How to merge PDFs without uploading them anywhere

Is it safe to merge PDFs online?

For a restaurant menu and a flyer, sure. The trouble is that the PDFs people combine most often are exactly the sensitive ones: a rental application with three payslips attached, a mortgage pack, a passport scan plus proof of address, a medical report for an insurance claim. Between them they carry your full name, home address, tax or social security number, bank details, salary and signature.

Almost every popular merge site works the same way behind the button. Your files are uploaded, written to a server disk, processed there and left behind a download link for a while. Whether they are deleted after one hour or a day is their policy, not your choice. Even with an honest operator, the real exposure is in the parts you cannot see:

• Copies you don't control. Logs, backups and the cloud providers a service runs on can hold onto data longer than the headline promise.
• Links travel. A download link that gets forwarded, synced or saved in a browser history opens the document for as long as it lives.
• Breaches. A server that receives thousands of ID scans a day is a tempting target, and an ID plus a payslip is most of what someone needs to attempt fraud in your name.
• Someone else's data. If the PDF holds customer or employee records, sending it to an unvetted third party may break your employer's data protection rules (and, in the EU or UK, GDPR) without you realising.

A sensible rule: documents with no personal data can go anywhere. Everything else should be merged on your own machine.

Combining PDFs on a Mac with what's already installed

macOS is the easy case. It has two built-in ways to do this.

Preview

1. Open the first PDF in Preview and show the sidebar with View › Thumbnails (Option + Cmd + 2).
2. Drag the next PDF from Finder into the sidebar, dropping it exactly where its pages should go. Edit › Insert › Page from File does the same.
3. Drag thumbnails to reorder and press Cmd + R to rotate a sideways scan.
4. Save with File › Export as PDF under a new name.

Finder Quick Actions

Select several PDFs in Finder, right-click and choose Quick Actions › Create PDF. You get one new file containing all of them, in the order they appear in the window. If order matters, prefix the names first: 01-application.pdf, 02-payslip-may.pdf.

Windows can't merge PDFs on its own

Plenty of guides dance around this, so plainly: Windows has no built-in way to combine PDFs. File Explorer can't do it and neither can the PDF viewer in Edge.

The workaround you'll often see, Microsoft Print to PDF, doesn't merge anything. It prints one document at a time and produces one file per document, so selecting five PDFs and choosing Print gives you five new PDFs. It also rebuilds each file as if it were paper: bookmarks vanish, links stop working and form fields are frozen.

That leaves three honest options on Windows: pay for Acrobat, install a third-party program and trust it, or use a tool that runs inside the browser without sending the file anywhere.

How a website can combine PDFs offline

Modern browsers can read a file from your disk into memory and run code that processes it right there. The Utiliobox PDF toolbox is built that way. It uses two open-source libraries, pdf.js to read and draw pages and pdf-lib to assemble the new document, served from the site's own domain the first time you need them. When you drop files in, the browser reads them locally; when you click download, the merged PDF is built on your device and saved straight to your downloads folder.

You shouldn't have to take that on faith. Here are two ways to check.

Test 1: pull the plug

1. Open the tool, add any harmless PDF and wait a couple of seconds: as soon as a file is loaded, the tool finishes downloading its engines. Remove it afterwards.
2. Turn off Wi-Fi or switch to airplane mode.
3. Now load your real documents, merge them and download the result.

If it works with no connection, your files cannot have gone through a server. For extra peace of mind, keep the network panel from test 2 open when you reconnect: nothing "queued" gets sent afterwards either.

Test 2: watch the Network panel

1. Open developer tools (F12 on Windows, Option + Cmd + I on a Mac) and switch to the Network tab.
2. Tick Preserve log and clear the list.
3. Load your PDFs, merge them and download.
4. Look at the Method and Size columns. An upload shows up as a POST or PUT request roughly the size of your file.

What you will see: the engine scripts loading (files such as pdf.min.mjs and pdf-lib.min.js, first time only), requests from the ads on the page, and blob: or data: entries, which are the thumbnails and the finished file being created in memory. What you won't see is any request carrying your megabytes out.

Merging and reordering, step by step

1. Go to PDF Tools and drag your files onto the drop zone, click to pick them, or paste with Ctrl + V. Dropping them in the right order saves dragging later.
2. Every page becomes a numbered thumbnail. Hover over the number to see which file a page came from.
3. Drag thumbnails to reorder. On a phone, the arrow buttons on each card do the same job without fighting your thumb.
4. Rotate sideways scans with the rotate buttons, and remove stray pages with the cross, or tick several and use "Delete pages".
5. Forgot something? Drop it on "Add more files". Its pages join at the end and you move them where they belong.
6. On the Organise & merge tab, click Download merged PDF. The file is saved as utiliobox-merged.pdf; rename it to something the recipient will recognise.

Images can go into the same document: a JPG, PNG or WebP photo of your driving licence becomes one more page, wherever you place it. And Reverse order fixes a stack that went through the scanner upside down in a single click.

What survives a merge, and what doesn't

Page content yes, document structure no

Pages are copied as they are: text stays selectable and searchable, and nothing is recompressed, so quality is untouched. But the output is a brand-new document. Bookmarks (the outline in the sidebar) from the source files are not carried over, and neither are the title and author stored in the file's properties.

Fillable forms and digital signatures

Don't count on a fillable form staying fillable after merging. Fill it in first, check the answers are visible in the result, and send the original as well if it really matters. Digital signatures (certificate-based ones, not a scanned squiggle) follow a rule that applies to every tool, not just this one: the signature covers the exact bytes of the original file, so any merge invalidates it. The signature image may still show, but it no longer proves anything. If a signed document has to arrive intact, attach it separately.

Mixed page sizes

Each page keeps its original dimensions, so a US Letter contract merged with an A4 annex from a European office ends up with both sizes. On screen you'll barely notice; when printing, the printer scales each sheet. Watch out for phone photos, though: a 4,000-pixel-wide image becomes a very large page. It looks fine in any viewer, but crop or downsize the photo first if it's going to be printed.

Troubleshooting

• "Password-protected." The tool doesn't open encrypted PDFs. If you know the password, open the file in Chrome, enter it, and use Print › Save as PDF to get an unencrypted copy (with the print-to-PDF losses described above), then load that copy. If printing is also blocked, ask the sender for an unrestricted version.
• Thumbnails appear but the download fails. Typical of PDFs that open without a password but carry permission restrictions, which banks and government portals love. Same fix: save an unprotected copy and merge that.
• The browser runs out of memory on huge scans. Your device sets the limit. An ordinary laptop handles about 100 MB and 500 pages comfortably; a phone, far less. Compress each scan on its own in the Compress tab, which is designed for scans, then merge the slimmed-down files.
• The merged file is bigger than expected. Merging doesn't shrink anything: expect roughly the sum of the originals, sometimes a little more. For heavy scans, compress afterwards. For text documents, don't, because compression turns pages into images and the text stops being searchable (the tool warns you first).
• Duplex scan with all the even pages at the end. There's no automatic interleave. Use "Reverse order" if the batch came out backwards, then drag the even pages into place.
• "The file may be damaged or incomplete." Usually a half-finished download or a truncated email attachment. Download it again and make sure it opens in your normal viewer.

Other jobs on the same working document

Whatever you load stays open as one working document, so you can chain operations without sending anything anywhere:

• Split & extract: by ranges (1-3, 5, 8-), one page per file, or only the pages you've ticked. Multiple outputs arrive together in a .zip.
• Compress: re-encodes each page as an image at the quality and resolution you pick. Great for scans; if the result is larger than the original, the original is kept.
• Convert to images: exports pages as PNG or JPG at 96, 150, 200 or 300 dpi.
• Numbers & watermark: numbering in six positions and three formats, starting from any number, plus diagonal text such as "CONFIDENTIAL" with adjustable size, opacity and angle. It uses the standard PDF font, which covers accented Latin characters but not scripts like Cyrillic or Chinese.