Legal professionals interact with PDFs more intensively than almost any other profession. Contracts, pleadings, discovery documents, deposition transcripts, regulatory filings, and client records are all PDF-centric. Yet the PDF skills most lawyers have — open, read, print, sign — barely scratch the surface of what the format enables and what legal practice requires. This guide covers the PDF workflows that matter most in legal practice: electronic filing system requirements, digital signatures under the ESIGN Act and UETA, proper redaction (versus inadequate black-box coverage), Bates numbering for discovery, and metadata hygiene before filing or sharing sensitive documents. Each section addresses both the technical mechanics and the professional responsibility implications — because in legal practice, a PDF workflow failure can be a malpractice exposure.

Electronic filing: court-specific PDF requirements

Federal courts and most state courts require PDF filings. The requirements vary significantly by jurisdiction: **Federal courts (CM/ECF system)** — The PACER/CM/ECF system requires PDF filings with specific constraints: - Text-based PDFs only (not scanned images) for filings that contain selectable text — courts need to search, copy, and process filings - Maximum file size: typically 25 MB per document (varies by court) - Bookmarks required for documents over a specific page count in many districts - Hyperlinks permitted in many courts — internal links to exhibits, external links to cited cases - Fonts must be embedded **State court requirements** — Vary widely. Some states use their own filing systems; others adopt CM/ECF-based platforms. Check each court's local rules. Key things to verify: - Acceptable PDF version (most accept PDF 1.4+) - Whether PDF/A is required (some archival courts now specify PDF/A) - Page orientation and margin requirements - Whether searchable text is required **Signature on filings** — Most electronic filing systems accept the typed `/s/ Attorney Name` convention as an "s-signature" under the federal CM/ECF Administrative Procedures Manual. This is distinct from a cryptographic digital signature — it is a convention, not a technical signature. Actual cryptographic PDF digital signatures may also be required or accepted depending on the court and document type. **Best practice for court filings:** Generate PDFs from Word using the PDF export (not print-to-PDF) to ensure a proper text layer. Embed all fonts. Verify the PDF is not scanned by selecting text — if you cannot select text, the filing will fail or be rejected.

Digital signatures for legal documents

Two distinct signature types are used in legal PDF workflows — understanding the difference is essential: **S-signatures (typed /s/ convention)** — Widely accepted in US federal court filings and many jurisdictions for attorney-of-record signatures on pleadings. Requires no cryptographic infrastructure. Is a convention (an agreement among parties that `/s/ Jane Smith` represents a signature), not a technical guarantee. Provides no tamper evidence. **Electronic signatures (DocuSign, Adobe Sign, etc.)** — Platform-level signatures where the signing platform (acting as a trusted third party) certifies that a specific person signed at a specific time. Governed by the Electronic Signatures in Global and National Commerce Act (ESIGN) and the Uniform Electronic Transactions Act (UETA). Legally binding for most commercial contracts. The platform generates a PDF certificate of completion as evidence. **Cryptographic PDF digital signatures** — X.509 certificate-based signatures embedded in the PDF itself using the PDF digital signature mechanism. Provide tamper evidence (any modification after signing invalidates the signature) and identity verification (the certificate identifies the signer). Required by some regulatory contexts, financial filings, and cross-border transactions under eIDAS. For most transactional legal documents (contracts, NDAs, settlement agreements), a DocuSign or Adobe Sign electronic signature satisfies ESIGN/UETA. For regulated filings, notarised documents, or international contracts, consult the specific jurisdiction's requirements.

Redaction: the only approach that actually works

Legal redaction is a recurring source of high-profile failures. The 2019 Manafort filing (see our case study) and the 2006 AT&T/NSA case are two of dozens of documented instances where "redacted" documents revealed their hidden content. The only redaction method that permanently removes content is **permanent content stream redaction** — where the underlying text and image data is physically removed from the PDF file, not just covered visually. **Inadequate methods that create a security illusion:** - Black rectangle annotation over text — the text is still in the content stream, selectable by Ctrl+A - Black-filled white box placed over text — same problem - Covering with an image — text beneath is still present - "Whitening out" content — still in the file **The correct approach:** 1. **Adobe Acrobat Pro Redaction** — Tools > Redact > Mark for Redaction, then Apply Redactions. This physically removes the marked content from the content stream and replaces it with an opaque rectangle. The Apply step is irreversible and modifies the PDF structure permanently. 2. **Command-line: pikepdf or qpdf** — For automated redaction of structured data (account numbers, dates). 3. **Verify after applying** — Select all text (Ctrl+A), copy, paste into a text editor. The redacted areas should produce no selectable text. Try to select text directly over the redaction box. For discovered documents, US law under FRCP Rule 26(b)(5) requires proper redaction — inadvertent disclosure of privileged content due to inadequate redaction may be sanctionable.

Bates numbering: stamping discovery documents

Bates numbering is the practice of stamping sequential identifiers on each page of a legal document production — typically in the format PARTY-000001 through PARTY-NNNNNN. Bates numbers enable precise citation in depositions, briefs, and trial exhibits, and are standard in US federal litigation discovery under FRCP Rule 34. In PDF, Bates numbers are added as: - **Header or footer text** — Added to each page's content stream or as a text annotation. Adobe Acrobat's Bates numbering tool (Tools > Edit PDF > Header & Footer) supports this natively. - **Watermark overlay** — Added as a semi-transparent or opaque text stamp across each page. **Tools for Bates numbering:** Adobe Acrobat Pro: Tools > Edit PDF > Bates Numbering > Add — supports prefix, number of digits, start number, position, and font settings across a batch of PDFs. Command-line (PDFtk): ```bash # PDFtk doesn't do Bates directly; requires a script to apply # Common approach: generate watermark PDFs per page, then stamp ``` Python (ReportLab + pypdf): Generate a Bates-numbered overlay PDF and stamp it onto each page. Nuix, Relativity, and other e-discovery platforms handle Bates numbering automatically during document production and maintain a production log cross-referencing Bates ranges to source documents.

Metadata hygiene before filing or sharing

PDFs carry metadata that can expose information lawyers never intend to share. Before filing or sending a PDF to opposing counsel or a court, review and clean: **Document Information Dictionary fields** — Author (often the firm's internal document author name), Creator (the application that created the source document, which may reveal the law firm's internal software), and Subject may contain internal categorisation. **XMP metadata** — Extensible Metadata Platform data can contain document creation date, modification history, software version, and unique document identifiers. See our PDF metadata guide for the full list. **Tracked changes and revision history** — Word documents converted to PDF may carry XMP revision history entries identifying editors and edit times. Always use "Accept All Changes" and "Delete All Comments" in Word before generating the PDF. **Hidden layers** — PDF Optional Content layers may contain hidden content not visible in the default view but accessible to anyone who enables the layer. **Tools for metadata removal:** - ExifTool: `exiftool -all= document.pdf` removes all metadata - Adobe Acrobat Pro: Tools > Redact > Sanitize Document (removes metadata, JavaScript, hidden layers, and embedded content in one step) - QPDF: `qpdf --no-warn document.pdf output.pdf` preserves document content but clears some metadata For client-confidential documents, run sanitisation as a mandatory pre-send step, not an afterthought.
PDFs are mission-critical in legal practice, and the risks of misusing them — inadequate redaction, metadata exposure, invalid electronic signatures — are not theoretical. The workflows described here — proper court filing formats, legally sound electronic signatures, technically correct redaction, systematic Bates numbering, and metadata hygiene — form the foundation of professional PDF handling in a legal context. Each failure mode has documented, real-world consequences. Building these practices into a firm's standard document workflow, rather than relying on individual attention, is the professional responsibility standard.