Join separately scanned PDFs into one file

A claim file scanned in five sittings

Tomas is putting together a water damage claim. Over one week he fed paper into the office copier whenever he had a spare minute, and its scan-to-email button sent him a separate file each time: scan_20261027_0914.pdf with the four-page claim form, scan_20261028_1702.pdf with a two-page repair quote, six single-page receipts scanned one by one, and a final batch with the landlord's letter, front and back. That is nine files and roughly 31 MB, all in colour at the copier's default resolution.

The insurer's portal takes exactly one PDF of no more than 10 MB, and the adjuster wants the claim form on top. File names built from timestamps say nothing about content, two receipts came out sideways, and the duplex pass left an empty back page behind every receipt. None of that is hard to fix, but each problem is far easier to catch while the scans are still nine small files than once they are one long one.

Tidy each scan before it joins the pile

Spend five minutes on this list first. Every item is quicker on a short file than inside the finished bundle.

• Open every file and note what it really is. Write a short running order on paper: claim form, quote, receipts by date, letter. Do not trust timestamp names to sort themselves.
• Turn sideways pages upright. Rotate PDF fixes a landscape receipt that the feeder read the wrong way round.
• Drop empty reverse sides and accidental double scans. Organize pages shows thumbnails, so a blank page or a sheet scanned twice is easy to spot and remove. The guide to deleting empty pages covers the faint grey pages that are not quite blank.
• Turn photos into PDF pages. If a receipt exists only as a JPG from your phone, put it through Images to PDF first. The merge tool accepts PDFs only.
• Check for a password prompt. Scans rarely carry one, but a file someone forwarded to you might. The merge tool refuses locked PDFs.

Joining the scans in PDFGym

With every part clean, the join itself takes a minute in Merge PDF:

1. Open the merge page and drop all the cleaned files onto it at once, or choose them with the file picker. Each one appears in a list with its page count and size.
2. Compare those page counts with your notes. A file showing twelve pages where you expected one receipt is the moment to stop and look.
3. Put the list in your running order with the up and down arrow buttons beside each file. The order of the list is the order of the pages in the output, file by file.
4. Read any notice under a file name. If a source carries bookmarks, form fields, custom page labels or attachments, the tool says so before it runs, because those may not carry over.
5. Press the merge button. The new file is reopened and compared with the sources for page count, page order, page sizes and selectable text, and a result that fails is not offered for download.
6. Download it. The name is taken from the first file with "-merged" added, so rename it to something the adjuster will recognise, such as Tomas-Hruby-claim-water-damage.pdf.

What survives the join

Scans are pictures of paper, and joining them is a copy, not a redraw. This is what each part of a scanned source becomes in the output.

Bringing the bundle under the portal limit

Only now does size matter. Open the joined file in Compress PDF and give it a ceiling. For a 10 MB portal, typing 9.5 MB leaves a margin in case the portal counts a megabyte as a flat million bytes. The tool reads every candidate back and checks its text against the source before offering it, then shows the final size and a preview next to the original.

Colour scans of mostly white paper usually leave room to work with, because the page images are where the weight sits. If the run stops short of your number, the result says by how much and offers one more pass that redraws page images smaller, with softer photos as the price. Shrink once, on the finished bundle: squeezing each scan first means every image is encoded twice and the parts never add up to the number you need.

Scan-pile slip-ups and their fixes

• Sorting by file name. Copier names follow the moment of scanning, so a receipt rescanned on Friday lands after the letter. Order by content, using your written list.
• Fronts and backs in separate files. A manual duplex job often gives one file of odd pages and one of even pages. Merge them, then drag the thumbnails into sequence in Organize pages.
• A colour scan of a black and white page. Grey text on a coloured cast weighs far more than it needs to. If a page is worth rescanning, choose greyscale on the copier.
• Sending the copy before checking it. Scroll the joined file from start to finish once. A missing back page or a duplicated receipt is obvious at that speed.
• Keeping no originals. Hold on to the separate scans until the claim is accepted. If the adjuster asks for one receipt alone, you have it.

When a browser merge is not the best route

If you have not scanned yet, let the scanner do the joining. Most copiers and feeder scanners can save a whole stack as one multi-page file, which skips this job entirely; the guide to multi-page scanning covers those settings. If the documents need searchable text and the scans have none, use the scanner software or a desktop app with text recognition, because joining here adds no text layer.

Some recipients want each document separately: a portal with one upload slot per receipt, or a solicitor who asks for originals by post. Follow what they ask, even when one file seems tidier. And a pile of several hundred pages is better joined on a laptop than on an older phone, since every file sits in the browser tab's memory while the merge runs.