Merging PDFs is a structural operation: page objects are copied from several documents into one. Nothing is re-rendered, so the visual content comes through exactly as it was. But a PDF contains more than pages, and some of that other material does not survive the trip.
Here is an honest inventory, so you find out now rather than after sending the file.
What survives, reliably
Page content. Text, images, vector graphics, all at original quality. No re-compression, no re-rendering. A merged page is byte-for-byte the same drawing instructions as the source page.
Text selectability and searchability. Text stays text. Ctrl+F works across the merged document.
Page size and orientation. Each page keeps its own. Merge a landscape report into a portrait one and the landscape pages stay landscape — correct behaviour, though it surprises people who expect uniformity.
Embedded fonts. Carried across, so the document renders identically on machines that do not have the fonts installed.
Image resolution. Untouched. A 300 DPI scan is still 300 DPI after merging.
What is usually lost
Bookmarks and the outline tree. This is the big one for long documents. If you merge three reports that each had a navigation sidebar, the merged file typically has none. The pages are all there; the way of jumping between them is gone. Most merge tools, ours included, do not rebuild the outline.
Form fields and entered data. Interactive forms are the riskiest case. Field definitions may survive, may be dropped, or may collide — two source documents each with a field named "name" cannot both keep it. Any values already entered frequently disappear.
If you are merging filled-in forms, flatten each one first: open it, print to PDF, and the entered values become permanent page content that merges safely. You lose interactivity, which is usually the point.
Internal links. A link that pointed to page 12 of its original document either breaks or points to page 12 of the merged file, which is now some unrelated page. External links to websites usually survive; internal navigation usually does not.
Document metadata. Title, author, subject and keywords come from one source document or get reset. If metadata matters — for a library submission, say — set it after merging.
Attachments and embedded files. Files attached to a PDF are commonly dropped.
Digital signatures. Always invalidated, without exception. A signature certifies the document it was applied to; merging creates a different document, so the signature is no longer valid — that is the entire point of a signature. If you need signed documents combined, get the combined document signed.
What this means in practice
For the ordinary cases — combining scanned pages, assembling receipts, joining chapters, putting a cover page on a report — none of this matters. The pages are what you wanted and the pages are what you get.
Be careful when:
Merging does not compress
A common misconception: people expect a merged file to be smaller than the sum of its parts. It generally is not. If three 2MB PDFs merge to roughly 6MB, that is correct — the page content still has to be stored.
Some overlap is deduplicated in good implementations, mainly fonts used by several source documents. That saves a little. It is not compression.
If the merged file is too large, compress it afterwards as a separate step.
Order it before, not after
Set the page order by dragging thumbnails before you merge. Fixing the order afterwards means either starting again or reorganising the merged file with Organize PDF — both slower than getting it right the first time.