Merging multiple PDFs into a single file

Six items to send off: a completed form, two scanned proofs of address, a signed contract, an invoice exported from your accounting software and a certificate that arrived by email. The recipient wants one. A single file, in the right order, readable the moment it opens. The merge itself takes three seconds; it is the minutes before it and the check after it that decide whether the bundle is accepted or bounced back.

What merging actually does to your files

A PDF is not a picture of a document. It is a container of objects — pages, fonts, images, annotations, form fields — together with a cross-reference table that records where in the file each object sits. Merging means copying the objects from every source into a new container, then rebuilding that table.

The consequence matters: nothing is recompressed. A page scanned at 300 DPI stays at 300 DPI, vector text stays vector, an already-embedded JPEG photo keeps exactly its original quality. Unlike a conversion, a merge degrades nothing. The only bytes you save come from de-duplicating shared resources: if all six documents embed the same font, a decent engine stores it once. Beyond that, expect a file roughly the size of all the inputs added together.

Page order, the number one trap

Almost every botched merge fails for a daft reason: the order. Most tools sort files alphabetically, and the alphabet cannot count. scan_1, scan_10, scan_11, scan_2 — that is the sequence you will get if your filenames have not been prepared.

Two habits head the problem off:

• Number with two digits, leading zero included, before you even open the tool: 01-cover-letter.pdf, 02-contract.pdf, 03-appendix.pdf. Alphabetical sorting then becomes the correct sorting.
• Check how the tool behaves. No web tool remembers the order in which you clicked the files: the browser always hands over the selection in folder order. So it is the filename that decides, not the route you took through the dialogue box. Plenty of online tools simply keep the order the browser gives them, which is the folder's alphabetical order — preparing the names in advance is the one habit that works everywhere.

If you spot the mistake afterwards, there is no need to start from scratch: put the files back in the right order and run the merge again — one operation, rather than splitting the document and reassembling it.

Make everything consistent before you assemble

A real-world bundle is rarely made up of PDFs alone. There is the table exported to Word, the photo of a receipt taken on a phone, the certificate saved as JPG. All of it has to become PDF before the merge: run office documents through Word to PDF conversion and images through JPG to PDF conversion, which puts each image on its own page.

Three inconsistencies are worth checking before you assemble:

• Page sizes. A document exported from a US application comes out as Letter (216 × 279 mm) while the rest of your bundle is A4 (210 × 297 mm). On screen the difference goes unnoticed; in print the margins shift: on Letter paper an A4 page (297 mm tall against 279) loses its bottom edge, footer included; on A4 paper a Letter page (216 mm wide against 210) is trimmed at the sides.
• Orientation. A landscape table dropped in among portrait pages makes the reader tilt their head. It is not a technical fault, but it is a page many people will simply not read.
• Upside-down scans. A sheet feeder that swallows a page the wrong way round produces a page rotated by 180°. Rotation in a PDF is only an attribute stored on the page: correcting it does not re-encode any image and costs nothing in quality.

The size of the finished file

A few rough figures, so you know what to expect. A page of text exported from a word processor is typically 30 to 80 KB. A page scanned in colour at 300 DPI lands somewhere between 300 KB and 1.5 MB. A smartphone photo turned into a PDF page comes in at around 3 to 5 MB. In other words, a twenty-page bundle made up mostly of scans comfortably reaches 15 to 25 MB, while most email services refuse attachments beyond roughly twenty megabytes.

The answer is compression, but after the merge, not before: a single pass over the assembled file treats all the images consistently and benefits from de-duplication. Compressing the finished PDF works mainly on two levers: downsampling the images (going from 300 to 150 DPI cuts the size of a scanned page by a factor of three or four) and a more aggressive JPEG recompression. Vector text is left alone.

Pick the level according to the use: 150 DPI is ample for reading on screen, while 200 to 300 DPI remains necessary if the document is to be printed properly or put through OCR by the recipient.

What can break, and why

Password-protected PDFs

A file that demands a password just to open will simply be refused by the merge tool, which cannot read its contents. Some PDFs open without a password but carry usage restrictions that forbid page extraction: the result is the same. If the document is yours and you have the password, removing the protection before merging settles it.

Electronic signatures

A digital signature certifies the file as it stands, byte for byte. Dropping that file into a larger document inevitably breaks that fingerprint: the signature will show as invalid, if it does not vanish altogether. The practical rule is to reverse the order of operations: merge first, then have the assembled document signed. If a signed copy has to be passed on, send it as a separate file.

Fillable forms

Two different forms often contain fields sharing the same internal name (name, date, signature). After merging, those fields become linked: whatever you type in one appears in the other. The reliable workaround is to flatten each form before assembly, for instance by printing it to PDF, which turns the fields into fixed content.

Bookmarks, internal links and compliance

Navigation bookmarks and the clickable table of contents of the source documents survive to varying degrees depending on the engine, and internal links that used to point at ‘page 4’ now point to a different page. In the same way, if one of your files met the PDF/A archiving standard, the merged document's compliance is no longer guaranteed as soon as the other sources fall short of it.

The text inside scans

Merging does not create a text layer. A scanned receipt stays an image: you cannot search it, you cannot copy a line out of it. If the bundle needs to be usable, character recognition has to be run on each source, before assembly.

The final check, in one minute

Before you send, four checks are enough to avoid a round trip:

1. Does the total page count match the expected sum? A missing page points to a source file that was skipped or is corrupt.
2. Are the first and last pages the right ones? That is the quickest test of the order.
3. Does the file open in a reader other than your own — browser, mobile? A slightly malformed PDF sometimes passes in one application only.
4. Does the size fit within the limit of whatever you are sending it through?

Assembling your PDFs on Convertu

The PDF merge tool takes several files, lets you set their order with the up and down arrows, and returns a single document with no recompression of the pages; the maximum size accepted per file is shown in the upload area itself. Processing happens server-side, as it does for all the PDF tools: the files you upload are deleted within the hour. One conversion a day is free with no sign-up; beyond that, the €7-a-year subscription covers all nineteen server-side tools in the catalogue, with no commitment.