Everyday PDF workflows that stop documents going wrong

Merging: safe, as long as the order is deliberate

Combining files copies page objects into one document. Text stays selectable and fonts travel with the pages, so merging is one of the safest operations there is.

What goes wrong is sequence. When a portal asks for "CV, then degree certificate, then experience letters", a reviewer reads the first two pages and stops. Name your source files with a numeric prefix before merging, and check the first and last page of the output every time.

Reordering, rotating and deleting pages

These operations touch page metadata, not content, so they never degrade quality. Rotation in particular is worth doing properly: a sideways scanned page is legible to a human who tilts their head and unreadable to an automated check.

Delete blank trailing pages before submitting. They look like a truncated document and can push a required page past a stated page limit.

Editing text: what is actually possible

Text in a PDF is positioned glyph by glyph, with no reflowing paragraph model. That means short edits — fixing a date, correcting a name, redacting a number — are realistic, while rewriting a paragraph usually reflows badly and needs the original source document.

If the page came from a scan, there is no text to edit at all; you are looking at an image. Recognising which of the two you have takes one second: try to select a word.

Signatures that do not cover the page

A photographed signature on paper carries a white or grey rectangle around it. Pasted onto a form, that rectangle hides the printed line, the date field and sometimes a clause. Removing the background so only the ink remains, with transparency around it, is what makes a signature sit on the page rather than on top of it.

Sign in dark ink on white paper, photograph it flat in even light, remove the background, then scale to roughly the height of the signature line. Keep one clean transparent PNG and reuse it.

File size: the limit nobody mentions until you fail

Government and university portals commonly cap uploads at 500 KB to 2 MB, and photo fields sometimes at 50 to 200 KB. A phone scan can easily be 8 MB per page.

Reduce size by downsampling embedded images rather than re-exporting the whole document as one flat image, which destroys the text layer and makes the file unsearchable. Aim for legibility at 100% zoom, then check the smallest printed text is still readable.

• Scan documents in greyscale at 200–300 DPI, not colour at 600 DPI.
• Keep an archival full-quality copy and a separate compressed upload copy.
• Verify after compressing: open the file and select text.

Privacy: prefer processing that stays on your device

Payslips, identity documents, degree certificates and offer letters are exactly the files you should not casually upload to an unknown converter. Where a task can be done locally in your browser, do it locally. Where a service is involved, check what it retains and for how long, and strip anything unnecessary from the document first.

Frequently asked questions

Does merging PDFs reduce quality?

No. Merging copies existing page objects, so text and images are unchanged. Quality loss comes from compression or from re-exporting pages as images.

Why can't I edit text in my scanned document?

A scan is a photograph of a page with no text layer. You would need optical character recognition first, and the result is an approximation.

What resolution should I scan certificates at?

200–300 DPI greyscale is legible, archival-adequate and small enough for nearly every upload limit.