Suppose the selected rows are Part A, Part B, and Part C, but the intended output is C, A, B. Move Part C upward twice and verify the visible rows before merging. The result copies every page from C, then every page from A, then every page from B; it never interleaves them.
Order changes invalidate old output
If you move or remove a source after a merge, PDFMerge revokes the old download and requires a new merge. This prevents a visible order from disagreeing with a stale result link.
Inspect the downloaded PDF
Open the exact downloaded file in the reader used by its destination. Confirm the complete page sequence, page count, visible content, links, annotations, form behavior, accessibility, and any destination-specific requirements. Keep the sources: a successful page copy is not a certification that every PDF feature was preserved.
Understand the fresh document boundary
PDFMerge writes a fresh PDF containing copied visible pages. It does not promise preservation of document metadata, bookmarks, page labels, embedded files, portfolios, XFA, JavaScript, open actions, optional layers, tagged structure, PDF/A or PDF/X conformance, usage rights, encryption, incremental revisions, or digital signatures. A signature in a source does not sign the output.