A multi-part series on the fundamentals eDiscovery practitioners need to know about the preparation and production of ESI
So far, in our five Fundamentals Series, we have covered the fundamentals of identification and preservation, of collection, of processing, of early case assessment, and of review. We turn our attention now to the final pre-trial phase of an electronic discovery effort: production.
The final thing eDiscovery practitioners must be able to do during discovery is to produce ESI effectively. This is a requirement that is almost always fulfilled in collaboration with an internal expert or an external service provider, but it is important for practitioners to understand the range of possibilities and their differing requirements, limitations, and implications.
Depending on what is negotiated or required, ESI production may be as simple as printing out a few files, or as complicated as custom load files with extracted text and redacted, Bates-numbered page images. Certain source types (e.g., relational databases) may require decisions about what reports or exports to generate and how to do so. How materials are produced affects how long they take to prepare and how easily they can be searched, reviewed, and used later in depositions and at trial.
Negotiating production format, including details like whether and what metadata will be provided, can both ensure maximum usability of what you receive and preempt disputes over what you produce and how you produce it. Failure to understand and negotiate effectively in advance still leads to frequent disputes today.
Beyond simply being important, the ability to successfully prepare and deliver productions of relevant ESI may also be an ethical requirement for attorneys to fulfill their duty of technology competence. For example, the California duty of technology competence for eDiscovery, which has been widely discussed as a useful model for all attorneys, explicitly names proper production of ESI as one of its nine core requirements: “produce responsive non-privileged ESI in a recognized and appropriate manner.”
The first production decision that needs to be made is the format or formats in which the relevant, non-privileged documents will be produced. That decision will determine the workflow that follows for actually preparing, validating, and delivering the production. Broadly speaking, there are four primary production formats available: paper, near-paper, native, and near-native.
In a paper production, the materials to be produced are printed out and produced as paper documents as in a traditional document production. Per-page Bates numbering, other endorsements, and redactions may be applied before or after printing.
While superficially simple, paper productions of ESI materials can still create technical issues. For example, review markup and other annotations in documents may not appear, auto-date fields may update to the print date, formulae underlying spreadsheet values may not be visible, and speaker notes for presentations may not be included. Moreover, some types of documents are not well-suited to standard, letter-size printing (e.g., spreadsheets, databases, media files). Additionally, paper productions are not searchable and do not include metadata, which may itself be relevant.
In a near-paper production, the materials to be produced are converted to image files that simulate printed, paper versions of the documents. Each page image can then have per-page Bates numbering, other endorsements, and redactions applied before production. Such image collections are paired with a load file that records document breaks, provides selected metadata for the documents, and potentially, includes extracted text for searching. This collection of images and related information can be loaded by the recipient into a document review tool like Relativity.
A near-paper production is a popular choice that combines some of the benefits of a paper production (e.g., per-page numbering, redactions) with some of the benefits of a native or near-native production (e.g., associated metadata, searchable text). It also retains some of paper’s drawbacks (e.g., questions of what’s visible, unsuitable document types) and creates some new technical issues of its own (e.g., time and cost of image creation, reconciling extracted text with image redactions).
In a native production, materials are produced in their native formats, as they are created and kept in the ordinary course of business. Such native file collections may also be paired with a load file containing extracted metadata, searchable text, and other information. This collection of native files and related information may be loaded by the recipient into a document review tool like Relativity.
A native production can generally be prepared with less time and expense than a near-paper production, and native productions eliminate questions of what’s visible in printouts or images and of what metadata is included. Native productions are not without drawbacks, however, as per-page numbering, other endorsements, and redactions are generally not possible. Moreover, some types of materials may not be reasonably usable in their native format (e.g., email databases, chat logs), and you must be very careful to ensure review of all metadata and hidden content before production. There is also some risk of inadvertent alteration of the native files by the recipient during their review.
A near-native production involves the conversion of native files into another electronic format that approximates the native format. For example, native instant message logs might be separated by conversation and converted into HTML or XML files, or only a static subset of a large relational database might be produced. Near-native productions are typically paired with a load file containing extracted metadata, searchable text, and other information. This collection of near-native files and related information can be loaded by the recipient into a document review tool like Relativity.
A near-native production carries similar advantages and drawbacks to a native production. Certain document types can be presented in a more useful way than they can with a true native production, but you also reintroduce questions of what’s visible, of how it’s presented, and of what metadata is included. Moreover, per-page numbering, other endorsements, and redactions are still generally not possible.
In reality, most productions today utilize a combination of near-paper, native, and near-native approaches, applying different handling to different document types to maximize later usability. For example, most documents in a production might be produced as near-paper images (to facilitate per-page numbering, endorsements, and redactions), with spreadsheets being produced as native files (to facilitate readability and formulae access), and with mobile device data being produced as near-native files (to facilitate readability and usability). Which combination of production formats is right for your matter will depend on the composition of your materials, your available technology, your negotiated agreements, and other factors.
Upcoming in this Series
In the next Part, The Nitty-Gritty and Other Reduplications, we will continue our review of production fundamentals with a discussion of other important format considerations, including load files, numbering, and redactions.