Redact PDF v1.0

DocPivot Redact PDF is a browser based tool that permanently deletes sensitive text from a PDF instead of covering it with a black box. A drawn rectangle is only an annotation layered on top of the page, so the words underneath stay selectable, copyable, and extractable by any PDF library. This tool removes the content from the document itself, finds it for you by keyword search or automatic pattern detection, and runs entirely inside your browser, so the unredacted original never leaves your device.

What DocPivot Redact PDF Does

Redact PDF marks sensitive content, deletes it from the file, and proves the deletion worked before handing back the finished document. You load a PDF, mark what has to go, and the tool rewrites each affected page so the marked words no longer exist anywhere in the file structure. Nothing is uploaded, no job is queued on a server, and no API key is involved, because all rendering and rewriting happen in the browser tab you already have open. Scanned files are covered too, including the searchable layer a document picks up from OCR PDF.

Who uses it: legal teams preparing filings, HR staff sharing investigation records, finance teams circulating statements, healthcare administrators releasing patient records, and journalists publishing source documents. The common thread is a document that must be shared in full except for a handful of names, numbers, or addresses.

The problem it solves: a covered name is still a readable name. Two public examples show the cost. A 2019 court filing in the Manafort case exposed redacted passages the moment someone highlighted the black bars and pasted the text elsewhere, and in December 2025 readers stripped redactions from the Department of Justice Epstein release with nothing but copy and paste. The pattern repeats because visual concealment and data removal are two different operations, and DocPivot performs the second one.

Why DocPivot Redact PDF Deletes Instead of Covering

A PDF stores text as drawing instructions, not as a picture of words. When a shape is placed over a paragraph, the viewer paints the shape after the text, which hides the words on screen while every character stays in the content stream underneath. Select the area and copy it, or run a text extraction library over the page, and the hidden content returns intact.

Two operations actually remove data. The first deletes the text drawing instructions from the page, so the characters are gone while the rest of the page stays selectable and searchable. The second rebuilds the page as an image at 200 DPI with the black bars painted into the pixels, which destroys everything on that page including any invisible layer. Redact PDF uses the first method wherever it can be verified and falls back to the second everywhere else. If you ever want to audit a file you received from someone else, running it through PDF to Text is the fastest way to see whether the redactions hold.

Key Benefits of DocPivot Redact PDF

• Nothing is uploaded. The unredacted file stays on your machine for the whole session, which removes the awkward step of sending a confidential original to a third party before removing the confidential parts.
• Removal is verified, not assumed. Each marked area is tested against the untouched page before the file is written, and a page that cannot be cleaned precisely is rasterized instead.
• Sensitive data is found for you. Detectors flag emails, phone numbers, card numbers, national insurance numbers, bank account numbers, and IP addresses across the whole document.
• Scans are handled correctly. Scanned pages carry an invisible OCR text layer that ordinary redaction leaves behind, and this tool destroys it along with the visible image.
• Metadata does not survive. The output is built as a new document, so author, title, subject, and keyword fields are absent at byte level.
• No account and no cap. The tool is free with no daily limit and no watermark, available in eighteen languages.

Core Features of DocPivot Redact PDF

• Box marking. Drag a rectangle over any region to remove every text run it covers, which is the fastest option for addresses and table cells.
• Freehand brush. Sweep over irregular shapes such as handwritten signatures or margin notes, where a rectangle would take too much of the surrounding page.
• Search by term. Type a name, an account reference, or a case number and every occurrence across the document is marked at once.
• Run aware text matching. Search reads each page in document order rather than item by item, so a name split across two internal text runs is still found. That split is invisible on screen and is the single most common reason search based redactors miss a match.
• Six PII detectors. Switch on automatic detection for emails, phone numbers, card numbers, national insurance numbers, bank account numbers, and IP addresses.
• Checksum validated card matching. Card numbers are validated before they are flagged, so a sixteen digit order reference is not treated as a payment card.
• Mandatory review step. Every mark is a preview, and saving stays blocked while marks are pending, so the file cannot be written while you are still deciding.
• Overlay labels. Print a marking such as REDACTED or a statutory exemption code onto the bar, sized to fit the area.
• Metadata stripping. Document properties are dropped during the rebuild, a step Adobe places behind its separate Sanitize action inside Acrobat Pro.
• Rotated page reporting. Pages with a rotation applied are named in the results rather than skipped silently.

How DocPivot Redact PDF Works

1. Mark, search, or detect. Draw a box, sweep the brush over an irregular area, type a term to catch every occurrence, or turn on the detectors and let the tool sweep the document for you.
2. Review everything first. Marks are previews only, and the save action stays locked while any mark is pending, because until you commit the file still holds the original words.
3. Remove by the right method. Precise removal deletes the text drawing instructions from the content stream. Rasterization rebuilds the page from pixels at 200 DPI with the bars burned in.
4. Prove it or fall back. Each marked area is checked against the original page, and if any area covers something that is not removable text, the entire page is rasterized. A partial success counts as a failure here.
5. Save and download. The finished document appears on the result panel ready to download, with unmarked pages carried over untouched.

DocPivot Redact PDF holds the whole document in browser memory, so for very large files, splitting it first with Split PDF, redacting the section that matters, then rejoining it using Merge PDF keeps performance comfortable. Extract Pages from PDF works well when only a short exhibit needs releasing.

When to Use DocPivot Redact PDF

Reach for this tool whenever a document will leave your control carrying parts that must not travel: anything filed publicly, disclosed to an opposing party, published online, or forwarded outside your organization.

• Court and regulatory filings. Remove witness names, minors' details, and account numbers before a document enters a public docket.
• Subject access requests. Strip third party personal data from records before releasing them to the person who asked.
• Vendor and client sharing. Send a contract with commercial terms intact while pricing schedules or personnel details come out.
• Medical and insurance records. Take patient identifiers out of a report that still needs its clinical content.
• Financial statements. Delete account and card numbers from statements used as proof of address or proof of income.
• Published source documents. Protect a source before a document goes into an article or a research dataset.

Skip it when a document only needs to be locked rather than edited. If the goal is controlling who can open the file, Password Protect PDF is the correct tool, and if form fields and annotations simply need freezing in place, Flatten PDF does that without touching the text layer.

How DocPivot Redact PDF Chooses a Removal Method

The method is decided per page by what that page contains, not by a setting you have to understand in advance.

Scanned documents almost always land in the rasterization column, which is the correct outcome. If a scan has been through optical character recognition at any point, the searchable layer it gained is exactly the layer a careless redactor leaves intact.

DocPivot Redact PDF in Real Workflows

Law firm preparing an exhibit

Context: a paralegal has a forty page deposition exhibit that must be filed with witness names removed.

• Search for each witness surname and mark every occurrence at once.
• Review the highlighted matches page by page before committing.
• Save, then confirm the names return nothing in a text search of the output.

Result: the filing goes out with names deleted from the file rather than hidden behind bars, and the remaining text stays searchable for the court.

HR sharing an investigation summary

Context: a manager needs the findings of a complaint investigation without learning who raised it.

• Turn on email and phone number detection to catch contact details in the appendix.
• Brush over the complainant signature block, which is a graphic rather than text.
• Save and share the cleaned copy internally.

Result: the signature page rasterizes automatically while the body pages keep their text, and no unredacted copy ever reached an external server.

Finance team releasing a bank statement

Context: a statement is needed as proof of income, with account and card numbers out.

• Enable card number and bank account detection across all pages.
• Add a box over the sort code line that detection did not cover.
• Download, then run the result through Compress PDF if rasterization inflated the size.

Result: the statement proves income without carrying payment credentials into a third party inbox.

Journalist publishing a leaked contract

Context: a scanned contract will be published alongside an article, with the source protected.

• Mark the signature page and the header carrying the sender's details.
• Let the tool rasterize the scanned pages and destroy the OCR layer with them.
• Publish the output, which also carries no author or title metadata.

Result: neither the hidden text layer nor the document properties can identify the source after publication.

Honest Limits of DocPivot Redact PDF

Rasterization costs a page its text layer and makes the file larger. The tool warns you before you commit, and pages that keep precise removal are unaffected.

Detectors are deliberately conservative. A plain seven digit number is not treated as a phone number, and an unvalidated sixteen digit string is not treated as a card. A missed match is recoverable by searching for it, whereas a wall of false positives makes people switch detection off entirely, which is the worse failure.

Rotated pages are skipped by search and detection and reported in the results rather than passed over silently, because their coordinates and the page text coordinates disagree. Mark those pages by hand.

A hand drawn box removes whole text runs, so it can take neighboring words on the same line with it. The behavior over removes rather than under removes, which is the correct direction for this job but worth knowing before you draw a tight box in the middle of a sentence. There is also no batch mode yet, so documents are processed one at a time, and no text editing: if the goal is to change wording rather than delete it, Edit PDF Text is the tool for that. Pages that need reorienting first should go through Rotate PDF.

How DocPivot Redact PDF Compares to Other Redaction Tools

The distinguishing factor is not whether removal is permanent, since several mainstream tools do delete content properly. It is the combination DocPivot offers: no upload, verified per page removal, and automatic detection in a free tool.

• PDF24 processes files on its own servers and converts every affected page into a raster image when the document is saved, so removal is thorough but the whole page loses its text and the original still had to be uploaded. There are no usage limits or watermarks.
• Sejda removes content properly and keeps the rest of the page selectable, but files are uploaded, and the free web tier allows three tasks per hour with a 50 MB and 200 page ceiling.
• iLovePDF also uploads, states that files are deleted after two hours, and applies no daily task cap on the free plan. Its redaction tool caps at 400 MB and one file per task, and keyword search is available in its mobile app.
• Adobe Acrobat offers the deepest feature set, including pattern presets for phone numbers, card numbers, social security numbers, emails, and dates, plus a Sanitize step for hidden data. Redaction is available only in the paid Pro, Premium, and Studio tiers.

Pattern based detection is common in paid desktop software and rare in free browser tools, which is where this implementation sits. Metadata stripping is a separate manual step in Acrobat and happens automatically here. Once a file is clean, Watermark PDF can mark the released copy, and Remove Metadata from Photo handles the same hygiene problem for standalone images.

DocPivot Redact PDF Frequently Asked Questions