference other sections. Pulling one clause out of a contract may misrepresent its context. Either include the full context or add a note in writing that explains which sections are omitted.

Reports where every section depends on every other section. Some documents are meant to be read end to end. Extracting does not produce a cleaner version — it produces a misleading version.

Combining extraction with other workflows

Extraction often sits inside a larger flow. A few common combinations:

Extract, then merge. Pull the relevant section from several different source documents with Split, then combine them into one focused packet with Merge.

Extract, then compress. A small extracted file is already small, but if the source had large images, compression can shrink it further for email.

Extract, then rotate. If the source had mixed orientation and only some pages were relevant, rotate with Rotate PDF after extracting.

Extract, then convert. If the reader needs the extracted pages in Word, extract first (to keep the PDF small) and then convert the small extract.

Common extraction mistakes

Using printed page numbers instead of PDF page numbers. Always confirm the extraction with the PDF reader's built-in counter.

Forgetting to update the filename. A file called "split-output.pdf" in a shared drive helps nobody.

Extracting a one-page file when a three-page context is needed. One page is clean; three pages may be clearer.

Sharing the extracted file without checking cross-references. Extracted documents can contain references that no longer make sense.

Extracting to avoid dealing with redaction. If sensitive information is on the pages you are pulling out, extraction does not protect it. Use actual redaction for confidential content.

A few realistic scenarios

Client asks for the deliverable summary. Your final report is fifty pages. The summary lives on pages three through six. Extract those four pages, label the file with the project name and "Summary", and send.

Lender asks for your tax return's first page. Open the PDF, extract page 1 as a one-page file, rename it "Tax-Return-2025-Page-1.pdf", and attach.

Submitting a single chapter of your thesis to a conference. Extract the chapter, add a cover page with the chapter title, convert it to a standalone PDF, and check it reads end to end.

Forwarding a specific invoice from a year-end bundle. Extract the invoice page, rename it with the invoice number, and send just that page to your accountant.

Put it into practice

Extracting pages is one of the most common PDF tasks and one of the easiest to do badly. Plan the page list before you upload. Think about whether context is needed. Check the output end to end. Rename the result so someone else can tell what it is.

When you are ready, open the Split PDF tool, enter the page range you need, and save the extract as a new file. Every upload is encrypted in transit and deleted automatically within two hours, so you can extract pages from sensitive documents without leaving a trail.