Why is a rectangle drawn over text not enough?

A drawing can hide a name on screen while leaving the original text or image available underneath. This tool rebuilds the output without copying the source object graph, but you must still mark every occurrence and inspect unmarked content.

What does the output check establish?

The exported PDF is reopened and text in the marked regions is checked. A failed processing or final-verification step prevents download. This does not prove that you selected every confidential detail; review the whole document yourself.

Which pages retain selectable text?

Only supported simple text pages using standard fonts retain selectable text outside the marks. More complex supported pages are rendered as new images and listed in the result, including some unmarked pages. Those pages lose search and text selection and can change resolution or colour.

Can I redact a scanned page?

Supported scans are rendered to new images with the marked regions removed from the output pixels. The original image objects and OCR layer are not carried over. Check the visible result and scan detail because the flattened pages are no longer searchable.

What happens to comments, forms and attachments?

Source annotations, form definitions, links, bookmarks, attachments and document metadata are not copied into the new PDF. Visible information outside your marks can remain as pixels, so inspect rendered comments and form values too.

Does the tool locate every repeated name automatically?

No. You mark the areas yourself. Check headers, footers, tables, images and other repeated occurrences throughout the document before exporting and again before disclosure.

Can I revise my marks?

You can remove a box by clicking it or clear the marks before processing. The operation produces a downloaded copy and does not overwrite the source PDF.

Are document bytes uploaded?

The browser opens and rebuilds the document locally. The website's code and assets still have to load. Local processing does not replace a careful review of what you are about to share.

PDF

Redact PDF

Mark what must be removed, create a fresh PDF, and review the result before sharing it.

Private by design. Your files never leave your device

Drop a PDF here, or click to browse

PDF — the marked text is deleted, not covered over

Tip: Ctrl+V / ⌘V to paste

What this page helps you do

Create a fresh PDF with content removed from the areas you mark. Source document objects, attachments and metadata are not copied; the export checks marked text regions and lists pages converted to images.

How to use the result

Review every page, mark the complete sensitive areas, and read the export result. Search for repeated names or values and inspect headers, footers, annotations and other places where the same information may appear.

Before you rely on it

Reopen the copy and inspect every marked area; test search and copy-paste where text remains selectable.
Review unmarked content and repeated details visually, especially on flattened pages that can no longer be searched.

Limit to keep in mind: Only supported simple text pages retain selectable text. Complex pages, including unmarked ones, may be flattened with changes to resolution or colour. The tool does not find every sensitive occurrence; failed processing or verification refuses export.

Mark the information, wherever it appears

Review the entire document before placing the removal areas. A customer name can appear in an address block, a table and a footer; marking one occurrence does not deal with the others.

Search can help find text, but it cannot account for writing inside pictures or every visible detail. Use a page-by-page review suited to the information you intend to disclose.

Understand the new-document workflow

A successful export builds a fresh PDF. Supported simple text pages are recreated with marked text removed. More complex supported pages are rendered to new images with the removal areas applied to their pixels.

The source object graph, attachments, bookmarks, form definitions and document metadata are not carried into that fresh file. Information visibly rendered outside your marks can still remain, and some unmarked complex pages also need flattening.

Read verification and flattening notices

The finished PDF is reopened and text inside each marked region is checked with a different engine from the writer. A processing or final-verification failure prevents the download.

Only the supported simple-text route keeps selectable text. Inspect the reported flattened pages visually, since search and copy-paste no longer examine their words as text. Unsupported scaling or encryption can prevent processing.

Inspect the actual copy you will send

Open the downloaded PDF, check all pages and confirm that the intended recipient can still read the retained information. Flattening can change colour, resolution and file size, so review those as well as the removal areas.

Keep the original private. If you subsequently extract pages, merge files or compress the result, review that final output again rather than relying only on an earlier preview.

Related tools

Organize PDF

Extract just the pages you need to share.

Compress PDF

Shrink a PDF while keeping its text selectable.

Remove Metadata

Remove camera and location metadata from image files.

Word to PDF

Convert a .docx without uploading it anywhere.

Frequently asked questions

Why is a rectangle drawn over text not enough?: A drawing can hide a name on screen while leaving the original text or image available underneath. This tool rebuilds the output without copying the source object graph, but you must still mark every occurrence and inspect unmarked content.
What does the output check establish?: The exported PDF is reopened and text in the marked regions is checked. A failed processing or final-verification step prevents download. This does not prove that you selected every confidential detail; review the whole document yourself.
Which pages retain selectable text?: Only supported simple text pages using standard fonts retain selectable text outside the marks. More complex supported pages are rendered as new images and listed in the result, including some unmarked pages. Those pages lose search and text selection and can change resolution or colour.
Can I redact a scanned page?: Supported scans are rendered to new images with the marked regions removed from the output pixels. The original image objects and OCR layer are not carried over. Check the visible result and scan detail because the flattened pages are no longer searchable.
What happens to comments, forms and attachments?: Source annotations, form definitions, links, bookmarks, attachments and document metadata are not copied into the new PDF. Visible information outside your marks can remain as pixels, so inspect rendered comments and form values too.
Does the tool locate every repeated name automatically?: No. You mark the areas yourself. Check headers, footers, tables, images and other repeated occurrences throughout the document before exporting and again before disclosure.
Can I revise my marks?: You can remove a box by clicking it or clear the marks before processing. The operation produces a downloaded copy and does not overwrite the source PDF.
Are document bytes uploaded?: The browser opens and rebuilds the document locally. The website's code and assets still have to load. Local processing does not replace a careful review of what you are about to share.