Redact PDF Online: Hide Client Details

Redact Client Details With the Tool

To redact a PDF, open the file above, mark the details to remove and prepare a redacted copy. All document processing runs in your browser. Your PDF and the details you enter stay on your device.

1. Choose an unlocked PDF and select Open PDF. Use the source preview to move through its pages.
2. Enter an account number, tax file number (TFN) or other repeated detail, one per line. Select Find and mark all to mark matching text throughout the PDF.
3. Inspect the marks on every page. Text matching removes whole text fragments, so a mark can cover neighbouring words in the same fragment.
4. For a scan or a missed detail, enter the area's position and size as percentages of the displayed page. Select Mark area on this page.
5. Select Prepare redacted copy. Visit every output page, confirm your inspection and download the PDF. Reopen the saved file before sharing it.

Area coordinates start at the displayed page's top-left corner. For example, left edge 10%, top edge 10%, width 30% and height 10% marks that rectangle on the current page. Leave a margin around each detail.

A page without searchable text displays a warning to use area redaction. This tool has no optical character recognition. A searchable page can also contain scanned portions, so inspect every page visually.

Matching ignores letter case and spaces. Punctuation still matters. Find and mark all replaces the previous text matches while keeping manually added areas. Clear this page's marks removes both kinds on that page.

The tool accepts one unlocked PDF up to 25 MB and 40 pages. It refuses forms, signed documents, annotations, attachments and layers. Complex or unusually large pages can exceed its processing limits.

Redaction vs Covering Text

Redaction permanently removes the selected content from the shared copy. Drawing an opaque rectangle over retained text can leave that text available to search, copy or extraction software.

Incorrect: place a black shape over an account number and save the original PDF objects underneath it. Correct: apply redaction and create an output that excludes the original content.

This tool renders pages at 144 dots per inch, blacks out the marked pixels and creates a new PDF using only those final page images. Original text objects and hidden text never enter the output.

The output also loses selectable text outside the redacted areas, along with bookmarks and document links. Inspect small print and figures at full size. Image conversion can increase file size.

Search the downloaded copy for a removed detail and try copying text from it. Both checks must return no original text. These checks cannot detect a missed detail still visible in an image, so inspect the pixels too.

What to Redact on a Bank Statement

Redact information the recipient does not need, while preserving the evidence required for their task. Keep the original statement separately and share a clearly named redacted copy.

• Remove a TFN when the recipient has no lawful need for it. The OAIC's TFN guidance explains its restricted collection and use.
• Hide full account numbers or card numbers when the recipient needs only a partial identifier. Keep any digits needed to match the statement to the application.
• Remove an address, client name or customer identifier from a copy used for an unrelated example or analysis.
• Redact unrelated transaction descriptions only when the recipient permits it. Leave evidence relevant to expenses, liabilities or income intact.

A lender can need the account holder, statement period, balances and transaction history to assess an application. Do not hide debts, spending or transfers to change the picture of the client's finances.

Follow the receiving lender's document requirements before obscuring identifiers or transactions. A redacted analysis copy and the evidence required for a credit application can need different information.

Redacting Before AI Uploads

Remove unnecessary client details before uploading a document to an AI tool. Redaction reduces the information you disclose, but remaining transactions or contextual details can still identify someone.

Data masking hides or replaces a value, such as displaying only the last digits of an account number. Its protection depends on whether the underlying value remains accessible.

Pseudonymisation replaces an identifier with a code, such as Client A. A separate matching record can reconnect that code to the client, so protect that record separately.

De-identification requires a broader assessment of whether someone is reasonably identifiable from the remaining information and its context. The OAIC's de-identification guidance explains why removing names alone can be insufficient.

Redact direct identifiers first, then inspect what remains. A distinctive transaction, employer or property address can identify a client without their name. Share only what the analysis needs under your brokerage's approved AI process.

Automatic text matching only finds the details you supply. It does not detect every type of personal information. This tool makes no PCI DSS compliance or freedom-of-information certification claim.

Redacting in Adobe, Word and Outlook

Adobe Acrobat Pro has a dedicated PDF redaction process. In Adobe's redaction instructions, open All tools, choose Redact a PDF and mark the text or areas. Apply the redactions, sanitise hidden information and save a new copy.

In Word, delete the sensitive text from a separate document copy before exporting it. Remove comments and resolve tracked changes so deleted material is not retained as a revision.

Run Word's Document Inspector through File, Info, Check for Issues and Inspect Document. Microsoft's Document Inspector guidance covers properties and hidden information. Inspect the exported PDF again. A black Word shape over text can preserve recoverable content.

For an Outlook message, print the relevant message to PDF, then redact that PDF before sharing it. Inspect sender and recipient addresses, quoted replies and signatures as well as the message body.

If you forward an email instead, delete unnecessary content from the new draft and remove original attachments. Microsoft confirms you can edit forwarded text and attachments before sending. Attach the inspected redacted copy only.

Remove PDF Metadata

PDF metadata can contain the author, document title and software used to create the file. Other hidden document content can include comments, attachments or earlier revisions.

Keep Remove document properties selected to exclude the original properties and the new file's software and creation-time properties. The output is rebuilt from page images, so original hidden objects and revisions are excluded.

If you clear that setting, the tool copies the original title, author, subject, keywords and software names when present. Those values can contain client details even after visible text is redacted.

Metadata removal does not erase a name printed on a page. Redact visible details separately, inspect the downloaded copy and keep its filename free of unnecessary client information.