Merge PDF
Built in your browser · nothing is uploaded
Merging copies every page of every file, in the order you set, into one new document. The pages themselves are untouched; same text, same fonts, same resolution, so the merged file is roughly the sum of the originals and nothing is re-compressed on the way through.
How to merge PDFs
What survives a merge is the part most tools decline to state. The pages themselves come through untouched: text stays selectable and searchable, images keep their resolution, and nothing is re-rendered or re-compressed. What does not come through is everything that belongs to the document rather than to a page.
That distinction is not a policy, it is the shape of the file format. A PDF keeps its outline, its form definition, its accessibility structure tree, its embedded attachments, its page labels and its metadata in one catalogue at the top of the document, and merging builds a fresh catalogue containing only pages. So bookmarks disappear, form fields stop being fields even where the boxes still draw, a tagged and accessible document comes out untagged, and the title and author fields are replaced with the merging library’s own name and today’s date. Internal links are the case that surprises people, because they look as though they should survive: the link annotations do get copied, but each one carries a reference to the page it pointed at, and that reference is copied alongside rather than redirected, so the link ends up aimed at an orphaned copy of the old page and goes nowhere. Reinstate anything you need in the finished file.
A digital signature is a different kind of loss and is not recoverable at all. A signature covers a byte range of one specific file, so the moment a new file exists there is nothing for it to certify. Sign the merged document rather than merging signed ones.
Encryption is the case to know about before you try it, because it fails without saying so. The parser these pages use contains no decryption at all. It is asked to ignore encryption, and that suppresses the error rather than unlocking anything: an encrypted file loads, the page count comes out right because the document structure is readable, and every page’s content stream stays encrypted. The merged result is then a fraction of the size it should be, with pages that render blank or broken in every reader, and the panel reports it as a success. That is equally true of a file which only forbids printing or editing and of one which needs a password to open. Take the protection off first, in whatever produced the file, and merge the unrestricted copy.
What people use it for
- Rejoining a contract after one page was printed, signed and rescanned
- Stitching appendices onto a report before it goes out
- Joining a cover letter to a CV so they arrive as one attachment
- Assembling certificates and transcripts into a single application file
- Rebuilding a document that arrived as one file per page
Questions
No. It runs on your own device, so the limit is your device’s memory. Hundreds of pages are routine.
No. Pages are copied intact, not re-rendered. The merged file is about as large as the originals added together.
Because nothing was compressed, and because shared resources are copied per source rather than pooled: merging two documents that embed the same font embeds it twice. Run the result through the PDF compressor if size matters.
No, and that goes for both kinds. Nothing here decrypts a PDF. Whether the file needs a password to open or only forbids printing and editing, the encryption stays on the page content, so remove the protection first and merge the unrestricted copy.
Nothing visibly fails, which is the problem. The page count reads correctly, the merge reports success, and the file you download is a fraction of the size it should be with pages that will not render. Open the output before you send it anywhere.
No. Both live in the document’s catalogue rather than on its pages, and merging builds a new catalogue holding only pages.
No. The link annotations are copied, but each one’s destination is copied with it rather than redirected, so it points at an orphaned duplicate of the old page instead of at the page now in the merged file. Cross-document links to a URL are unaffected.
They are dropped. The producer and creator fields are replaced with the merging library’s name and the creation date is set to now, so check the document properties before sending a merged file anywhere formal.
The tags are lost. The structure tree that screen readers follow is stored once for the whole document, so a merged file is untagged and has to be re-tagged if accessibility matters.
No. The parser runs in this page. You can disconnect from the network and it still works.