Salary data, bonus figures, and compensation details are among the most sensitive numbers in any business document. When an HR report, an offer letter template, or a payroll summary needs to be shared with stakeholders who should not see individual pay figures, redacting those numbers from the PDF is the correct and often legally required approach. Applying a black box or scribbling over the numbers in a markup tool does not actually remove the underlying text, anyone who knows how to copy and paste or inspect the PDF content stream can recover what was supposedly hidden. Proper PDF Redaction permanently deletes the selected content from the file structure, making recovery impossible through any normal means.
Why Drawing Over Numbers Does Not Work
The most common mistake people make when trying to hide salary figures is drawing a black rectangle over the numbers using a PDF annotation tool or a screenshot markup application. These tools place an opaque shape on an annotation layer above the text, but the text itself remains completely intact in the PDF content stream underneath. Anyone can open the file, select the area under the black box, copy the text, and paste it into a text editor to reveal the supposedly hidden figures. This is not a software bug or a security vulnerability, it is how PDF annotation layers are designed to work by specification. Annotations are non-destructive by design, intended to mark up a document without altering its underlying content.
The screen recording or screenshot approach is equally ineffective. Someone who receives a PDF with black bars drawn over numbers can simply open the file in a different viewer that renders annotations differently, or use a PDF inspection tool that extracts the text layer while ignoring the annotation layer entirely. In legal and corporate contexts, accidental disclosure through improper redaction has resulted in significant financial and reputational damage. Proper redaction uses a dedicated redaction tool that marks the content for removal and then permanently deletes it from the document structure when the file is saved.
Using a PDF Redaction Tool Step by Step
Open the PDF in a tool that supports true redaction. Browser-based platforms like WukongPDF include a redaction mode specifically designed for this task, and the same functionality exists in desktop applications such as Adobe Acrobat Pro. Look for a Redact or Mark for Redaction tool in the toolbar, not a drawing tool or a shape tool. Select the redaction tool, then click and drag to draw a rectangle over each salary figure or compensation number that needs to be removed. The tool highlights the area, usually with a colored overlay, to indicate it has been marked for redaction. At this stage, the content is still present in the file and can be recovered if you close the document without applying the redactions.
After marking all the figures, look for a button labeled Apply Redactions, Finalize, or Burn. This is the critical step that permanently removes the marked content from the document structure. Do not skip this step and simply save the file with the marks still visible as colored overlays. Marks alone do not remove anything, they are visual indicators of intent, not completed actions. Once applied, the redaction marks are replaced with black boxes or the appearance you configured, and the underlying text is irreversibly deleted from the content stream. The file size may even decrease slightly after application because the redacted text and any associated font data have been removed.
Redacting Patterns of Numbers Across Multiple Pages
Salary figures often follow recognizable patterns that a PDF redaction tool can search for automatically. Dollar signs followed by digits, numbers formatted with commas as thousands separators, and percentages near the words bonus or comp are all searchable patterns that appear consistently in compensation documents. Use the redaction tool's search function to find all instances that match a pattern, then mark them for redaction in a single pass. This is both faster and more thorough than manually scanning each page for numbers, and it eliminates the risk of human oversight where a figure in a footnote or a page margin goes unnoticed.
For numbers that do not follow a predictable pattern, work page by page using the thumbnail sidebar as a systematic guide. Scroll through each page in order, marking every occurrence of compensation data. Pay special attention to footnotes, appendix tables, and email signatures embedded in the document, since salary information sometimes appears in these locations outside the main body. A single missed instance in a footnote on page 47 of a 60-page report can be enough to breach confidentiality obligations and trigger notification requirements under data protection regulations.
Redacting Adjacent Text and Context Clues
Redacting the salary number but leaving the job title and department creates a situation where the remaining context narrows the possible salary range significantly. A redacted figure next to Senior VP of Engineering in the Boston office makes the approximate range deducible from publicly available industry salary surveys within a margin of fifteen to twenty thousand dollars. When redacting compensation data, consider whether the surrounding text needs to be removed as well. Job titles, years of experience, performance ratings, and location data all provide context that can be combined to triangulate the redacted figure with surprising accuracy.
The following table provides a decision framework for evaluating which context elements to redact alongside the raw numbers:
| Context Element | Redact? | Reasoning |
|---|---|---|
| Individual name | Yes | Directly identifies whose salary is shown |
| Specific job title | Usually | Narrows group, especially in small depts |
| Department (small, <10) | Yes | Small team makes individual salaries obvious |
| Department (large, 100+) | Optional | Large population provides anonymity |
| Location + title combo | Yes | Combined, these uniquely identify roles |
Verifying the Redaction Before Sharing
After applying redactions but before sending the file, verify that the content is truly gone using methods that simulate what an inquisitive recipient might try. Open the redacted PDF and select all text with Ctrl-A, then paste it into Notepad or another plain text editor. If any redacted numbers appear in the pasted text, the redaction was not applied correctly and the content is still recoverable. Search the document for the dollar sign, the word salary, and any keywords associated with compensation that might appear near figures you intended to remove. Click any visible black boxes and attempt to copy the area underneath by dragging the selection tool across the redacted region.
Check the document metadata as well. Some PDF creators embed information in the metadata fields that may include file paths containing employee names, department codes, or even salary ranges used as document titles. Use the document properties panel to inspect the title, author, subject, and keywords fields. Clear or overwrite any that contain identifying information. This extra step is frequently missed in real-world redaction workflows, and a metadata field with an employee name can undo the PDF Privacy protection applied to every visible page of the document.
Keeping an Unredacted Original Securely
Redaction is permanent and irreversible by design. Once applied and saved, the original numbers are gone from that copy of the file with no recovery path. Before starting the redaction process, save a copy of the unredacted original in a secure, access-controlled location separate from the copy you plan to share. This preserves the full data for authorized use while ensuring the distributed version does not leak it. The unredacted copy should be stored with the same level of access control as the original HR system or payroll database from which the data came, including audit logging of who accesses the file and when. For particularly sensitive compensation data, consider whether the document should exist in an unredacted digital form at all, or whether the redacted version should be the only version distributed outside the core HR team.
Redacting PDFs on a Schedule for Routine Reports
If your organization produces regular reports containing compensation data that must be redacted before wider distribution, building a repeatable process saves time and reduces error rates. Create a checklist document that lists every field and data pattern that requires redaction in each report type. Update the checklist when report formats change. Run the same checklist against every report before distribution rather than relying on memory or spot-checking, which inevitably misses an instance eventually.
Consider whether the redacted report should include a visible redaction log or certification. In regulated industries, some compliance frameworks require documenting what was redacted and why. Check your organizational policy or industry regulations to determine whether you need to maintain a record of redaction decisions. If a log is required, save it as a separate document rather than embedding it in the redacted PDF, since the log itself may contain references to the figures that were removed.
One additional precaution: after completing the redaction and verification steps, open the redacted PDF on a different device or operating system than the one used for redaction. A PDF that passes all verification checks on a Windows desktop might still reveal redacted content when opened on an iPhone or in a Linux PDF viewer that handles the document structure differently. Cross-platform testing takes a few extra minutes and provides an additional layer of confidence before the redacted document leaves your control.