How to Extract Pages From a PDF (and Reorder Them While You Are There)

Somebody sends you a 60-page contract and you need the clause about deposits. A bank sends a statement and you need page 2. A colleague sends a report and you need the two pages with the charts on them.

Every one of those is an extract: you want a document that contains some of the pages and nothing else.

What extracting actually does

The Extract PDF Pages tool has one field, and you fill it with the pages you want:

• A single number - 7
• A range - 4-9
• A mixture - 1, 4-9, 12
• And, if you want, the order you want them in

The result is a new PDF built from those pages. The pages you did not name are not carried over - they are never written into the file at all. What you download contains exactly what you asked for, in the order you asked for it.

That last part is the useful bit, and it is easy to miss: the order of your list is the order of the output. Write 9, 3, 4 and you get those three pages in that order. That is the fastest way to reorder a selection without touching the rest of the document - the alternative is to reorder the whole file, and for a long document that means writing a list as long as the document.

Extract or delete: they are not the same

Both operations rebuild the file, so neither one carries over the pages you left out. The difference is what the result is:

So delete when you want "the document minus these pages" - the full report with the confidential annex removed. Extract when you want "these three pages, as their own file" - the pages to email, the appendix to attach, the selection to reorder.

The delete-pages guide covers the other side in detail.

Why it matters when the pages are sensitive

Because the pages you did not name never reach the output, extraction is the operation to reach for when a page has to be gone rather than merely absent from the view.

The pages are copied into a brand-new document. Anything you did not list - the signature page, a payslip, a page from somebody else's file, a page of somebody else's medical detail - is not in the file you are about to send.

There is one thing extraction does not do, and it is the same thing for every PDF tool: it does not touch the metadata. The document information dictionary travels with the file, not with the pages, so a fresh extract still carries the author, the title and the creation date of the original. For a file going to someone you have never met, strip the metadata as well.

The numbering trap

The page numbers are the page numbers of the file, not of your reader's thumbnail.

Most readers count from the first page you can see, which is often a cover, a title page or a fax header. The file counts every sheet, including the blank one the scanner produced. A document with a cover and a contents page has its real first page at number three.

The reliable way to avoid the guesswork is to split the document into single pages first, look at the list, and then extract the ones you want. It is one extra step, and it removes the off-by-one entirely.

Two mistakes worth naming

Extracting the wrong page because the count was off by one. This is the common one, and it has a simple fix: open the PDF in a reader that shows the page number, and check the page you are about to name.

Expecting the result to contain the pages you did not name. An extract is not a "select these pages" toggle - it is "build a file from these pages". If you want the rest of the document too, you want delete instead.

A short checklist

• Check the page numbers against the file, not against the reader's thumbnail.
• The order you write is the order you get - that is how you reorder a selection.
• You can repeat a page if you genuinely want it twice; anything you do not name is not in the result.
• The original is untouched. You get a new file, and nothing was uploaded.
• Extract the pages, then strip the metadata if the file is going to a stranger.
• Keep the original. There is no undo, and a re-scan is an afternoon.