The eDiscovery Paradigm Shift: 5/1/11

Wednesday, May 4, 2011

Riding the Wave of Early Case Assessment

This past weekend, I finally emerged from a rather lengthy project and therefore I now have a bit of time to get back to my Blog. At the top of my list was the desire to comment on an article entitled, “Riding the Waves of Early Case Assessment” published on April 26, 2011 on the Law.com. The author, George Rudoy is CEO of Integrated Legal Technology, a legal technology and services consultancy, based in New York, does an excellent job of framing some of the issues surrounding Early Case Assessment (ECA) and adds some biting yet very true commentary on some of the early technology winners. My only real complaint with the article is that he didn’t solicit any input from me (just kidding). I guess that I will need to invite him to join my LinkedIn network and also join my Early Case Assessment Association Group (http://www.linkedin.com/groups?gid=1827179&mostPopular=&trk=tyah).

First of all, Early Case Assessment (ECA) is both a very clever marketing term that has no doubt propelled vendors such as Clearwell way beyond the point where they should have been had it not been for the term (BTW – Clearwell’s success will be studied at Harvard business school in the years to come as an excellent example of how to successfully position a product in an emerging market). On the other hand and as many have pointed out to me as I have ranted on about the fact that ECA is actually a technology, the practice of ECA (whether you call it that or not) has been around for many years and is nothing new to the experienced and blooded litigator. I concede that fact. However, I also believe that technology now exists (actually has existed) for many years that can in fact advance the practice of ECA. It is just not being applied yet to what I would call the eDiscovery subset of the Information Governance market.

I have been an enterprise technology junkie and serial entrepreneur for many years and therefore I have an excellent understanding of what technology exists and what it can and cannot accomplish. And, since entering the litigation services/eDiscovery market about five years ago full time, I have been continuously amazed at the lack of the “market’s” interest in utilizing very pertinent and established technologies from other enterprise types markets (e.g. Information Governance, BPM, etc.). In addition, I had the great fortune of working with Dr. Herb Roitblat several years ago and learned more about semantic search and related technology such as predictive coding than was probably necessary for any normal person (even a technology junkie such as myself). Having this knowledge base as a foundation, I have developed what would the market would consider to be a fairly radical concept of ECA and where it fits in the EDRM. However, after reading George’s article, I think that I may have some company.

Without going into a long drawn out description of where ECA technology fits in the EDRM, just let me state that I believe we should be following an iterative process (i.e. like the Agile software development model) where we are constantly analyzing the entire dataset to determine what we should be looking for, how we are doing, risk assessments, etc. And, today’s available technology lends itself to this approach. Further, as we move more and more ESI to the Cloud, this approach becomes even more practical, less expensive and much more productive. In the coming weeks, I hope to set aside some time to address my model in more detail. So, stay tuned.

The full text of George Rudoy’s article is as follows:

Can you really conduct early case assessment, or is it simply a fancy name for a more traditional indexing and culling model? What exactly is "early case assessment" (ECA)?

While working on a project for one of my corporate clients, I had to delve deeper into this question and reached out to a few of my old industry friends and colleagues -- Gyorgy Pados of Capital Legal; Richard Cohen, of Renew Data; Rick Lutkus and Lori Chavez of Seyfarth Shaw; Michelle Mahoney of Mallesons Stephen Jaques; and Jonathan Maas of Ernst & Young -- to gain a better perspective on the widely debated phenomenon.

As the novelty of the ECA concept starts to wear off and more and more practical evaluation and understanding develops, it’s time to reflect on the positioning of ECA on the Electronic Discovery Reference Model (EDRM) framework. Is it mature enough yet to be used in the best way -- and embraced? What capabilities are properly framed "ECA"?

The legal technology industry uses ECA in various ways. ECA could refer to appliance, software, or software as a service to quickly collect/index, de-duplicate, and search electronically stored information (ESI) to provide fast analysis of its content. The concept is a platform or service to facilitate a quick fly-over of the ESI collection to gather key reports of data make-up, information, and communication histograms patterns, concepts and themes, keyword analysis, and evaluations. Most ECA products provide dynamic probing and data-evaluation facilities to arrive at the relevant ESI, very much like the traditional fact-finding early-case-evaluation approach to assess the strengths and weaknesses of the case. But is this the right way to define ECA, or should it mean something different, such as analytic capabilities prior to collection or indexing of data?

The premise of early case assessment is to give legal teams the ability to conduct up-front, fast, intelligent data gathering, with probative queries on the dataset to reduce it to a relevant universe that can be assessed. Legal teams have a need to see "what they got" faster than what traditional EDD services typically can provide.

Let us consider two often-mentioned vendors that offer early case assessment: Clearwell Systems and Guidance Software.

Clearwell has done well in quickly turning the need for ECA into widely used capabilities both prior to and after collection, and while it is debatable whether the company was first to hit the market with it, it is hard to dispute that it was the first commonly used tool. Guidance has been in the e-data business for many years, and its EnCase eDiscovery product has been in the e-discovery market since 2002. While a little late to the post-collection review stage, Guidance's tool can be effectively used for identification, preservation, collection, processing and producing load files for review platforms.

So let's compare and contrast.

EnCase offers ECA functionality prior to, during, and after the collection on the basis of its Optimized Distributed Search ("ODS") technology, which allows IT administration to place servlet on individual workstations and gain complete visibility to the end-points without having to index or migrate data. A high percentage of its customers are using these capabilities in support of multiple types of investigations (inappropriate activity, remote intrusion, e-discovery), and it would be valid to characterize EnCase as a comprehensive general-purpose discovery and forensic engine. Its ability to reach every node in a network, including desktops, laptops, and servers, make it a popular collection and processing tool.

However, as previously mentioned, the tool has had analysis and first-pass review capability for less than one year, and it lacks full-blown review capabilities.

The Clearwell Legal Discovery Solution claims to have "90% data culling rates, and improved defensibility in court that addresses each stage of the e-discovery lifecycle from identification and legal hold through review and production, delivering significant cost savings and improved case outcomes". Clearwell boasts quite an impressive list of clients (a certain reward for being the first on the scene in terms of in-house review capabilities) and enjoys instant recognition when the topic of ECA in the review context is mentioned.

However, there is a price to be paid for beating everyone else to the punch. Clearwell is now facing stiff competition and is rumored to lose potential clients over its pricing as well as limited data volume per appliance. It doesn't offer metadata-only collections, which are valuable for pre-collection analytics. Defensibility is also coming into question, with recent customer satisfaction issues (NDLO and Microsoft cases) challenging the supremacy of the product. While well defended by Clearwell, such challenges widen the opening for other products to capture some of their market share.

While the debate on the usefulness and effectiveness of ECA continues, it should be noted that how the term is defined continues to evolve. Some of the service providers adapted quickly enough to have ECA as part of their "arsenal" a few years ago. Many of these tools were designed to filter metadata after collection and help the company decide how much a case will cost. This is a noble objective, but not completely in line with the original intent of ECA -- which was to help an organization determine its risk exposure and make strategic decisions about a case based on that analysis.

RenewData, which spent considerable time and effort researching ECA, analytics, and keyword-development techniques, took its time selecting a suite of ECA tools and chose not to go with Clearwell, Guidance, etc. The company's desire was to deliver tools that were more in line with the definition of ECA based on review capabilities, and it felt that what was already on the market wasn’t quite there yet. Instead, it focused on building two proprietary content analytics tools -- Vestigate and Anagram -- which can together help clients sift through large amounts of data quickly and relatively inexpensively, and make the strategic decisions that ECA promises.

Rich Cohen, president of RenewData, explains: "We have taken a unique approach to ECA, one that focuses specifically on the language of the matter, versus leveraging metadata filtering or complex mathematical formulas to simply churn through documents quickly. The result is a much more mechanical, transparent process that delivers far greater insight and greater certainty that you’re making strategic case decisions based on all relevant data available."

Considering the risks at hand, this could be a significant step forward -- to be able to determine your risk exposure early on in a case, based on what human beings are seeing in the language rather than point to a "black box" purchased off the shelf. Clearly, ECA capabilities both pre- and post-collection continue to evolve.