Bates Numbering

Bates numbering is the practice of stamping a unique, sequential alphanumeric identifier on every page of every document produced in litigation, regulatory proceedings, or formal document exchanges. The identifier — known as a Bates number or Bates stamp — enables all parties, counsel, witnesses, and the court to reference the same specific page of the same specific document without ambiguity. In complex litigation involving hundreds of thousands of documents, Bates numbering is the foundational document management convention that makes the production process tractable.

The standard format consists of a prefix of two to five letters identifying the producing party, followed by a seven-digit sequential number, padded with leading zeros: for example, ABC0000001 through ABC9999999. Variants exist by jurisdiction, court, and practice: some practitioners use date codes embedded in the prefix, some use shorter or longer number sequences, and some federal courts or government agencies specify their own format requirements. When multiple parties are producing in the same case, each party uses a distinct prefix to prevent numbering collisions.

The name derives from the Bates Manufacturing Company, a 19th-century office equipment manufacturer that produced mechanical rubber stamp numbering devices — the "Bates stamps" — that were the original technology for applying sequential numbers to documents before electronic document management existed.

Bates numbering is one of the least glamorous but most practically consequential aspects of litigation practice. Its absence or misapplication creates concrete, serious problems across the entire lifecycle of a lawsuit.

Federal Rules of Civil Procedure requirements. FRCP Rule 34(b)(2)(E) requires parties to produce documents in the form in which they are ordinarily maintained, or in a "reasonably usable form." While the rule does not explicitly mandate Bates numbering, case law and court practice have established Bates numbering as the standard implementation of "reasonably usable form" for document productions. Courts have sanctioned parties for productions that were so poorly organized or labeled that opposing counsel could not meaningfully use them — a disorganized native-file production without page-level identifiers is routinely cited as a discovery violation.

Court citations and deposition exhibits. In deposition practice, witnesses are examined about specific pages of documents. The only reliable way to ask "please turn to page 3 of the document marked as Exhibit 14" without confusion is if Exhibit 14 has a Bates number on every page. Attorneys who fail to Bates-stamp exhibits before depositions create transcripts full of ambiguous references that are difficult or impossible to use at trial. Federal appellate courts have noted cases where appellate records were muddled because trial exhibits lacked consistent page-level identifiers.

Trial exhibit management. At trial, exhibits must be presented to witnesses, opposing counsel, the jury, and the court simultaneously. Bates numbers allow all parties to confirm that everyone is looking at the same page of the same document. Trial exhibit lists reference documents by Bates range rather than document title for this reason.

Chain of custody and authentication. In evidentiary terms, a Bates-stamped production creates a chain of custody record: this production set, stamped with these numbers, was produced by this party on this date. Courts and opposing counsel can verify that the document they received is the same document as the one cited in a brief or at trial. Gaps or inconsistencies in Bates numbering can be exploited to challenge the authenticity or completeness of a production.

AI-assisted document review and Bates. When AI systems perform document review — coding documents for relevance, privilege, or key issues — the AI's output must be tied to specific Bates numbers. An AI that identifies a document as responsive to a particular request must do so by reference to its Bates number, so that the responsive document can be included in the production set with a verifiable record connecting the AI coding to the specific pages produced. Failure to maintain this Bates reference chain creates an audit trail gap that opposing counsel or a reviewing court may use to challenge the adequacy of the review.

The practical consequence of Bates errors — gaps, duplicates, inconsistent prefixes — is that they generate discovery disputes and motion practice. Resolving these disputes takes attorney time that is billed to clients or absorbed by the firm. In contested matters, opposing counsel will use Bates errors as leverage in sanctions motions or as evidence of a sloppy production.

How It Works (Technical)

In modern eDiscovery practice, Bates numbering is applied at the production stage rather than at the collection or review stage. The workflow is as follows:

1. Collection and processing. Documents are collected from custodians (email servers, hard drives, cloud storage) and processed by an eDiscovery platform to extract text, metadata, and file properties. At this stage, documents have internal system identifiers (document IDs) assigned by the platform, but not yet Bates numbers.

2. Review. Attorneys review the processed document population for relevance, privilege, and other coding categories using the eDiscovery platform's review interface. Documents are coded by their platform document ID during review.

3. Production set creation. When review is substantially complete, a subset of documents — those coded as responsive and not privileged — is designated for production. The production set is assembled in the eDiscovery platform.

4. Bates stamping and production. The eDiscovery platform applies Bates numbers to every page of every document in the production set. For native files (email, spreadsheets) that are being produced in native format, a Bates number may be applied to a placeholder page or slipsheet rather than to the native file itself, since native files do not have "pages" in the traditional sense. For documents converted to PDF or TIFF for production, the Bates number is stamped directly on each page.

5. Production delivery and log. The Bates-stamped production is delivered to opposing counsel, typically on an encrypted drive or through a secure portal. A production log (sometimes called a production transmittal) records the Bates range, the production date, and any redactions applied.

The technical challenge in large-volume productions is ensuring that the Bates counter increments correctly across the entire production set, that no numbers are skipped or duplicated, and that the Bates prefix is consistent throughout. Most eDiscovery platforms handle this automatically, but human configuration errors — entering the wrong starting number, using the wrong prefix, or running multiple production jobs with overlapping number ranges — are a common source of Bates errors.

For multi-party or rolling productions (where documents are produced in multiple tranches over time), the platform must be configured to continue the Bates counter from the last number in the previous production, not restart from 1. Failure to configure this correctly produces duplicate Bates numbers across tranches — a serious production defect.

How Legal AI Vendors Address It

Relativity is the industry standard eDiscovery platform for large-case document review and production. Its Bates stamping functionality is highly configurable: production administrators can set prefix, starting number, number of digits, and character encoding, and can run production quality control checks to verify Bates sequence integrity. Relativity generates a detailed production log for every production set. Limitation: Relativity requires trained administrators to configure productions correctly. The platform's flexibility is also a source of configuration risk — incorrect prefix or number settings are easy to enter and not always caught before the production runs. For law firms without a dedicated eDiscovery team or a Relativity-certified professional on staff, production errors are a recurring problem.

Everlaw is a cloud-native eDiscovery platform with a strong Bates stamping workflow. Its production module walks users through prefix and numbering configuration in a cleaner interface than Relativity, reducing the risk of configuration errors for less experienced users. Everlaw automatically generates a production summary showing Bates range, document count, and volume. Limitation: Everlaw's configurability in production — particularly for complex redaction and endorsement rules — is less extensive than Relativity's, which can be a constraint on very large or complex matters where highly customized production requirements must be met.

Logikcull is a self-service eDiscovery platform aimed at the mid-market: small to mid-size law firms, solo practitioners, and in-house legal teams that need to manage discovery without enterprise eDiscovery infrastructure. Its Bates stamping workflow is simplified and designed to be usable without specialized eDiscovery training. Limitation: Logikcull's Bates stamping features are less granular than Relativity or Everlaw. For complex productions with non-standard Bates prefix requirements or rolling productions across many tranches, Logikcull's simplified interface may not provide sufficient control, and practitioners may need to verify production output more carefully.

Casepoint is an AI-powered eDiscovery platform with strong production capabilities for government, enterprise, and law enforcement use cases. Its production module supports complex Bates stamping configurations and integrates with its AI document review and analytics layers, maintaining Bates references throughout the review-to-production chain. Limitation: Casepoint's UI for production configuration has a steeper learning curve than some competitors; users without platform-specific training may misconfigure production settings.

DISCO eDiscovery provides automated Bates stamping as part of its managed review and production workflow, with a focus on reducing manual production administration. Limitation: DISCO's production workflow is optimized for its native review environment; importing external review data and producing with Bates numbers assigned in a different platform can require additional configuration.

How Lawyers Should Verify and Apply It

1. Establish your Bates prefix convention at the start of every matter. Before the first production in any case, document the agreed Bates prefix and starting number in your case management file. If opposing counsel and your client have agreed on a production protocol, confirm Bates format is addressed. For multi-party litigation, confirm with all parties that prefix assignments do not overlap.

2. Configure rolling productions to continue from the last Bates number. For any case where documents will be produced in multiple tranches, configure your eDiscovery platform to continue numbering from the prior production's last number rather than resetting. Document the ending Bates number of each production in your matter file and verify before running each subsequent production that the platform is set to begin from the correct starting point.

3. Run a production quality control check before delivery. After generating a production set and before delivering it, verify: (a) the Bates range in the production log matches the expected count; (b) open several random documents in the production set and confirm Bates stamps appear on each page; (c) confirm no Bates numbers are duplicated within the production set by reviewing the production log. Most platforms support automated gap and duplicate checking — use it.

4. For AI-assisted review, confirm that review coding is linked to Bates numbers in the production output. When using AI document review tools, verify that the AI's relevance or privilege coding is exported to your eDiscovery platform in a format that preserves the link between coding decisions and specific Bates numbers. Audit a random sample of 50–100 documents to confirm that the documents produced under the AI coding decision are the same documents that the AI reviewed.