If you scan a book or a magazine flat on a scanner, you get two pages on every scan. That's fine for skimming but annoying for everything else: page 14 of the PDF is really pages 26 and 27, phones show it at half the size it should be, and printing gives you tiny text on a sideways sheet.
Split in half cuts every spread down the middle and gives you one page per page. A 3-page file becomes 6 pages. It runs in your browser, and the file isn't uploaded.
How the cut works (and why nothing gets blurry)
You might expect a tool like this to take a picture of each half and build a new PDF from the pictures. It doesn't, and that matters.
Every PDF page has a box that says which part of the page is shown, measured in points from the corner. Split in half copies each page twice, then sets that box on the first copy to the left half and on the second copy to the right half. The content itself, the scanned image or the text and drawings, isn't touched. Your viewer shows the part inside the box and ignores the rest.
So there's no re-rendering and no quality loss. If the original has real text, or a text layer from OCR, it stays selectable and searchable. Links and comments are kept on the half where they sit.
The flip side is that the hidden half is still in the file, just outside the visible area. For a book scan that doesn't matter. But don't use this to "remove" half a page you don't want anyone to see. Some apps may still find that text when you search or copy everything. For that, use Redact PDF, which really removes content (the guide to redacting a PDF properly explains the difference).
Splitting the pages step by step
- Open Split in half and drop your PDF on the upload box. A password-protected file asks for its password first.
- Next to "Split direction:", choose "Left / Right" for side-by-side spreads, or "Top / Bottom" for two pages stacked on one sheet.
- Look at the preview. The dashed line is where the cut goes. Use the arrows to check a few other pages.
- If the gutter isn't in the middle, drag on the page to move the line. The hint below shows the split as percentages, such as 52% / 48%, and "Reset to the middle" puts it back.
- Check the line under the file name, for example "120 page(s) will become 236 pages", then click "Split pages".
The result is a single PDF called yourfile-split-in-half.pdf.
The cut line sits at the same position on every page, anywhere between 10% and 90% of the page. If your scans drift (the book moved a bit between pages), pick a position that works for most of them and trim the rest afterwards with Crop PDF. If the spreads are visibly tilted, straighten them with Deskew PDF first. A straight cut through a crooked gutter clips a strip of text from one side.
Covers and pages that shouldn't be cut
Most book scans have a few pages that aren't spreads: the front cover scanned on its own, a fold-out, the back cover. Cutting those in half would be wrong.
"Automatically detect pages to exclude" is ticked by default under "More options". It looks at the shape of each page. For a left/right cut, a page only counts as a spread if it's wider than it is tall (for a top/bottom cut, taller than wide). Anything else is kept whole. A portrait cover scanned by itself is left alone without you doing anything.
The shape test can't tell everything, though. A wide cover, or a single page scanned sideways, looks like a spread. For those there are manual options:
- "First page" and "Last page" keep the covers whole.
- "Exclude pages" takes page numbers and ranges, like 1,8-10,13,15- (where 15- means page 15 to the end).
The preview marks excluded pages as "kept whole", and a line under the file lists them. If every page ends up excluded, the tool tells you instead of producing the same file again. With a portrait-only document and auto-detect on, you'll see a message that none of the pages look like spreads. Untick "Automatically detect pages to exclude" if you really do want them cut.
Booklets and right-to-left documents
Two more options cover the cases where cutting is easy but the page order comes out wrong.
Re-paginate from booklet scan is for a stapled booklet that you took apart and scanned sheet by sheet. A booklet isn't printed in reading order. In an 8-page booklet made of two folded sheets, the outside of the first sheet holds pages 8 and 1, its inside holds 2 and 7, and the second sheet has 6 and 3 on one side and 4 and 5 on the other. Cut that in half and you'd get 8, 1, 2, 7, 6, 3, 4, 5. This option puts the halves back into the order 1 to 8. It expects the scans in sheet order, outside then inside of each sheet, and every page gets cut, so the exclusion options are switched off while it's on.
Right to left document is for Arabic, Hebrew and other right-to-left books, where the right half of a spread comes first. Without it, every pair of pages ends up swapped.
If the PDF has bookmarks, "Discard bookmarks" also appears. Left unticked, each bookmark points to the first of the two new pages made from its old page.
When this is the wrong tool
Split in half changes pages, not files. If you want separate files (one per chapter, say), split the pages here first, then use Split PDF or Split by bookmarks on the result.
It also cuts every page at one position. A scan where some spreads are left/right and others top/bottom needs two passes, with the "Exclude pages" box doing the sorting. And if the two halves come out in a strange order that neither booklet nor right-to-left fixes, Organize pages lets you drag them into place.
Last thing to check: open the result and look at the first and last few pages. That's where covers and blank endpapers tend to catch people out.
Also useful: Deskew PDF · Crop PDF · Organize pages
