Starting With More Data Shouldn’t Lead to Reviewing More Data

Posted on
trees shedding data

By Howard J. Reissner, Esq.

Yes, we are now starting with much greater quantities of raw data that is subject to discovery.

No, that does not mean that more documents need to be reviewed by attorneys.

What has been the common thread in eDiscovery over the past year? In a nutshell; the practical and economic necessity to substantially reduce the quantity of data requiring direct attorney review. Period.

The latest Amendments to the FRCP (effective December, 2015), had their origins in the realization that the ever-increasing amount of data now being created and stored was often skewing the legal process. The costs of the discovery process were becoming a central issue as to whether litigants would be denied the right to have their cases decided on the merits, and not denied the just, speedy and inexpensive determination of their case (Rule 1).

Always with the Best Intentions

The primary intent of the Amendments is to enable the discovery process to be faster, more focused, reduced in scope, and thereby less expensive – ultimately allowing cases to avoid being derailed by the costs and burdens of discovery.

Consistent with this theme is the effort to clarify document preservation requirements, allowing the routine deletion of data stored without the fear of sanctions at some point in the future. We believe that one of the practical implications of the Amendments, the accelerated number of days to comply with scheduling requirements, is having an impact on the eDiscovery industry that was not fully anticipated.

In essence, the reduced timeframes required for obtaining knowledge of a litigant’s document systems and data within their control has had an effect similar to the adoption of the 24 second shot clock in the NBA 60 years ago. There is a lot less time now to pass the ball around. The “scope” of the potential discovery for a case now needs to be addressed at the very outset of that case (if not before, when possible). Rule 26(b)(1) directs that discovery requests to be focused on non-privileged evidence that is relevant to a claim or defense AND (emphasis added) is proportional to the needs of the case.

Also, information necessary to support claims or defenses needs to be identified and understood pursuant to Rule 26 (a)(1) Initial Disclosure. The practical result is that discussions regarding the scope of discovery are taking place much earlier in a case. Outside counsel now need to be knowledgeable of the legal issues and relevant data under the control of their client to enable competent responses at 26(f) conferences and to early requests for discovery on every case.

M&A Activity Complicates the Data Landscape

To further complicate matters, the low interest rate environment in the global economy the past few years has facilitated a substantial amount of corporate merger activity. Low financing costs have made it less expensive than ever to buy, rather than internally develop business growth. This activity has created new corporate structures encompassing disparate legacy data management systems. From an information governance perspective, data increasingly resides in multiple locations, often with no common connectivity or searching capabilities.

The Ball is Moving Much Faster Now

What are some of the unintended consequences of the well intentioned changes to the Rules that have compressed the time periods for data identification, analysis review and production? I am willing to offer that they have blind-sided technology providers, litigants and attorneys who had been relying on tools and work-flow developed during a more leisurely, less complex, less cost sensitive era.

While the Rules were being amended over a period of several years, fundamental changes in technology was rapidly expanding the volumes of data created, the complexity and variety of information, and the decreasing costs of storage (e.g. cloud back-up). In essence, technology developments are far outpacing the ability of the law to keep pace. The law is always fighting the battle with the last war’s weapons.

Several years ago, it was generally accepted that organizations needed to implement information governance processes that disposed of information on a routine basis. The general concept that this is good corporate policy has not changed. Information that is not legally (or for good business purposes) required to be retained should be discarded. The practical effect will be to reduce the costs of locating, processing, searching and producing the information when subjected to a discovery request.

Storage Lockers Drop in Cost

Until quite recently, one of the main economic reasons supporting this concept was the high cost of storage (i.e. hardware, software, security, expert personnel). While the need to appropriately eliminate data to avoid unnecessary costs at the outset of future litigation has not changed, the acceptance of inexpensive cloud base storage has encouraged the current rationale of “let’s just keep it all because it’s a cheaper solution than figuring out what to discard and what to retain”.

To summarize, the corpus of information potentially responsive to discovery requests has skyrocketed, and time frames to respond have become compressed. Nonetheless, corporations are insisting vigorously to their outside counsel that due to much tighter budgets, the number of documents requiring attorney review should not be increasing. This has accelerated the emphasis on the utilization of technology earlier in cases to identify a more discrete universe of potentially responsive data.

It’s Important to Have the Right Tar for the Specific Job

So what is the most efficient, cost effective way of limiting the quantity of data that must be reviewed by attorneys?

At least a few times a week I see an article describing new methodologies of “TAR”. I consider the TAR designation a catchall phrase for various technology and work-flow paradigms that attempt to identify, cull and categorize information throughout the discovery process. Properly implemented, TAR workflows during pre-review can eliminate a large percentage of the non-relevant data, thereby allowing a more targeted, relevant subset of the data that ultimately is reviewed.

I am confident that in the right types of cases (and data sets), TAR methodologies such as conceptual search, categorization, clustering, “predictive coding”, and “continuous active learning”, have proven effective in substantially reducing the original corpus of information that requires human review. Additionally, in the latest emerging technology frontier, much progress has been made in utilizing Artificial Intelligence to identify relationships and threads in data sets that may not be apparent when viewed in a traditional linear manner.

But strip away all of the above TAR methodologies to their bare essentials; they are not effective if not designed and utilized in a manner that provides complete, accurate results in a consistent, defensible platform. As important, the technology and work-flow should be user- friendly and understandable to attorneys and their support staff.

For the majority of litigation cases, advanced TAR is not being utilized. That is because most attorneys do not wish to become technologists; an eDiscovery platform that can accurately process all of the data, provide detailed reports, enable an easy to navigate work-flow, and supported by knowledgeable people is the often unmet “gold standard”.

Implications for Attorneys, Litigation Support, and Service Providers

The practical realities of today’s economic environment are that corporations demand streamlined, efficient discovery, with as few documents requiring attorney review as possible, at the lowest cost. The challenge for service providers is to achieve this result without compromising on work quality (and completeness), and of course, client service. Industry service providers who did not anticipate the current environment and failed to adapt their work-flows, technology and personnel resources to meet these challenges are failing to meet these heightened levels of client expectations.


To conclude, the current challenge for e-discovery service providers is to offer highly efficient, easy to navigate technology solutions supported by expert personnel. The size and complexities of the raw data universe will only increase. Attorneys, most who are not technophiles, are being challenged to adapt to this new era, whether by choice or practical demands.

Easy to navigate, intuitive technology, combined with expert support, will for the vast majority of cases be the most efficient way to confidently reduce the quantity of documents requiring attorney review. This is good news for the market participants that have invested their resources in creating flexible, scalable technology while noting the increased importance of responsive, client support.


Leave a Reply