Interleave PDF Pages

The scene: a flatbed or sheet-fed scanner that only scans one side. In front of it, a 120-page signed agreement that has to go digital today. So you improvise, like everyone does. Run the stack through once and save the fronts. Flip the stack, run it again, and save the backs. Now you own two PDFs — pages 1, 3, 5, 7 in one file and pages 2, 4, 6, 8 in the other — and no document.

Manual reordering is the road most people start down, and the road most people abandon around page 30. Dragging thumbnails alternately between two windows takes an afternoon, invites exactly one silent mistake that ruins every citation after it, and has to be redone from scratch when the scan turns out crooked. The task is mechanical. That is the giveaway. Mechanical tasks should not have human hands inside them.

Interleave PDF Pages does the whole repair in one move: two files in, one correctly ordered document out, alternating page by page. It runs in your browser, asks for no account, uploads nothing, and includes the offset option for the odd-page and mismatched-count cases that trip up simpler tools.

Interleaving is a specific merge pattern. Where a standard PDF Merge concatenates — all of file one, then all of file two — an interleave alternates. First page of A, first page of B, second page of A, second page of B. Applied to your fronts-and-backs scans, the output reads 1, 2, 3, 4 straight through.

The tool's few controls exist for the real-world messes around that clean definition:

• Which file leads — fronts first is the usual order, but a test page confirms it before you commit 120 pages.
• An offset for unequal files, because 61 fronts against 60 backs needs to know where the odd page lives.
• A pass-through mode for leftover pages — the unpaired final page joins the end of the document instead of being silently dropped.
• Instant preview of the first interleaved pages, so a back-to-front error announces itself in five seconds rather than after the download.

That last one deserves a moment of appreciation. The costliest interleaving failure is not a crash; it is a silent success that ordered the pages wrong. A preview that shows pages one through six before you save is the difference between a document you trust and a document you re-check page by page.

Here is the full routine for the classic case — a double-sided original scanned as two one-sided passes. Ninety seconds, most of it spent on the sanity check.

1. 1

   Scan the fronts

   Run the stack through in reading order and save the result as fronts.pdf. Note the page count.

2. 2

   Scan the backs

   Flip the stack — the usual way is to turn it over left to right, keeping the same top edge — and save backs.pdf. If the scanner rotated everything 180 degrees, rotate the file first.

3. 3

   Load both files

   Open them in the tool. It shows both page counts and previews the first alternating pages immediately.

4. 4

   Set the order and offset

   Fronts lead. If the counts differ by one, set the offset so the unpaired page falls where the original had it — usually the end.

5. 5

   Verify on a known page

   Pick a page whose content you know — a signature page, a numbered form — and confirm it lands where it should in the preview.

6. 6

   Interleave and save

   Download the merged document. Spot-check page 2 and the final page, then get on with your life.

A PDF's pages live in a page tree — the document's table of contents for pages — and each page is an object that may be shared between documents by reference. Interleaving never touches the page contents themselves. The tool builds a new page tree whose order alternates between your two sources, copying page objects across with pdf-lib's copy operations.

That is why the output is lossless. A scanned page is an image stream with a size, a resolution, and compression settings; moving it into a new order changes none of them. Your 300-dpi page 47 leaves the output at 300 dpi, byte-for-byte the same image, merely standing in a different position in the queue. Interleaving is re-shelving, not re-printing.

The offset logic is equally unglamorous: when file A has n pages and file B has n-1, the alternation pairs A1-B1, A2-B2, and so on to A(n-1)-B(n-1), then appends A(n) alone. When B is short at the front instead of the end — rare, but real, usually from a mis-fed first sheet — the offset shifts the pairing so the alignment matches the physical original.

Because the work is object bookkeeping rather than image processing, it is fast and memory-polite. Two 200-page scans interleave in a couple of seconds on ordinary hardware. The browser does the parsing and the writing; there is no server, no queue, and no file size police other than your own RAM.

A note on metadata, because careful people ask: the output document carries a fresh structure assembled from the two inputs, so document-level metadata like a title or author field starts fresh rather than inheriting from either source. Page content is untouched; bookkeeping is rebuilt. If the merged file needs a proper title for records or filing systems, set it in whatever organizer you keep documents in — or number the pages and consider the document formally born.

The examples below are the three shapes this problem actually arrives in. Two are clean; one has a twist that the offset setting exists for.

Interleave is a scalpel, and the failure mode of every scalpel is being used as a hammer. A few situations that look like interleaving problems but are not:

• Two different documents. Interleaving a contract with a totally unrelated receipt file produces a genuinely comic and legally confusing result. Plain concatenation is what you want; that is the merge tool, not this one.
• Already-correct duplex output. Modern duplex scanners produce one file in the right order. If your PDF reads correctly, it does not need this tool — check by jumping to page 3 before you 'fix' anything.
• Two copies of the same document. Occasionally people scan a stack twice by accident and hope interleaving will 'unwind' it. It will not; it will dutifully alternate duplicate with duplicate. Splitting and deleting one copy is the actual repair.
• Insertion rather than alternation. When you need one page from file B inserted after page 12 of file A — a missed sheet, a re-scan — split file A at 12, then merge A1, B, A2 in sequence. Interleaving all of B would be wrong at scale.
• Fragmented page images. A torn sheet scanned in two pieces is a stitching problem, not an ordering problem. Different tool, different day.

The diagnostic question for all of these is the same: what order do the pages need to be in, and does that order alternate? If yes, interleave. If no, one of the neighboring tools — split, merge, rotate — is the right one, and they are all a click away.

Keep the table close during your first big run. Nearly every interleaving question that has ever been asked resolves to one of those four rows, and the five-page test tells you which one in under a minute.

Genealogists arrive first. Family letters, baptism records, and courthouse documents are one-sided by nature, and the person digitizing them usually owns a simplex scanner because duplex ones cost real money. Fronts and backs of a 90-page estate file is precisely the Tuesday this tool was built for.

Small offices and administrators follow: signed agreements scanned on the office's aging multifunction, printed forms collected at events, medical intake packets with printing on both sides. Students digitize handwritten notes and old lab handouts the same way. Accountants meet the problem during paper-heavy audits, and paralegals meet it whenever a client's 'PDF' turns out to be two files of alternating pages produced by exactly this workaround on exactly this kind of scanner.

There is a quieter group too: people doing a one-time digitization project — clearing a parent's paperwork, archiving a club's records, finally scanning the recipe box — who will use this tool once, intensely, and never again. That is fine. It is why there is no account to create and no trial to cancel: single-use rescue jobs are first-class citizens here, not leads to nurture.

The economics of these projects explain the tool's bluntness. A one-time digitization of a few hundred pages does not justify buying a duplex scanner or a document-management subscription; it justifies an hour, a simplex scanner, and a utility that solves the ordering problem in one click. When the project ends, the tool ends with it — nothing installed, nothing to cancel. The best tools for occasional rescues are the ones that leave no furniture behind.

If your scanner or office multifunction has an automatic document feeder with a duplex mode, the entire problem evaporates at the source. Look for a setting labeled two-sided, duplex, or 2-sided in the scan dialog, select it, and the machine pulls each sheet through twice — front and back — delivering one file in correct order. It is genuinely that boring once enabled, and genuinely common to discover it was disabled all along.

Vendor scan software sometimes offers its own combine fronts and backs feature, which promises exactly what this tool does. In practice those wizards are slow, occasionally reorder pages anyway, and almost always upload nothing — but they do demand their proprietary desktop app stay installed and updated. A browser tool that works the same on every machine, and leaves nothing behind, tends to age better.

When simplex is forced on you — fragile paper, a bound volume that will not feed, a flatbed with no stacker — plan the workaround instead of improvising it. Scan all fronts in reading order into one named file. Flip the stack the same direction every time (turning it left edge over right keeps the top edge consistent, which matters later). Test the whole pipeline on five sacrificial sheets before committing the real run. The five-minute rehearsal is the only part of this workflow that is ever optional, and it is the part nobody regrets.

For runs beyond a hundred pages, pre-flight checks stop being fussy and become cheap insurance. Everything on this list takes seconds; every item on it has a corresponding disaster story.

• Confirm both page counts against the physical stack. Fronts plus backs should equal your sheet count times two, adjusted for blank reverses.
• Run the five-page test with the exact flip technique you plan to use for the full run.
• Check that backs are right-side up and read in the same direction as fronts before touching the big files.
• Give the files unambiguous names — fronts.pdf and backs.pdf — because interleaving stack.pdf with newscan.pdf is how documents get mangled.
• Note a landmark page: the signature page, the odd-colored sheet, anything you can find in two seconds in the output.
• Never overwrite your source scans. The interleave writes a new file; let it.
• After saving, jump straight to your landmark page and to a random page in the middle before declaring victory.

Total elapsed time for all seven checks: under two minutes. Compared with re-scanning 180 pages because page 87 was a mirror image the whole way down, it is the cheapest insurance in the document world.

Notice how often a real repair chains two of these: rotate the backs, then interleave. Split, then merge around an inserted sheet. The chain is normal — documents fail in combinations, so fixes come in combinations. Every step runs locally and every step is free, which makes chaining a plan rather than an expense.

Interleaved material skews private. Estate files, medical packets, signed agreements, personal correspondence — the entire category is things people digitize for their own use, not for publication. Uploading two files of it to an unknown server so it can be re-ordered and downloaded back feels like hiring a stranger to shuffle your photographs.

So the tool does not do that. Both PDFs parse in your browser, pdf-lib performs the page-tree surgery locally, and the interleaved result writes directly to your disk. The network tab shows no file traffic, and the tool keeps functioning with Wi-Fi off — a pleasant property when your scanning project is happening at the archive, at a relative's kitchen table, or in an office with a strict data policy.

It also means there is no page cap invented to sell a premium tier, because there is no tier. The constraint is your machine, and your machine is better at this than most people expect: re-ordering page objects is cheap. The expensive parts of scanning were always the paper handling and the attention, and those are behind you by the time you get here.

The questions below cover the whole territory — scanner quirks, odd counts, and what the tool does with your files. For anything not covered, the five-page test answers it faster than any FAQ can.

Interleave is one station in a small assembly line. PDF Merge concatenates the documents that never needed alternating. PDF Split extracts sections or creates insertion points. Add Page Numbers to PDF gives the finished document citable order — and once a rebuilt file carries numbers, you will finally believe it is whole. All local, all free, all account-free.