Mixed scans often contain a few landscape receipts, forms or reverse sides among correctly oriented pages. The safest simple workflow is to identify those source page numbers in a viewer first, then express exactly that finite selection in PDFRotate.
Make a page inventory first
Record the total page count and every page that needs the same relative turn. For example, 1, 3-5, 8 means exactly five source pages. Do not infer the list from thumbnails in another file version. PDFRotate rejects duplicate, descending and out-of-range entries rather than guessing. If some pages need clockwise rotation and others need counterclockwise rotation, run separate operations and inspect between them.
Compare boundaries, not only one page
After download, compare the first selected page, the first and last page of each range, and at least one untouched neighbor. Confirm page count and order did not change. A page that looks correct in one viewer is useful evidence, but it does not prove forms, links, annotations, tags or signatures remain suitable in another destination.
A new PDF needs an independent check
PDFRotate changes page rotation entries and serializes a separate file. It does not promise byte identity or preservation of every document-level feature. Metadata, outlines, page labels, attachments, optional layers, tagged structure, PDF/A or PDF/X claims, usage rights, encryption, incremental revisions and digital signatures may change or stop applying. Forms, annotations, links, crop boxes and accessibility relationships can depend on viewer behavior. Keep the untouched source, open the download in another trusted viewer and inspect every affected page before using it.