Crop PDF margins and understand what stays hidden

You can crop a PDF to remove distracting margins from view. Cropping changes the area a reader sees, though, and can leave the material outside that area in the file. This one-page example lets you compare a retained cropped PDF with a copy whose full page view was restored.

See what the crop hides

Download the practice original and the cropped PDF. The page contains two synthetic text markers: PRIVATE MARGIN NOTE 73 near the left edge and INVOICE BODY farther inside. In the cropped view, the body marker remains visible and the margin note disappears.

Now open the restored PDF. It was made from the retained cropped file by resetting the visible page boundary. Both markers appear. The note came from content still present in the cropped file; restoring the view did not require replacing it with the practice original.

The text extracted from the cropped PDF contains both markers, too. A note missing from the page view, or from a viewer’s search results, is therefore not proof that it has been removed from this file.

Read the crop coordinates

A PDF page has a MediaBox for its full page boundary and a CropBox for the area presented to the reader. The example’s unrotated page has a MediaBox of [0, 0, 612, 792] and a CropBox of [200, 72, 540, 720]. The four CropBox numbers are the lower-left x and y coordinates, followed by the upper-right x and y coordinates, in points.

That makes the visible area 340 points wide (540 − 200) and 648 points high (720 − 72). Relative to the full page, the crop hides 200 points on the left and 72 points on each of the right, bottom, and top edges.

The two text markers were drawn at the same vertical position, y=400. The margin note starts at x=20, outside the crop’s left boundary of x=200. The body text starts at x=300, inside it. The left boundary hides the margin note while keeping the body text inside the visible window. pypdf’s cropping documentation describes these page boxes as a way to control the visible region.

The crop coordinates and the margins describe the same rectangle in different ways:

Edge	CropBox coordinate	Margin from the full page
Left	200	200 points
Right	540	612 - 540 = 72 points
Bottom	72	72 points
Top	720	792 - 720 = 72 points

If a dialog asks for margins, entering 540 as the right margin would confuse the right boundary’s position with the amount to trim. For this page, the right margin is 72. The MediaBox stays 612 by 792 points; the narrower 340-by-648-point view comes from the CropBox. The deliberately large left margin hides the example note. Choose margins from the content of your own document.

Crop a page in Acrobat

In Adobe’s documented desktop procedure, choose Edit, then Crop page. Draw a rectangle around the area you want to keep and double-click inside it. In Set Page Boxes, adjust the boundaries, choose the page range, and select OK.

Try one page first. Check its edges for content you meant to keep, such as captions, page numbers, annotations, or parts of a diagram, before applying the crop to a range. Pages with different sizes or layouts may need different boundaries. In particular, copying this example’s 200-point left setting to another PDF could hide useful material.

Cropping is suitable when the aim is a cleaner visible page. If the aim is to remove confidential information, the example shows why a crop is insufficient: its hidden note was still extractable and became visible when the CropBox was restored. A change in the saved file’s byte size would not, by itself, establish that the note was removed.

Check the supplied file locally

If you are comfortable running Python, you can inspect the cropped example without relying on a viewer’s search. Install pypdf with pip install pypdf, save crop-hidden-text.pdf in your working folder, and run this code there:

from pypdf import PdfReader, PdfWriter

reader = PdfReader("crop-hidden-text.pdf")
page = reader.pages[0]

print("MediaBox:", list(page.mediabox))
print("CropBox:", list(page.cropbox))
print(page.extract_text())

writer = PdfWriter(clone_from=reader)
writer.pages[0].cropbox = writer.pages[0].mediabox
writer.write("crop-restored-local.pdf")

The first three prints let you compare the page boxes and extracted text with the observations above. The final lines set the first page’s CropBox to its MediaBox and write a new file named crop-restored-local.pdf. The same restoration operation was run locally on the supplied cropped file: the saved output retained both text markers and restored the full page box. The code leaves the downloaded input unchanged.

Common crop questions

Can I undo a crop in a saved PDF?

For the supplied cropped PDF, resetting its CropBox to its MediaBox brought the margin note back because the note remained in that same file. Keep an original copy when editing your own PDF; the example does not establish that every cropped document retains recoverable content.

Does a slimmer-looking page mean the PDF was redacted?

No. The supplied cropped page looks narrower while text extraction still returns PRIVATE MARGIN NOTE 73. Its confidential-looking marker remains recoverable despite disappearing from the view.

Can I use the same crop on every page?

You can select a page range in Acrobat’s documented workflow, but inspect representative pages before doing so. A boundary that clears the margins on one layout may cut into content on another.

What if text extraction returns nothing?

pypdf cannot extract words from an image. A scanned, image-only page may therefore return no text while its contents remain visible. An empty text result alone does not establish that the page contains no hidden or visible information.

Source and test scope: The retained cropped output dates from September 29 and was parsed locally on October 7. The practice original was reconstructed on October 7 and is not the historical input file. The restored sample was produced from the retained cropped file; the Acrobat steps above come from Adobe’s documentation, with no recorded Acrobat interface trial.