Why "Print to PDF" is not a merge
The advice Windows users are most often given is to print several documents to the Microsoft Print to PDF driver. It appears to work and quietly does real damage.
Printing rasterises the page through the print pipeline. The output is a picture of your document: the text can no longer be selected or searched, links stop working, bookmarks are gone, and the file is usually larger than the originals it replaced. For anything that will be read back, searched, or converted to Word later, it is the wrong tool.
It also cannot combine files in one pass — you have to print each document in turn into the same job, which is where the page order usually goes wrong.
If a document has already been through Print to PDF, the text is not recoverable without OCR. Go back to the originals.
Merging on a work computer with no admin rights
This is the common case, and it is the strongest argument for a browser tool. Corporate builds block installers, and the free PDF software most people reach for either needs admin rights or bundles something the IT team will object to.
A web page needs neither. It runs in the browser that is already approved, with no plugin and no elevation, and because the file is processed locally there is no upload traffic for a proxy to inspect or block.
That last point matters for the other reason too: a document you are not allowed to install software to handle is usually a document you are not allowed to upload to a free website either.