You've got a PDF and you want fewer pages in it — or more accurately, you want to end up with a different set of pages than the one you started with. Three tools can get you there: delete, extract, and split. They overlap enough that "which one do I use" is a genuinely confusing question for anyone working with PDFs occasionally rather than constantly.
The short answer:
- Delete when you want to keep most of the document and remove a few specific pages.
- Extract when you want to keep a few specific pages and discard the rest.
- Split when you want to end up with multiple PDFs instead of one.
The longer answer covers why those rules of thumb work, what each tool actually does to your file, and the edge cases where the choice matters more than it first appears.
What each tool actually does
Delete PDF Pages takes a PDF in, lets you mark pages to remove, and gives you back the same PDF minus those pages. Order is preserved. If you delete pages 3 and 7 from a 20-page document, you get an 18-page document where pages 1-2 stay where they were, what was page 4 becomes page 3, and what was page 20 becomes page 18.
Extract PDF Pages takes a PDF in, lets you mark pages to keep, and gives you back a new PDF containing only those pages. The output can be a single combined file or one file per page, bundled into a ZIP.
Split PDF takes a PDF in and breaks it into multiple new PDFs based on rules: every N pages, at specific page numbers, by file size, or one file per page. The result is always multiple files, never one.
Organize PDF is the wildcard — it can do all three, plus reorder pages by drag-and-drop. It's the right tool for a mixed job: remove these three pages, also reorder these two, and maybe rotate that one. For pure-delete or pure-extract jobs it's overkill and the specialized tool is faster.
When delete is the right pick
Pick Delete PDF Pages when:
- You want to keep most of the document. A 200-page report with two pages of blank inserts is a delete job. Mark the two blanks, remove, done.
- The pages to remove are obvious by sight. Cover pages, blanks, separator slides, "intentionally left blank" pages, scanned cover sheets, advertisement inserts in saved articles — anything you'd pick out by clicking on a thumbnail.
- You don't want to think about what to keep. With 200 pages to preserve and 3 to drop, deleting is the natural framing. Listing out the 200 pages you want to keep would be silly.
A good test: if your natural sentence is "I want this PDF, but without page 7", delete is the right tool. If your natural sentence is "I want page 7 from this PDF", extract is the right tool.
When extract is the right pick
Pick Extract PDF Pages when:
- You want just a few pages out of a long document. A 400-page manual where you only need pages 200-220 is an extract job. Deleting the other 380 is technically possible but absurd.
- You're cherry-picking from a single source. Signature pages, proof-of-purchase pages, the one diagram you actually care about out of a long technical doc, the table of contents from a textbook scan.
- You need each extracted page as its own file. Extract has a "Separate PDFs" mode that produces one file per page. Useful for uploading each page separately to a system that wants individual PDFs.
The other test: if you find yourself selecting more than half the pages, you're probably reaching for extract when delete would be faster. Click the small set, not the large one.
When split is the right pick
Pick Split PDF when:
- The output is multiple files. That's the defining characteristic. If you'd be done with one file at the end, split is the wrong tool.
- The pages are partitioned by a rule, not picked individually. "Split every 10 pages" produces 10-page chunks without you selecting anything. "Split at page 50, 100, 150" produces four files. The rule does the work.
- You're chunking by file size. Long scanned documents that exceed a recipient's attachment limit get split into 20 MB pieces. The tool decides where the page boundaries fall.
Split is the only one of the three that creates documents rather than reshaping a single one. The other two preserve the "one-PDF-in, one-PDF-out" shape (extract's separate-PDFs mode is the closest exception).
The edge cases
Removing every page except a handful. This sits right on the boundary between delete and extract. The cleaner choice is extract: you're answering "which pages do I keep?" — that's the extract framing. The bigger reason: deleting a long list is error-prone (one missed page and you've leaked content); extracting a short list is self-checking (the output has exactly the pages you clicked).
Splitting at a single page. Splitting a PDF at page 50 produces two files: pages 1-49 and 50-end. If you only need the second half, extract pages 50-end is cleaner — you get one file directly, not two files with one to throw away. Use split when you actually want both halves.
Removing pages from a signed PDF. Any structural edit to a PDF invalidates a certificate-based digital signature, because the signed byte range no longer matches. This applies to delete and extract equally. If signature validity matters, work on an unsigned copy and re-sign after editing.
Removing pages from a password-protected PDF. All three tools need to read the page structure, which encryption blocks. Use Unlock PDF first (you'll need the password), then edit, then re-apply protection with Protect PDF if needed.
A page that should be deleted but contains sensitive text in the metadata. Deleting a page removes its content from the visible document, but if the page contained sensitive metadata (annotations with private notes, embedded files attached to that page), those go with it. If the goal is privacy rather than cleanup, the safer tool is Redact PDF — that explicitly removes content from a page while keeping the page structure intact.
What the three tools share
Common to all three:
- Lossless on page content. None of them re-encode the pages they keep. Text stays as text, images stay at their original resolution, fonts stay embedded.
- They work on a copy. Original file is unchanged on your device; the operation runs on the uploaded byte stream and returns a new file.
- They can't unlock a PDF first. All three need an unlocked PDF as input.
- They preserve embedded data on retained pages. Annotations, form fields, hyperlinks on the pages that survive the operation carry forward intact.
The differences are entirely in how you describe what you want: the page set, the framing, and how many files come out.
A quick decision tree
When you're unsure which tool to reach for:
- Will you end up with more than one file? If yes, split.
- Otherwise: are you keeping more than half the pages? If yes, delete the few you don't want.
- Otherwise: extract the pages you do want.
- **Doing several different operations at once (delete + reorder
- rotate)?** Use Organize PDF — it bundles everything into one pass.
The first question is the most useful: it separates split from the other two cleanly. The "more than half" rule on step 2 is just a heuristic for "what's easier to click" — if you'd rather select the keeper set than the discard set, extract is the right framing regardless of the count.
Common questions
Will deleting pages reduce the file size?
Usually yes, proportionally to the content. Removing a 5 MB image-heavy page from a 30 MB document drops the file to roughly 25 MB. Removing a near-empty page from the same document barely moves the needle. If size is the goal, follow up with Compress PDF — the combination of "delete bulky pages, then compress what's left" typically wins over either operation alone.
Can I undo a page deletion?
Not on the output file. Keep the original; the edit runs on a copy and produces a new file. If you delete the wrong page, go back to the original and start over.
Do these tools work on scanned PDFs?
Yes — they operate on the PDF page structure, not on text content, so it doesn't matter whether the pages contain real text, scanned images, or both. The visible content of each page is preserved as-is.
Can I delete pages from multiple PDFs at once?
Not in a single operation. Each tool takes one PDF at a time. For repeated bulk operations across many files, Workflows can chain a delete-pages step into a larger pipeline and apply it to a batch — but the per-file selection has to be the same across all files (e.g. "always delete the cover page"), since the workflow can't know which pages to drop on each individual document.
What's the difference between delete and redact?
Delete removes the entire page from the document — it no longer exists in the output. Redact keeps the page but blacks out specific text or regions on that page. If you want the page gone, delete. If you want the page to stay but a specific portion to be unreadable, redact.
Will hyperlinks and bookmarks survive?
Hyperlinks targeting pages that get deleted will land on whatever took that page's place (since page numbers shift), which is usually not what you want. Bookmarks pointing to deleted pages will dangle. If your document relies on internal navigation, do a final pass through it after the edit to confirm the targets are right — or use Organize PDF which can re-attach bookmarks during reordering.
Wrap-up
The three tools answer three different questions about a PDF:
- Delete: Which pages don't I want?
- Extract: Which pages do I want?
- Split: How do I break this into multiple files?
When the question is well-phrased, the tool picks itself. The most common mistake is reaching for the wrong framing — selecting 180 pages to "extract" when you really wanted to delete 20, or splitting a file at one point when you only needed one half of it. Stop and phrase the goal as a sentence before opening any tool; the right choice usually falls out of how you describe the result.