E-discovery helps law firms to identify relevant content and provides information oversight in legal cases. E-discovery solutions automate the electronic discovery (e-discovery) process, which is outlined in the Electronic Discovery Reference Model (EDRM) diagram. They are essential to managing the ever-changing nature of e-discovery processes. E-discovery processes, triggered by a lawsuit or internal investigation, change as litigation progresses because the nature of the dispute and individuals involved are not constant. This article introduces the EDRM diagram, describes the e-discovery process, and describes available technologies that law firms can adopt.
EDRM was founded by attorney George Socha and technologist Tom Gelbmann in May 2005 to address the lack of standards and guidelines in the e-discovery industry. Currently, EDRM is part of Duke Law School’s Center for Judicial Studies. EDRM created the EDRM diagram (see Figure 1) to map the stages of e-discovery and as of 2018, is an industry-wide standard for managing the e-discovery process.
From the diagram, it is clear that e-discovery is an iterative process. This is due to the ever-changing nature of e-discovery, requiring law firms to constantly revisit information identified, preserved as litigation progresses.
According to Gartner, the first four processes (“left-hand side”), namely identification, preservation, collection and processing are IT-centric processes as they involve heavily technical data handling and analysis. Legal teams typically manage the last five processes (“right-hand side”). Successful e-discovery projects arise from all teams working in a collaborative and coordinated manner.
(a) Information Governance
Successful information governance involves conceiving interoperable processes. Law firm stakeholders must:
(i) Understand the value that business processes create;
(ii) Have basic knowledge of information technology that manages information;
(iii) Have oversight of legal, security, privacy and RIM risks; and
(iv) Be aware of the information lifecycle which lies at the centre of these processes.
Identification entails identifying potential sources of relevant electronic-stored information (ESI), which may include business units, people, IT systems and paper files. Key players should be interviewed to identify relevant information sources they possess. Additionally, IT and records management staff should be interviewed to establish how relevant data is retained, accessed and protected. In the litigation progresses, the relevance of information should be re-accessed and new relevant information identified.
Relevant ESI should be collected and stored in a legally-defensible, auditable, efficient, and proportionate manner.
ESI is collected in all forms and formats, ranging from e-mails to videos to compressed container files. Sometimes, they may even exist in physical storage devices such as tapes. Hence, ESI needs to be restored or processed before subsequent work can be done.
Processing comprises four sub-processes, namely, Assessment, Preparation, Selection and Output. Assessment filters out irrelevant data. Preparation involves converting the raw data into a format that allows for specific items to be searched and selected. Selection removes redundant and duplicate data, leaving behind only the most relevant of information to be output. Output finally prints the resultant data to be reviewed by attorneys or software and allows variances to be accounted for and if necessary, corrected.
Document review identifies responsive documents to produce and privilege documents to withhold. The review outcome is highly dependent on the legal strategy employed by litigators. It generally involves:
(i) Understanding the scope of the review;
(ii) Establishing procedures to manage reviewers; and
(iii) Identifying the right vendor, tools or platforms.
Analysis involves material fact-finding, identifying gaps in content, defining accurate search criteria amongst others. For effective analysis, legal teams should work with data analytics experts to best define search criteria that produces the most relevant results and materialise content loopholes for resolution.
(c) Production & Presentation
The large amount of ESI created and stored leads to increased research in how data that has been collected and reviewed is produced in civil litigation and regulatory investigations. Legal teams have to ensure the data produced is compliant with Singapore’s Evidence Law and is presented in a manner compatible with Singapore court procedures.
E-discovery technology automates and supports e-discovery processes by targeting and automating various stages of the ERDM.
As suggested earlier, data can be stored on a variety of platforms and formats. Forensic data collection technology automates the manual labour involved in gathering information that may be potentially relevant to an e-discovery matter, regardless of storage medium. Collected data should then be backup, tagged, and hashed to ensure data integrity before the information is subsequently used.
As discussed above, processing converts raw data from the various storage mediums to structured data that can be selected, manipulated and eventually printed in a format friendly to review software. Processing technology automates this complicated task, duplicating and converting large amounts of data within minutes to days, depending on the size of the data.
Legal hold entails preserving all potential evidence after receiving information record request from a government entity or reasonably anticipating an audit, investigation or litigation. In such scenarios, law firms have a legal duty to preserve ESI. ESI is preserved by holding information in-place (content held in original respiratory) or collection (content copied into and encrypted in secondary storage). System audit logs further protect data integrity. Besides the system itself, legal hold management processes include questionnaires, workflow and personnel tracking.
Predictive coding or technology-assisted review involves using machine-learning algorithms to reduce the number of non-responsive or irrelevant documents that need to be reviewed manually. It forms an integral part of Early Case Assessment, which aims to surface relevant ESI that may form potential evidence early in a legal matter.
The use of predictive coding has been approved judicially. In , a U.S. magistrate judge approved predictive coding to be used. This was similarly done across the Atlantic Ocean in the UK high court case of . With the increased adoption of predictive coding technology in the legal industry, it may only be a matter of time before Singapore courts follow suit.
Review technology supports legal practitioners in evaluating processed ESI for relevance and contextual privilege. Through a series of user-defined and computer-assisted workflows, practitioners can rate the relevancy of data and output readable datasets with proper redaction for the requesting entity. It is different from technology-assisted review software in that the latter uses machine-learning to perform reviews semi-automatically.
E-discovery software might be lowering the workload of lawyers, but it certainly is not intended to replace them. For example, while technology advocates might argue that TAR is inherently superior to human review, ultimately, a balanced approach might be better as e-discovery technologies merely magnify human judgment. However, what is there to magnify if no humans are judging? As U.S. Magistrate Judge Andrew J. Peck of the Southern District of New York puts it, “It’s a combination of the technology, the people involved and the workflow process.”