There was a lot of talk this year at LegalTech about corporate legal departments bringing e-discovery in-house. The purported justifications for doing so are many, ranging from reduced costs and risks to increased control, efficiency and compliance. However, since the e-discovery lifecycle is very broad in scope, the real world benefits realized by bringing any particular components of that lifecycle can vary greatly from company to company. While most software vendors and resellers will have you believe that all your problems can be solved and your costs vastly reduced if you just implement their single magic solution, this is rarely the case. In order to realize actual returns, the decision to bring e-discovery in-house must be much more thoughtful and calculated, with a focus on actual needs, capacity and strategies. Deciding which aspects are suitable for internal management by any particular company requires a full understanding of the many different cost levers, the resources and expertise that each component requires and the pros and cons of doing it yourself versus outsourcing.
Where to Start
The early phases of the e-discovery lifecycle are primarily focused on identification and preservation of electronically stored information (ESI) that may be implicated in new legal matters. In fact, the expectation that companies have a repeatable and defensible e-discovery process to ensure the proper identification of ESI is now well settled in case law. These tasks cannot just be passed off to outside counsel. The preservation burden is a shared one.
To meet these obligations, it has become common for companies to implement internal processes and technology for tracking new matters, mapping ESI storage systems, issuing and tracking litigation hold notices and managing the many tasks that go along with these processes. Given the repeatable nature of these tasks across all legal matters, and the need to interface with so many different internal resources in order to be effective, it makes a lot of sense for companies to bring these processes in-house to manage themselves. Doing so also allows them to have increased oversight of counsel regarding e-discovery disclosures, pleadings, the meet-and-confer process and standardized objections relating to costs and burdens.
The latter phases of the e-discovery lifecycle, on the other hand, may be less suitable for companies to manage on their own. These include collection, processing and hosting as well as document review and production. The primary reason most companies have shied away from these latter phases is the simple fact that they require highly specialized tools and expertise. There are also issues of scalability, defensibility and economics.
Consider the collection of ESI. In order to properly preserve metadata and preserve the chain of custody for authentication purposes, the collection of ESI in legal matters requires forensically tested tools operated by trained persons with specific skills in the handling of evidence for legal disputes. These needs are only compounded by the number and diversity of systems that need to be collected from across a corporate enterprise. The in-house IT staff rarely have the proper tools and training to competently collect ESI on their own.
Even where a company’s staff can do this, there may also be strategic reasons for not wanting them to do so. These include not wanting to create fact witnesses out of internal employees by injecting them into the chain of custody. While they may be highly skilled at their jobs, they may not always be the ideal candidates for written or verbal testimony, should something go wrong with the collection process. There may also an issue with letting a “fox guard the henhouse,” which can arise where the very people accused of wrongdoing in the legal dispute are then tasked with preserving and gathering the ESI that is central to defending those allegations.
This isn’t to say that companies should never collect their own ESI. It may be valuable for a company to build these skills as an in-house capability. The point is simply that when deciding to do so, counsel must understand all the implications and requirements, and ensure that they are deploying the proper tools by the properly trained personnel. Otherwise any cost savings may be quickly overshadowed by increased spoliation risks. For these reasons, unless faced with a constant barrage of collection needs that warrants building an internal forensic team equipped with its own arsenal tools, most companies have chosen to outsource ESI collections to specialized service providers.
Processing and hosting ESI for document review present similar challenges, though here the primary obstacles are scalability and the large capital investments in technology that are needed. ESI processing involves taking the hundreds if not thousands of differing native ESI file types collected and ingesting them into a single software program designed to extract both the native file metadata and the textual content in a normalized and standardized format. It will often include the creation of static images of the documents, such as TIFF or PDF formatted files. As you can imagine, this all requires a large investment in infrastructure and highly specialized software. It also requires a team of specialists with significant training.
It is easy to process standard file types when all goes well. It is very complex and difficult to address exceptions and issues caused by nonstandard file types, corrupt, encrypted or password-protected files, malware and viruses and the hundreds of other issues that arise when things do not go as planned (which is always). Processing several hundred gigabytes of ESI for a case with a tight deadline can easily require an entire farm of the fastest virtual servers and a team of 10 to 15 highly trained people to fully complete the task in a few days’ time. There are also security controls and workflow processes that must be put into place, as well as overall project management and tracking processes. Given the complexity, it is very easy to make mistakes and have things go wrong, the consequences of which can be enormous. Just dropping one file from the workflow can call into question the integrity of the entire process.
After processing is performed, the extracted data, native files and/or static images of the original files are all exported for loading into a document review platform for further culling and review. Loading the data in itself is not terribly complex, and the hosting software is relatively light compared to processing. However, properly managing quality control in the review process and executing searches for data culling using keyword filters require specific expertise in search term development and testing as well as an understating of statistics and sampling. This is especially true if the aim is to deploy any of the advanced analytics functionality such as predictive coding or conceptual clustering.
Given the low proportion of cases that typically make it to the document review and production stages across the entire litigation portfolio, scaling is also a significant issue. If only a small percentage of cases reach this stage – and it is unpredictable how many and what size they will be – it is hard to predict what hosting resources might be needed in any given time period. Data privacy issues only compound this, as you also may not know where on the globe those resources will be most needed to meet cross-border privacy requirements. Furthermore, it may be hard to predict the logistics of dispersed outside legal teams, which also have very rigorous demands for reliable, fast and around-the-clock data availability, with zero tolerance for down time.
The Bottom Line
That these later phases of e-discovery are less suitable to bring in-house does not mean that this should never be done. In the right circumstances, with the right litigation footprint, technological resources and human capital, it may indeed make perfect sense. Though, as pointed out in AlixPartners’ fourth Annual Litigation and Corporate Compliance Survey, the majority of companies still deal with e-discovery on a case-by-case basis, and many still leave it entirely to their outside counsel. The trends seem to indicate that bringing the later phases in-house is only working for the largest global companies in highly regulated industries such as finance, pharmaceuticals and insurance. Even for most of these, a hybrid model of in-house collection and processing capability with some hosting for early case assessment along with service providers handling overflow and larger high-risk cases seems to be preferable to an end-to-end in-house model.
In order to see added value, it is important to choose a preferred provider who is truly a partner and not just vendor performing discreet tasks at a cost. To this end, per-unit pricing should not be the primary focus of the selection process, and in fact may be fully replaced with flat fee arrangements or an annual subscription model that is not unit-based at all. These types of alternative fee arrangements often look beyond per-unit metrics and provide incentives that are instead tied to efficiency and cost savings. While the overall pricing itself may be derived from measured historic metrics, there are also adjustments for efficiency improvements. Cutting last year’s budget by 30 or 50 percent may be a laudable goal. But it is a false gain if 50 percent of the work done last year was inefficient and unnecessary to begin with. It is far better to work with a partner who can recognize these inefficiencies and help with constant improvement as the relationship grows. Smarter workflows and processes result in higher cost savings and predictability than picking a low price vendor running antiquated workflows who is over-collecting and processing as a result.
This is especially true where the efficiencies are built in across matters and consider the whole picture of litigation across the company. For example, it is not uncommon to see situations where each new case is started from scratch without regard to the fact that much of the same data had already been collected, processed and reviewed in previous matters. In contrast, a true partner manages corporate e-discovery by leveraging a variety of cutting-edge data management techniques, analytics tools, cross-matter management technologies and managed review workflows that deliver a higher level of efficiency and service at a lower overall cost. As a result, they can also provide the cross-matter metrics needed for measuring improvements over time and continually lowering overall spend.
In summary, while it may be advantageous to bring some aspects of e-discovery in-house, it is rather rare for a company to want to bring all aspects of it fully under their roof. The most common model that seems to work for most companies is to focus efforts on the early phases. Building a repeatable, defensible preservation process that is consistently communicated and followed by outside counsel, and also provides them the tools and resources needed to consistently represent the company, not only reduces costs in this area, but it can also have a significant impact on downstream budgets as well. Companies with larger litigation portfolios may even see value in bringing some of the later phases in-house, especially collection for the more common systems in routine matters. However, before doing so a full needs assessment should be conducted that considers all the technical and business requirements. Only then can it be determined where to properly draw the line between internal capacity and partnering with a service provider who is dedicated to teaming with the company to build the best mousetrap possible.
Published April 15, 2017.