PDF crop boxes: visible margins and hidden content
Understand MediaBox, CropBox, TrimBox and BleedBox. Learn why changing PDF margins can hide page content without removing it from the file.
By ToolsNow · Published
A PDF crop box tells a viewer which part of the page to display. Changing it can hide margins while leaving the original content in the document.
Use the PDF crop tool to adjust page presentation. If you need to remove confidential information, use a redaction workflow and verify the exported file instead.
The five boxes
Each one is a rectangle in PDF user space, and they nest from largest to smallest.
MediaBox is the only mandatory one. It defines the physical extent of the page, the sheet of paper. Everything else lives inside it, and it’s what “A4” means in a PDF.
CropBox defines the region a viewer displays. When it’s absent it defaults to the MediaBox, so most pages behave as though it doesn’t exist. When it’s present, your reader shows only that region. This is the box that changes when you crop a PDF.
BleedBox is a print-production box: the area content extends to before trimming, which gives the guillotine a margin of error so a background colour runs cleanly off the edge instead of leaving a white sliver.
TrimBox is the intended finished page size after trimming, meaning what the reader ends up holding.
ArtBox marks the extent of the meaningful content, and gets used when placing a page inside another document.
For everyday work only the first two matter. The other three are print-shop plumbing, and most PDFs never set them at all.
What cropping actually does
Setting the CropBox changes which part of the page a viewer draws. The page content is untouched: every text object, every image, every vector path. The MediaBox is untouched. Nothing gets deleted.
Three consequences follow, and they’re worth internalising.
The file doesn’t shrink. The hidden content is still in there, so all you’ve saved is a few bytes of box definition. If you want a smaller file you want compression, which is a completely different operation.
Quality is untouched. Nothing gets re-rendered or re-compressed, so text stays selectable, images stay sharp, links keep working and the text layer survives. Cropping is one of the very few genuinely lossless PDF operations.
The content can be recovered. Reset the CropBox, which is a menu item in several editors, and the margins come straight back, intact.
Cropping is not redaction
This gets its own heading, because the mistake is common and the consequences have been severe. There’s a long history of published documents where sensitive material was “removed” by covering it with a black rectangle or by cropping the page, then recovered in seconds by selecting the text underneath or resetting the view.
If information has to be gone, it has to come out of the content itself, not just out of view. Your practical options:
- Delete the objects in an editor that rewrites the file, not one that draws over it.
- Use a dedicated redaction feature. Acrobat has one that removes instead of covering.
- Rasterise the page, which throws away everything not visible. The compress tool does this as a side effect, and the cost is your text layer.
- Take the content out of the source document and re-export.
Cropping is for presentation: trimming scanner margins, removing a header, fitting a page to a screen. It isn’t a security control, and the crop tool here says so on the page instead of letting the ambiguity ride.
Practical cropping
Scanned pages are the most common case by far. A scanner captures the whole platen, so a scanned book page arrives with a grey border and a skewed edge. Cropping that away makes the document far more readable on a phone, where every wasted millimetre costs you legibility.
Removing headers and footers for quoting or reading is straightforward. Just remember the text is still in the file, so a cropped page run through a text extractor will still give you the header.
Rotated pages need care. A page carrying a rotation flag displays turned, so the edge you see as the top isn’t the top in the file’s own coordinates, and a crop tool that ignores this trims the wrong edge. The tool here rotates the margins to match what you’re looking at, so dragging a crop on the preview behaves the way it looks.
Mixed page sizes turn up constantly in merged documents. Apply margins relative to each page’s own dimensions instead of as absolute positions, and 20 mm comes off every page whatever size it is.
What ToolsNow changes
To be precise about the tool on this site: it sets the CropBox on the pages you select and changes nothing else. The MediaBox, the page content, the page order, the text layer, links and annotations all stay exactly as they were. The result opens identically in any reader, and anyone who resets the box can undo it completely.
That’s the honest description of what PDF cropping is. Anyone who describes it as removing content, or as a way to hide information, is describing something the format doesn’t do.
Sources and further reading
- ISO 32000 / PDF specification — page boundary definitions (MediaBox, CropBox, BleedBox, TrimBox, ArtBox)
- PDF Association — guidance on redaction versus visual concealment
- pdf-lib documentation — the setCropBox API used by this tool
Last reviewed 31 July 2026.
More help with this topic
- DPI, pixels and print size: which resolution do you need?
Calculate image dimensions from print size and resolution. Understand why changing a DPI label differs from adding pixels, and how PDF export resolution works.
- How to merge, split and reorder PDF pages
Combine documents, remove blank pages, fix sideways scans and split a PDF into separate files with a visual workspace and undo.
- Scanned PDF vs text PDF: copying, search and OCR
Check whether a PDF has an extractable text layer, understand what OCR adds to scans and troubleshoot text that will not copy or search correctly.
Published by ToolsNow. Read how tools and sources are checked.
Found a mistake or an outdated detail? Send a correction with the article title and the detail to review.