In the first of a six-part series, we outlined the purpose and core components of an e-discovery roadmap in the context of a corporate e-discovery readiness program. In part 2, we explain the e-discovery project model framework and discrete project lifecycle components that can have a measurable financial impact on the business. We also define terms and explain some of the metrics we use in our analytical exercises, which will be further addressed in later segments.
Many corporate e-discovery teams tell similar stories about their origins. With respect to the unnatural disaster, they can recall with great clarity all aspects of the scene. They commiserate with their comrades who shared that remarkable moment, when the case that no one expected, never thought would happen, was supposed to go away but didn’t, actually hit. In those days, the most proud moments were the result of managed chaos. Disbelief, confusion and incomprehension gave way to a numb acceptance of the scale, the complexity, the cost on the table. Experts hit the ground and commanded Orwellian technology that defied explanation. An entirely new language was discovered by some to have the miraculous effect of absolute incoherence upon repetition. And most bizarre, negotiating leverage appeared to be bestowed instantly on those who could illustrate complex mathematical algorithms with gumballs. It was a strange and extremely unconventional start to what might well now be a corporation’s most routinized, process-oriented legal operation.
For many corporations, similar experiences have prompted the introduction and subsequent evolution of a specialized e-discovery team, either adjunct to the legal department or within the technology/IT department. In-house e-discovery capabilities and subsequent strategic initiatives have generally focused on the acquisition of legal and technology expertise, implementation of discovery processes in-house (ad hoc or end-to-end) and the adaptation of sophisticated e-discovery technologies previously available only through highly technical third-party organizations.
Those that today “own” part or all of in-house e-discovery operations tend to report high confidence that their processes are successful because they save company money and avoid missteps that might otherwise result in costly sanctions. Most of these self-evaluations, while possibly true, remain nonetheless anecdotal. Is the hypothesis capable of testing? Is there room for improvement? How do you measure “success” in e-discovery operations? How do you know the value proposition between the “current” and “future” roadmap states?
In-house e-discovery operations can be objectively measured and a positive change in the quantifiable metrics, evidenced through our proposed e-discovery project model, can support the existing legal and business case and subsequent improvements to it. These same metrics may be repurposed in support of motions seeking to reduce the scope of discovery or to shift costs. Correctly deployed, the e-discovery project model can demonstrate, empirically, where the line exists between proportionality, on the one hand, and diminished returns, on the other.
The e-discovery project model frames the paths, procedures and protocols for the measurement of content analysis and production activities. An analysis of the corporate cost of e-discovery operations at discrete project lifecycle stages can enable continuous and strategic improvement of the e-discovery readiness program. Those same enterprise burden and expense run-rates will inform specific project forecasts and provide cost-per-volume and velocity benchmarks for project performance improvements. Given the recent attention to proportionality arguments for e-discovery cost containment, corporations should also look towards generating burden and expense forecasts based on benchmarked analyses of their e-discovery costs.
Project model management
Defining the “project” and the roles and responsibilities of legal and IT personnel is the first step in designing the management model. This is not always easy or straightforward. Anticipated litigation, existing litigation, third-party subpoenas, regulatory actions and internal investigations are likely to be managed by different legal or IT teams. Their differing policies and views may not align well with the proposed Project definition. Similarly, more distant stakeholders (e.g. data security, forensics, data services, and migration technology support units) are probably in competition for limited resources such as IT expertise and technology tools.
These challenges can be met successfully through establishment and maintenance of open communications and shared values between stakeholders. This is necessarily a give-and-take ritual, where the success of e-discovery project management will depend on centralization of a project pipeline, clear definition of authority — and support-roles at each project stage.
E-discovery project models vary from company to company, depending on internal and outsourced operational and technical factors. Some commonalities do, however, exist and that framework helps understand how they work and how to measure their performance. To that end, corporate e-discovery projects can be defined for purposes of cost analytics by their three segments: channel, scope and deliverables.
- Channel: Where data resides, how it is arranged and its format are important components that impact the costs and burdens associated with its identification, preservation, collection, processing, review and production. There are different data types. Unstructured data, for example, is unorganized, unrelated and occasionally non-machine readable (e.g. email, word processing, videos, audios, photos). Structured data, by comparison, is organized within a fixed record layout and extractable from a single flat file or relational tables (e.g. databases, spreadsheets). Archived data is a hybrid, where either unstructured or structured data is classified within a catalog and indexed for retrieval. Importantly for project management analytics, the costs, burdens and risks are different for each type of data, as are the tools and expertise required to extract and prepare the data.
- Scope: The legally permissible inquiry defines the scope of the undertaking. A document demand for one contract is much more limited in scope than a demand for “any and all” documents related to that contract. Depending on the channel, electronically stored relevant information “in scope” can be found by source (e.g., removable media, user workstation, network user file share); application or file type (e.g., email, user-created desktop files, system applications), and attributes (e.g., custodian, date range, content).
- Deliverables: The end-goal of an e-discovery project is to produce that which is sought, relevant and not privileged; in the form it was maintained. This end-product is called a deliverable. The cost of achieving deliverables is an important component to the cost reduction efforts. In example, a high dollar per deliverable ratio may indicate the scope was overly broad or that the model processed the data inefficiently.
Guidelines, procedures, templates and protocols may be developed and units of measurement established for the project lifecycle stages, which may include the following.
- Intake or initiation: First recognition of the project as a body of work, defines requirements of channel, scope and deliverables, and assigns resources.
- Preservation: Potentially discoverable data may be preserved in place or by collection, to ensure that content is excluded from routine disposition or inadvertent destruction.
- ESI Inventory: Electronically stored information within scope is typically inventoried by the custodian or information system/application from which the content is sourced and quantified by data volume.
- Scope refinement / cost allocation (proportionality): An effort to reduce scope or shift discovery costs to the requesting party typically based on a showing that the costs and burdens of pursuing the discovery sought outweigh the likely benefits to the requesting party of the effort. Put another way, a credible and persuasive argument demonstrating that the discovery effort demanded is “disproportionate” to the anticipate reward.
- Collection and chain of custody: Electronic content in scope for analysis and/or production may be collected through various methods depending on the specific project requirements, including self-collection by the custodian, custodian-directed collection, or selective or sweeping automated methods up to and including forensic collection. A chain of custody is established at points of collection and maintained for the electronic evidence through the project lifecycle.
- Analysis: While early case assessment and data modeling analyses for unstructured and archive content typically requires additional data processing and presentation in a review platform, structured data is often analyzed concurrently with quality assurance tasks following collection. Summary reports, data visualizations and gap analyses occur first in an analysis stage.
- Data processing and systematic culling: Any culling performed during extraction of the data, at the point of collection or post-collection, is generally defined by counsel to align with the fact pattern of the matter and legal strategy. Commonly, an external service may provide repeatable and consistent data processing and culling methods, including de-NIST, deduplication, date-ranging and other quantifiable attribute filtering.
- Search and retrieval: The identification of potentially responsive data within a collection is often efficiently accomplished by automated search and retrieval, based on language, numeric or other tokenized match, and which methods may include any of a number of assistive technologies and quality assuring protocols.
- Document review: The document review stage is generally considered the most time consuming and expensive phase for unstructured project deliverables and presents opportunities for creative review strategies to reduce costs and mitigate risk. As the efficient management of document review processes can have a direct financial impact to the corporation, we will address the functionality of efficient ESI technology and workflows in a future article in this series.
- Production workbooks: Production workbooks contain reports about the discovery processes — typically throughout its various stages — as required and defined by the contract with the vendor. Production workbooks help the data controller and counsel understand the status of document production throughout the discovery process and help to maintain workflow through the pipeline. Typical workbooks will contain detailed logs and status reports about most commonly recognizable discovery stages: identification, preservation, collection, processing, review and production. Production workbooks are generally considered protected from disclosure as attorney work product and, in some cases, attorney-client privileged communications.
- Closure: The final step in defining the e-discovery project model is to determine when a project work product is considered complete. Corporate records retention requirements and the continuing business value of work product may recommend a staged drawing down of e-discovery projects with substantial analysis, document review and production deliverables. The return of relevant work product to a central evidence repository may be designed to permit the strategic re-use and avoidance of inconsistent decisions on commonly harvested files and data types.
Identifying opportunities to improve
A corporate e-discovery readiness program requires a culture of continuous change and vigilance toward new precedents, rules, technological advances and operational improvements. Corporate strategic planning and capital investment cycles necessitate the analysis and definition of new initiatives at regular intervals in order to present business cases for investments in the e-discovery program. A standardized e-discovery project model will provide the corporation with the business intelligence to pursue cost containment and maintain a solid e-discovery readiness program.