PDF Redact

Cover sensitive regions with black redaction boxes.

What PDF Redact helps you do

Cover regions of a PDF with black boxes so the content underneath is no longer visible. Regions are entered as page, x, y, width and height in points, with the origin at the bottom-left corner. The boxes are drawn over the page, so this is a visual cover rather than a removal of the underlying text.

Redactions are drawn in your browser. Your document is not uploaded to Haivix.

How it works

A PDF stores positioned drawing instructions rather than flowing paragraphs, which is why editing one is different from editing a document. Page content, fonts, images and annotations are separate objects, and a page only carries a text layer if the producer wrote one. Most tools here rebuild or render the file in your browser with pdf-lib and pdf.js; the formats that need a full office engine, such as Word, Excel, PowerPoint, PDF/A and HTML, use the Haivix service and say so in their privacy note. Every tool writes a new file and leaves the original untouched. Because content is placed by coordinates, extraction often returns text runs in visual order rather than sentences, and columns or tables can arrive interleaved.

How to use PDF Redact

  1. Load the PDF and note the page numbers and positions to cover.
  2. Enter one region per line as page:x,y,width,height in points.
  3. Apply the redaction and check the count of regions drawn.
  4. Open the result and verify every sensitive area is covered.

When to use it

  • Hide personal or financial details before sending a document.
  • Black out confidential clauses in a shared contract.
  • Prepare a sample file with sensitive fields removed.
  • Cover a signature or account number in a screenshot.
  • Mask a name in a document that will be published.
  • Add visual covers to a fixed layout where coordinates are known.

When not to use it

  • Do not use it for truly sensitive data; the text under the box may still be extractable.
  • Do not expect it to remove metadata or attachments that contain the same data.
  • Do not use it when you cannot determine the coordinates; there is no draw-on-page UI.

Example

A region entered as 1:72,650,200,20 draws a 200 by 20 point black box on page one, 72 points from the left and 650 points from the bottom. The text underneath is still part of the content stream, so a copy-paste or text extraction can recover it. For real redaction, remove the content rather than covering it.

What you can do with PDF Redact

  • Draw black boxes at exact coordinates.
  • Target pages by number.
  • Apply several regions in one run.
  • Use point coordinates with a bottom-left origin.
  • Download the covered PDF.
  • Run the drawing locally in the browser.

Example: work on a copy and check the output

Keep the original file. These tools produce a download rather than editing in place, so a failed operation never damages the source.

Check whether the PDF has a real text layer before relying on extraction. A scan is an image, so text tools return little or nothing until OCR has run.

Unlock encrypted files first. A password-protected PDF cannot be reordered, cropped or annotated until it has been opened with the correct password.

Review the output before sending it on. Fonts can substitute, overlays do not always remove what is underneath, and page boxes can shift content near the edges.

Note the page size and orientation before laying anything out. A mixed document can contain portrait and landscape pages, and a tool that assumes one size may crop or offset the others.

Keep a note of the operation and its settings. If a recipient questions the result, reproducing it from the original is faster than guessing which options were used.

What to know before you start

  • The boxes are a visual cover; if the underlying text is not removed, it can still be recoverable, so use strong redaction when the data is truly sensitive.
  • Redaction does not remove metadata that may contain the same information.
  • There is no interactive drawing tool; coordinates must be entered manually.
  • Invalid page numbers or malformed lines are skipped rather than reported individually.

Common mistakes

  • Assuming a black box deletes the text underneath.
  • Using top-left coordinates when the tool expects bottom-left.
  • Entering millimeters instead of points.
  • Leaving the original file in a shared folder after redacting.

Troubleshooting

If a box appears in the wrong place, the origin or units are wrong.

If nothing is drawn, no valid region lines were parsed.

If the page number is out of range, that line is skipped.

If text is still selectable, the cover did not remove it; use a true redaction workflow.

Related standards and documentation

  • ISO 32000-2 defines the PDF format and its object model.
  • ISO 19005 defines PDF/A profiles for long-term archiving.
  • ISO 14289 defines PDF/UA accessibility requirements.
  • The PDF Association publishes technical notes on page boxes, annotation types and text extraction.
A copy, not an edit
  • Keep the source file.
  • Check for a text layer.
  • Unlock before editing.
  • Inspect the result.