A reading pack is a small act of editorial judgement. You have three long reports, a meeting in an hour, and a group of people who will not read three hundred pages between now and then. What they need is the chapter that matters from each report, in the order you will discuss them, in a single document they can actually open and finish. Assembling that by hand, printing pages, scanning them back, reordering them, is slow and error-prone. Pulling the exact pages out of each PDF, in the order you choose, takes minutes and produces a cleaner result.
Extract Pages pulls chosen pages out of a PDF into a single new document, in the order you list them. You can name pages like 3 and 1, or ranges like 1-3 and 5, and the result follows your list, not the original order. That custom-order feature is what turns extraction into pack-building: you are not just taking pages out, you are sequencing them for a reader.
Start from the discussion, not the documents
The most common mistake in building a reading pack is starting from the documents: opening the first report, taking its first chapter, then moving to the second report, and so on, in source order. The result reads like three beginnings stapled together. Start instead from the discussion. Write down the order of topics as they will actually be covered: the problem first, then the evidence, then the options. Then find the pages that serve each topic, wherever they sit in their source documents. A page from the third report may belong first, and a page from the first report may belong last. The pack serves the meeting, not the filing cabinet.
That order-first approach also disciplines length. A pack that mirrors its sources grows without limit, because every source feels important. A pack built from a topic list has a natural stopping rule: when every topic has its pages, the pack is done. Aim for a pack a busy reader can finish in the time available. Twenty well-chosen pages beat sixty dutifully included ones, because the twenty will actually be read.
Pull pages in the order you will present them
Extraction with custom order is the mechanical heart of the job. Suppose your topic order calls for the conclusion of one report before the evidence chapter of another. Listing the pages in presentation order, rather than extracting each report's pages separately and then merging, keeps the sequence visible in one step. Check the result by reading the pack's own page order, not the sources': does page one introduce the problem, does the middle develop it, does the final page land where the discussion should end? If the pack's flow feels wrong, the fix is to re-list the pages in a better order, not to rewrite anything.
Keep the source documents intact. Extraction creates a new document; it does not alter the originals. That matters for two reasons. First, you can rebuild the pack differently for a different meeting without damaging the sources. Second, anyone who wants the full context can still find it in the original report, which the pack should acknowledge. A short cover note naming each source and its page range is a courtesy that costs one paragraph and prevents confusion about where the pack's pages came from.
Worked example: a fictional charity trustee pack
The example is fictional: an invented charity, invented reports, and invented page numbers. Imagine the fictional Harbour Youth Trust, contact via hello@example.org, preparing a trustee meeting pack from three reports: an annual review, a funding proposal, and a safeguarding update. The trustees have thirty minutes of reading time, and the three reports total over two hundred pages.
The meeting will cover three topics in order: safeguarding first, because it needs a decision; funding second; and the annual review's headline figures last, as context. The organiser pulls the safeguarding chapter, pages 8 to 11, from the safeguarding update first. Next come pages 4 and 5 of the funding proposal, the summary and the budget table, in that order. Finally, pages 2 and 3 of the annual review, the headline figures, close the pack. The resulting document is eight pages long, in discussion order, and every trustee can read it before the meeting. No one was asked to find the relevant pages themselves, and no one received two hundred pages in the hope that the right eight would be found inside them.
Note what the pack does not do. It does not summarise the reports, and it does not replace them. A trustee who wants the full safeguarding update can still open the original. The pack's job is narrower and more valuable: it puts the right pages, in the right order, in front of the right people, before the meeting starts.
Check the pack before you send it
A pack assembled from several sources needs a check that a single-source document does not. Read the first and last lines of each extracted section to confirm nothing was cut mid-sentence by a page boundary you misjudged. Check that page numbers make sense: extracted pages keep their original page numbers if those numbers are printed on the page, so a pack whose printed numbers jump from 8 to 4 to 2 can confuse a reader who expects sequential numbering. A cover note explaining that the pack preserves original numbering, or a short contents line listing each section's source, removes that confusion.
Also check for duplication. Two reports often repeat the same background paragraph, and a pack that pulls from both can repeat it twice. Reading the pack straight through, as a reader will, catches repetitions that a page-list view hides. If two sections overlap heavily, keep the clearer one and drop the other. The pack is a curated document now, and curation includes cutting.
Honest limitations
Extraction copies pages as they are. It does not rewrite, summarise, or reformat them, so a pack assembled from different sources will look like what it is: pages from different documents, with different layouts and numbering. That is acceptable for a working handout, and it is honest: readers can see what came from where. Extraction also cannot pull pages that do not exist in a form the tool can read; scanned pages without a text layer still extract as images, which is fine for reading but not for searching. If your pack needs its pages joined with other material afterwards, Merge PDF joins files in your chosen order, and if you need to cut a whole document into parts rather than pull selected pages, Split PDF does that broader job. The distinction matters, and the next post in this series explains it.
FAQs
How do I save only certain pages from a PDF?
List the pages you want, such as 3 and 1 or 1-3 and 5, in Extract Pages, and download the resulting document. The extracted pages form one new file, in the order you listed them, and the original PDF is unchanged.
Can I extract pages in a different order from the original?
Yes. The result follows the order of your list, not the original document's order. That is what makes extraction useful for building reading packs and handouts sequenced for a presentation or meeting.
Will extracted pages keep their original quality?
Yes. Extraction copies the pages as they are into the new document, rather than re-rendering them, so their appearance and quality are preserved. Printed page numbers on the pages themselves stay as they were, so a cover note helps readers follow a pack drawn from several sources.
What is the difference between extracting pages and splitting a PDF?
Extraction pulls selected pages into one new document in your chosen order. Splitting cuts a whole document into multiple parts. If you want a few pages as one file, extract; if you want every part of a document as separate files, split. For a general guide, see how to merge and split PDFs.
human review pending client designation