Every attorney will eventually need to redact a document before filing it with a court. Social Security numbers, minor children's names, financial account numbers, dates of birth — FRCP 5.2 and state court rules require these identifiers to be removed or truncated. The process should be straightforward. In practice, it is the source of some of the most embarrassing and consequential mistakes in legal practice.
The core problem is that most attorneys use tools that create the appearance of redaction without actually removing the underlying data. This guide covers how to do it correctly.
The Most Common Mistake: Overlay-Only "Redaction"
The most frequent redaction failure is using a PDF editor's annotation tools — black rectangles, highlight marks, or text boxes — to cover sensitive text. These tools place a visual layer on top of the text, but the original text remains in the document and can be extracted by:
- Selecting the area and copying the text to a clipboard
- Using PDF text extraction tools
- Opening the document in a text editor
- Using accessibility tools that read the underlying text layer
This is not a theoretical risk. It has caused public redaction failures in federal court filings, high-profile corporate litigation, and government FOIA releases. If you are using any of the following methods, you are not redacting — you are hiding:
- Drawing black rectangles over text
- Using the highlight tool with a black color
- Placing opaque text boxes over content
- Using image editing to paint over text in a scanned document without removing the OCR layer
How to Actually Redact a PDF
Option 1: Adobe Acrobat Pro's Redaction Tool
Adobe Acrobat Pro includes a dedicated redaction tool that is separate from its annotation and drawing tools. The correct workflow is:
- Open the PDF in Acrobat Pro
- Navigate to Tools and select the Redact tool — not the Comment, Markup, or Drawing tool
- Select the text or areas to redact — Acrobat will highlight them in red to indicate they are marked for redaction but not yet applied
- Review all marked redactions to confirm you have selected the correct content
- Click "Apply Redactions" — this permanently removes the underlying text and replaces it with a black bar
- When prompted, also choose to remove hidden information (metadata, form fields, hidden text, comments)
- Save the file as a new document
The critical distinction is between marking and applying. Marking redactions highlights content for review. Applying redactions permanently destroys the underlying data. Many attorneys mark redactions but forget to apply them, leaving the text intact beneath red highlights.
Option 2: AI-Powered Redaction Software
Purpose-built legal redaction tools like RedactLaw automate the detection and application of redactions in a single workflow:
- Upload the document
- Select the categories of information to detect — PII, PHI, financial data, or custom categories
- Review the AI-detected items and confirm or adjust the selections
- Apply all redactions simultaneously
- Download the redacted document with all underlying text permanently removed and metadata stripped
This approach eliminates the most common failure modes: it detects identifiers that humans miss in large documents, it removes underlying text by default rather than overlaying it, and it handles metadata removal automatically.
Option 3: The Print-and-Scan Method (Last Resort)
Some attorneys — particularly those uncomfortable with PDF tools — redact by printing the document, using a physical marker to black out text, and scanning the result back to PDF. This method does work in the sense that the underlying text is not digitally present in the scanned image. However, it has significant drawbacks:
- Scanning degrades document quality
- The resulting document is not text-searchable unless OCR is applied — and OCR may re-create the redacted text if the marker did not fully obscure it
- It is extremely slow for multi-page documents
- It eliminates digital metadata but may not eliminate all readable text if the marker coverage is insufficient
This method should be considered only when no other option is available.
Post-Redaction Verification: The Step Everyone Skips
After applying redactions through any method, verify that the redaction is permanent before filing:
The copy-paste test. Select the redacted area with your cursor and attempt to copy and paste the content into a text editor. If any text appears, the redaction failed.
The text extraction test. Use a PDF text extraction tool or open the document in a text editor to search for strings that should have been removed. If the redacted text appears in the extracted content, the redaction is not permanent.
The metadata check. Open the document properties and review the metadata fields — author, title, subject, keywords, creation and modification dates, and revision history. Remove any metadata that contains or reveals redacted information.
The search test. Use the PDF viewer's search function to search for specific terms that were redacted. If the search returns results in redacted areas, the underlying text is still present.
These verification steps add less than five minutes to the process. Skipping them creates the risk of a public redaction failure that can result in sanctions, bar complaints, and client harm.
Handling Scanned Documents and Images
Scanned documents present additional challenges because the text may exist in multiple layers:
- The image layer — the visual representation of the page
- The OCR text layer — machine-readable text generated from the image
- Embedded metadata — EXIF data, scanner information, and processing history
Effective redaction of scanned documents requires addressing all three layers. Applying a black bar to the image layer while leaving the OCR text layer intact is functionally the same as an overlay-only redaction — the text is still extractable.
AI-powered redaction tools handle scanned documents by detecting text in the OCR layer, removing it, and simultaneously redacting the corresponding area in the image layer.
Batch Redaction for Large Filings
Single-document redaction works for simple filings, but litigation often requires redacting dozens or hundreds of documents for a single production or filing. Manual page-by-page redaction at this scale is not just slow — it is unreliable. Error rates increase with volume and fatigue.
Batch redaction tools allow attorneys to:
- Apply consistent redaction rules across an entire document set
- Process hundreds of pages in minutes rather than hours
- Generate audit trails documenting what was redacted and by whom
- Maintain consistency that is impossible to achieve with manual document-by-document processing
For any filing or production involving more than a handful of documents, batch processing is not a convenience — it is a quality control requirement.
Conclusion
Proper PDF redaction is not difficult, but it requires using the right tool for the right purpose. Annotation tools annotate. Drawing tools draw. Redaction tools redact. The distinction matters because the consequences of using the wrong tool — extractable "redacted" text in a public court filing — are severe and entirely preventable. Verify every redaction before filing, and consider whether your current tools are actually doing what you think they are doing.