Criminal Case Raises Interesting e-Discovery Search Issues

Once again, Judge John Facciola has authored an intriguing opinion on e-discovery, this time in a criminal case. United States v. O’Keefe, No. 06-249 (D.D.C. Feb. 18, 2008). The decision begins with criminal law specific issues, such as what discovery rules to apply, but quickly transcends them to address universal e-discovery questions. The most interesting issue was raised by the accused defendants who challenged the adequacy of the government’s keyword search. Judge Facciola held that this issue was too complicated for lawyers and judges to address without the aid of expert testimony. He implies that to do otherwise would be foolish, since he concluded that keyword search analysis is an area of e-discovery “where angels fear to tread.”

Obviously I am no angel, nor are thousands of other lawyers who must routinely deal with keyword search issues. Perhaps we are indeed foolish to run these searches without more help. As Alexander Pope said: “A little learning is a dangerous thing.” Here is the full quote from Judge Facciola’s opinion:

As noted above, defendants protest the search terms the government used. Fn6. Whether search terms or “keywords” will yield the information sought is a complicated question involving the interplay, at least, of the sciences of computer technology, statistics and linguistics. See George L. Paul & Jason R. Baron, Information Inflation: the Legal System Adapt?, 13 RICH. J.L. & TECH. 10 (2007). Indeed, a special project of the Working Group on Electronic Discovery of the Sedona Conference is studying that subject and their work indicates how difficult this question is. See The Sedona Conference, Best Practices Commentary on the Use of Search and Information Retrieval, 8 THE SEDONA CONF. J. 189 (2007) Given this complexity, for lawyers and judges to dare opine that a certain search term or terms would be more likely to produce information than the terms that were used is truly to go where angels fear to tread. This topic is clearly beyond the ken of a layman and requires that any such conclusion be based on evidence that, for example, meets the criteria of Rule 702 of the Federal Rules of Evidence. Accordingly, if defendants are going to contend that the search terms used by the government were insufficient, they will have to specifically so contend in a motion to compel and their contention must be based on evidence that meets the requirements of Rule 702 of the Federal Rules of Evidence.

For more on the law review article, Information Inflation, by Paul & Baron cited above, see my prior blog Information Explosion and the Future of Litigation. I have also previously written on the above cited Sedona Commentary on Search in The Myth of the Pharaoh’s Curse.

By the way, for those who have not read Rule 702 in a while, it says:

Rule 702 Testimony by Experts
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.

To my knowledge, O’Keefe is the first opinion to suggest that judicial review of alleged search deficiencies requires expert testimony. This may well portend a new type of expert in the future who opines concerning the search methods employed by parties to litigation. A whole new line of employment could open up for information scientists, e-discovery lawyers and others who specialize in the field of search.

The other issues in this case are interesting too, and shed some needed background on the search ruling. The defendant O’Keefe is a government official accused of taking bribes from co-defendant Agrawal for expediting visas to the U.S. for Agrawal’s employees. In a prior order, District Court Judge Friedman:

required the government to conduct a thorough and complete search of both its hard copy and electronic files in “a good faith effort to uncover all responsive information in its ‘possession custody or control.’” United States v. O’Keefe, No. 06-CR-0249, 2007 WL 1239204, at *3 (D.D.C. April 27, 2007) (quoting Fed. R. Crim. P. 16(a)(1)(E)).

The defendants received the submission from the government and responded with a motion to compel, complaining that the search was neither thorough, complete, nor a good faith effort as the court had required. The defendants sought to compel the government to provide them with detailed information on their search for both paper and electronic records, and to redo the searches.

Judge Facciola begins his analysis by noting that there is no rule governing criminal procedure as to appropriate format for production of documents. He finds it reasonable in this circumstance to apply Rule 34 of the Federal Rules of Civil Procedure. In his words: “it is far better to use these rules than to reinvent the wheel when the production of documents in criminal and civil cases raises the same problems.”

I am going to skip most of the paper search issues, although they are interesting, even from a civil perspective. Suffice it to say, the government just dumped boxes of paper documents with no folders or organization. That was not sufficient, and Judge Facciola ordered the government to meet with defendants and explain each document produced, the author, author’s title, recipient, date of creation and location of the document (where it came from).

The government opposed defendants’ motion to compel with an affidavit by a Visa Unit Chief which describes the search efforts that she and her five-member staff performed. There is no mention of attorneys participating in this search effort. All personal hard drives and active servers were searched, and the backup tapes going back two weeks were searched (they are erased after two weeks). Conspicuously absent is a description of other possible sources of ESI being searched, such as portable storage devices. Defendants do not, however, object to the limited scope of the search.

The only ESI search effort described is an automated keyword search. The Visa Unit Chief, whose qualifications to perform computer searches are not described, states that she used the following search terms: “early or expedite* or appointment or early & interview or expedite* & interview.” She also states that she had “[t]he Information Management Staff conduct[ ] the search of personal and hard drives because they have access to all drives from the network server, not just shared drives.”

Defendants argued that this ESI search was deficient, and did not comply with Judge Friedman’s order, for three reasons. First, the government did not interview the employees whose computers were searched. In other words, defendants accuse the AUSAs (Assistant United States Attorneys) who represent the government in this case of not discharging a key responsibility of the Zubulake duty, the duty to interview key players. See the Duty Page on the top of the blog that sets forth the responsibilities of legal counsel in electronic discovery. These widely accepted duties include the directive of Judge Scheindlin in Zubulake V to communicate with the “key players” in the litigation, in order to understand how they stored information. Zubulake v. UBS, 229 F.R.D. 422 (S.D.N.Y. 2004). I would add that such direct talks also allow you to understand the language the key players used so that you can more intelligently pick the words to use in a search.

Judge Facciola does not discuss this point directly, except in footnote six where he says that “if the search terms used actually captured everything there was to capture, such interviews would be unnecessary.” With respect, I think this misses the point, since interviews are a good way to develop effective search terms to begin with. When and if the search experts testify in this case, they will probably opine that it is impossible to develop search terms that will “actually capture everything there is to capture.” One hundred percent capture is not possible, not even close. Further, only by interviews and learning the linguistics involved can you hope to obtain the right keywords to improve the capture rate.

The second deficiency alleged was the government’s alleged failure to have any of the government employees search their own computers. Instead, the only search performed was a keyword search by the government’s search team. The employees know their own data the best. Why not ask them to search for it? You could also do a keyword search. How can a search be “thorough” and “complete” unless the custodians are involved in the search? Is that a “good faith effort to uncover all responsive information” as Judge Friedman ordered? Judge Facciola’s order does not discuss this issue, and I suspect that if the defendants renew their motion to compel, and this time support it with expert testimony, that the experts’ opinions will focus on this point.

The third deficiency is the government’s failure to disclose what software it used to conduct the search, and how it ascertained what search terms it would use. Defendants explained that they needed to know the type of software used to be assured that the metadata was properly preserved, and the search and collection were done according to standard forensic procedures. Now you can better see why Judge Facciola deferred ruling pending expert input, and otherwise ordered the parties to meet and confer, and attempt to resolve these issues through discussion with his active participation.

The defendants also alleged that the government failed to preserve ESI; but they did not offer any proof for that allegation, aside from paucity of production, which prompted Judge Facciola to say:

Defendants protest that there are inexplicable deficiencies in the government’s production of electronically stored information, but, as I have indicated in another case, vague notions that there should have been more than what was produced are speculative and are an insufficient premise for judicial action. See Hubbard v. Potter, No. 03-CV-1062, 2008 WL 43867, at *4 (D.D.C. Jan. 3, 2008). Accusations that the government purposefully destroyed what they were obliged to produce or knowingly failed to produce what a court ordered are serious. I must therefore remind the defendants of the wise advice given the revolutionary: “If you strike at a king, kill him.” If the defendants intend to charge the government with destroying information that they were obliged to preserve and produce pursuant to Judge Friedman’s order or the due process clause itself, they must make that claim directly and support it with an evidentiary basis—not merely surmise that they should have gotten more than they did. If they do not do so within 21 business days of this opinion, I will deem any such claim to have been waived.

Of course, the best way to obtain such evidence is through more discovery, a series of depositions actually. I wonder whether that is even possible in criminal cases. Still, you should not jump the gun and allege spoliation before you have proof. You should instead point to the sparsity of production as a justification for discovery on this issue.

Overall, in this opinion, Judge Facciola demonstrates a tolerance for the mistakes made by both sides, and a willingness to move on and look for practical solutions to the problem. Again, Alexander Pope said it well almost 300 years ago: “To err is human, to forgive divine.” Also recall Pope’s lesser known quote: “A man should never be ashamed to own that he is wrong, which is but saying in other words that he is wiser today than he was yesterday.”

That is an everyday reality in the fast changing world of e-discovery. It is especially true in the area of search, where newly developing forms of concept searching may soon make keyword searches obsolete. Judge Facciola mentioned this possibility in Disability Rights Council of Greater Wash. v. Wash. Metro. Area Transit Auth., 2007 WL 1585452 (D.D.C. June 1, 2007). For more information into this area, take a look at the West Legalworks CLE webinar I did with Jason R. Baron – Director of Litigation, U.S. National Archives and Records Administration; Doug Oard, Ph.D. – Associate Dean for Research, College of Information Studies, University of Maryland; and my co-chair of e-discovery at Akerman, Michael S. Simon. The 1.5 hour audio CLE is entitled The e-Discovery Search Quagmire: New Approaches to the Problem of Finding Relevant Needles in the Electronic Haystack and can be downloaded on demand from West. Doug Oard and Jason Baron and colleagues of theirs have been doing very interesting work in this area, including serious research concerning how keyword and Boolean searches measure against other types of search methods at the scholarly TREC Legal Track. Finally, you should also check out the excellent work being done on search by Anne Kershaw and her non-profit group e-Discovery Institute.

The enlightened attitude usually found in Judge Facciola’s e-discovery opinions is a product of his many years of experience, and his high level of expertise in this area. But be warned, very few judges today have this kind of deep understanding in e-discovery. Most have little or no judicial experience with e-discovery. To make matters worse, very few had any experience in this area before they came on the bench. In the sub-specialty of search, their experience is limited to Google, West or Lexis, which is completely different from e-discovery search in lawsuits and tends to give a distorted perspective. With no personal background in e-discovery, it is hard for many judges to appreciate the complexity and magnitude of the problems faced today by practitioners. For these reasons, and others, it is foolish to rush into court and expect the same kind of learned results from your judge. It is always better to try to work things out with opposing counsel, and only go to the judge for relief as a last resort.

e-Discovery Teams: Self-Organization and Development of Evidence Preservation Protocols

The preservation of discoverable ESI in litigation is a core activity of any e-discovery team. It is also a key component of risk management. Obviously, if you do not preserve electronically stored information, and it is deleted, then you will never be able to find it or collect it, much less review and produce it. Just ask the Bush White House about that; they failed to preserve over five million emails. The whole nine-step e-discovery process depends upon proper preservation. So too does risk management. Unless you are the White House, your failure to preserve after notice is a sure road to sanctions. Risk control in e-discovery begins with the routine employment of effective litigation hold procedures. This is the best way to minimize the chance of inadvertent or intentional destruction of relevant electronic records.

This is a difficult task in the best of circumstances. Intel’s email preservation losses in the AMD antitrust case demonstrate that. Intel was trying to implement a very complex litigation hold procedure to preserve relevant evidence, but despite strong efforts by its team, it lost thousands of emails. The loss was caused by a number of mistakes, including design flaws in the notice and collection procedures, and the failure to suspend an automatic file deletion program for certain key witnesses. Spoliation was also caused by the simple human error of forgetting to look at a second tab of an Excel spreadsheet listing more key custodians.

Even when no human errors are made, and the system design is near perfect, spoliation can still occur for a variety of reasons. For one thing, even though a suit may already have been filed, you still may not be able to determine what ESI is relevant and should be preserved, and what isn’t. Under today’s liberal notice pleading rules in federal court, it can be difficult, some might say impossible, to know exactly what ESI should be preserved and who should be notified of the preservation obligation. Pleadings can be so vague that the scope of a litigation hold is frequently a guessing game, especially at the beginning of a case when the duty to preserve is triggered.

The typical analysis to determine preservation scope is threefold: (1) what is the subject matter of the discoverable ESI that should be preserved; (2) who are the witnesses and custodians who may possess or control the discoverable ESI; and (3) what is the time frame of the discoverable ESI. In many lawsuits, it is not obvious from the pleadings what the dispute is really all about and what subjects could be relevant. In fact, most disputes usually morph a few times and develop new issues as the facts and law are better understood by the parties and the court.

This kind of subject matter and issue determination must be performed right after a suit has been filed. At this point, the defense lawyers may know little or nothing about the case beyond what is stated in a complaint. Even worse, the preservation duty may be triggered even before a complaint has been filed, when it is reasonably certain that litigation is likely. At that point, it frequently requires a crystal ball to try to guess all of the issues in a case. It is often just as speculative and risk filled to try to determine who the witnesses may be in the case, and what additional employees or third parties may be custodians of discoverable information, even though not direct witness. Even the time component may be vague, and you may not be sure how far back in time you should go, or how far forward.

You could say why not just call opposing counsel and find out, but that presupposes a knowledgeable and cooperative adversary. In reality, it rarely works that way. If and when they do return your call, which may be days or weeks later, and you are in fact given “advice,” it is often deliberately overly burdensome and oppressive, and you are sorry you asked. Gamesmanship is still very much alive and well in the adversary system.

In view of the importance of proper preservation to facilitate justice and avoid sanctions, and the inherent difficulties of guesstimating scope of discoverability at the beginning of a case, the development of good preservation protocols is a prime directive of every e-discovery team. Internal corporate teams must try to set up systems that will mitigate against these inherent uncertainties and risks. But does this mean it is the first thing a new team should focus on? Should the team begin its work by developing final protocols in this area? Should the top priority be development of procedures for rapid response to litigation by preservation notices, suspension of automatic file deletion programs, and the like?

No! The development of formal hold systems is important, but the first task of a team must be to self-organize. You have to have a team before you can play the game, much less win it. This means finding the right team players. It also means having them come together and attain a certain level of training and cohesion. Only then can they take on a play as complicated and important as litigation holds. The team must also secure adequate funding and senior management support for the team mission. It has to be a team of winners with a promising future, not a dead end. Only after these preliminary organizational steps have been taken can the team function effectively. Recruiting the right team members is more difficult than most people think. So too is securing upper management buy-in and budget approvals.

Team staffing requires careful selection of compatible people from three different sectors: law, IT, and management. The three-part nature of the ideal team is shown in the diagram to the left by IBM, with management referred to as “business.” IBM consultants, like almost everyone else in the field, advocate for internal e-discovery teams. The IBM white paper The Impact of Electronically Stored Information on Corporate Legal and Compliance Management, advises companies to:

Have a plan and a process for discovery of ESI that you can improve over time. Understand your end-to-end process from discovery to production and the implementation of “holds.” This encompasses methods and practices that make sense for your organization, understanding where technology is needed to facilitate or improve process efficiencies or quality of results, and identifying which specific technology capabilities are required to make your end-to-end process effective. It is best accomplished through a cooperative effort among legal, IT, and the line of business (LOB) organizations (see above figure).

You can have a team with just lawyers and techs. The first e-discovery team by Cisco in 2001 started that way. But today, most agree you should also have a management/business component in the core team. They can come from one or more of several different departments, including records management, HR, Finance, Risk Management, Compliance or Operations. There are many different variations. It all depends on the particular organization, its structure and corporate culture. IBM explains that the line of business members are needed to assist legal and IT to “set and manage the business priorities; establish the policies and best practices; and, enforce the organizational compliance.”

The culture of business, law and IT are very different. That is why careful selection of candidates from each sector is important. Not all in-house lawyers are cut out to work with computer techs. He or she may be a good lawyer, but computer-phobic. They went to law school precisely because they hated computers, math and science. The same applies to IT personnel. Many dislike lawyers, or are not comfortable with any kind of people activities. They would rather work on code for hours than attend a meeting. That is, after all, why most of them went into that field. Management types drafted into this project may have similar prejudices and see this as a dead-end assignment. They hear that it has something to do with records management, lawyers and IT, and their eyes glaze over.

So, believe me, finding the right people for an e-discovery team is not easy. You should not simply pick the people who appear to have some time right now to work on this. Their aptitude and cultural readiness is more important. Most companies do not have interdepartmental teams, so there will be little precedent for this kind of cooperative endeavor, and high turnover at first is not uncommon. For all of these reasons, significant time has to be spent on education at the beginning of the group’s existence. That is why my involvement in a team is usually front loaded, and my time lessens as the team self-organizes and becomes fully functional.

At the beginning, it is important to be sure everyone has a rudimentary understanding of the e-discovery process and the mission and future tasks of the team, including the need to design and implement good preservation protocols. These protocols and other projects will allow the team to better manage the high risks of litigation. They will also save the company boatloads of money, as I have previously blogged about, from reduced e-discovery costs.

The first members of a team typically consider whether additional team members are needed. There should be at least two representatives from each of the three sectors, and frequently more than that, depending on the size of the organization, its history, and the complexity of the organization. A team needs enough members to accomplish its goals, but not so many as to become cumbersome and unworkable. If a team must have many members due to the complexity of the organization, then it usually breaks down into sub-groups. When that happens, the initial education, communication and cohesion process becomes much more challenging.

After a team is fully formed and operational, and everyone is fully briefed and understands its missions and upcoming activities, one of its first tasks is to address litigation hold and collection procedures. Still, at this point, you are not yet ready to design final preservation and collection procedures. For most teams, that is still a year or so away. Instead you create interim protocols.

Some companies have some kind of identification, preservation and collection system in place, even if it is not in writing. Typically, it involves preserving computer files by telling employees to look through their computers for relevant files, and when found, save a copy onto a central location, usually a server. When there is already some kind of system like this in place, the team begins by studying the current procedures and looking for areas that need improvement. Usually existing systems fail to cover all ESI, fail to preserve metadata, fail to authenticate with hashing, and are otherwise of questionable legal validity. They are also usually not well documented, hard to follow, and fail to address many common contingencies.

More often than not, there are no protocols at all. Lit-holds have only been dealt with on an ad hoc basis by a number of different people, each with their own ideas on how to go about it, and what the law requires. There you basically start from scratch. But whether you already have written procedures in place, or are designing them for the first time, in either case you are only going to be able to create interim preservation procedures. Final procedures come much later in the work of the e-discovery team.

The procedures, even the interim procedures, must comply with the current laws. For instance, it is probably not legally sufficient to simply send employees an email telling them not to alter or destroy relevant files, and then hope for the best. Much more follow-up is required. Otherwise, if a mistake is made, and ESI is destroyed or altered, the company could face severe sanctions. See Cache La Poudre Feeds, LLC v. Land O’Lakes Farmland Feed, LLC, 2007 WL 684001 (D.Colo. March 2, 2007) and my prior blog on this case. For this reason, as a best practice, many companies are now moving to automated systems that supplement employee compliance. Also, as Zubulake V teaches, reasonable lit-hold procedures should include direct interviews with the key players in litigation. Zubulake v. UBS Warburg LLC, 229 F.R.D. 422 (S.D.N.Y.2004). Also see the Duties blog page above.

The hold procedures, even the interim procedures, must also be designed to cover all potentially discoverable information maintained by the organization, even the PST files that some pack-rat employees may have all over the place. Sometimes the preservation obligation may include back-up tapes, sometimes not. It depends on the computer systems and the particular case. Most of the time it will not be required. The hold and collection procedures must also be capable of preserving at least some of the ESI metadata. In some instances, it may not be acceptable to have employees copy files to a centralized repository, as that act in itself will change the file date metadata.

As this work begins, the team will become painfully aware of a number of deficiencies in existing systems, including problems with permitted computer use policies, retention policies, and computer systems. There is usually a lack of information about exactly what information the organization has, and where it is all located. [It is, in effect, a lack of system metadata, data-about-data.] Most companies have only a vague idea of what they know, and who or what knows it. Few have complete, up-to-date knowledge of where all of their electronic information is stored. If you do not know what information you have, nor where it is all located, how can you possibly preserve it, much less produce it as required by law?

This common deficiency must be addressed early on by the team. They must take a complete inventory of existing ESI and map it. Then they need to classify the ESI accessibility according to Rule 26(b)(2)(B). When the team looks, they always find that the organization has far more information stored in its systems than management realized, and that much of it is unneeded and should have been destroyed long ago. So then the team works on rehabilitating the organization’s overall electronic records management, and moves on to destruction of unneeded ESI according to Rule 37(f).

The many deficiencies in existing systems, coupled with the stringent requirements of the law, end up forcing the team to design hold procedures that are difficult to follow, time consuming, expensive, and disruptive. That is because they must work with what they have, and what they know. Frequently it is chaotic. For instance, each employee may or may not have their own email archives, PST files. These files may be located on desktop PCs, laptops, thumb-drives, or CDs kept at home. Some employees may have multiple PST files, all of which now need to be searched and protected from alteration. Some may have none, but they may use their personal email accounts for work from time to time. The process of studying current practices and computer storage systems, and trying to design standard hold procedures that will preserve all discoverable ESI, will inevitably highlight the need to change existing systems so that better, more efficient, hold procedures can apply.

The team will not want the stop-gap procedures to be permanent. They are, after all, the people who must follow these protocols and implement the lit-holds. Most teams will recommend adjusting future IT purchases to include preservation criteria, typically adding or improving archiving and indexing software. It may also mean purchasing new software specifically designed to implement and manage litigation holds. The process usually requires a radical overhaul of existing document retention policies and practices, and a tightening of permitted computer usage.

After the new policies and systems are in place, the team can then design a final protocol for preservation and collection. It will be far better than the patchwork program first developed, and far less likely to fail and expose the company to sanctions as in the Intel case. It will also be far easier for the now matured team to operate, and to realize significant cost savings in the subsequent steps of ESI collection, analysis, review and production.

When and Why Should You Start an e-Discovery Team?

Should every organization have an e-discovery team? If a company has only a few computers, say less than 100, or rarely gets sued and has less than 3 lawsuits going at a time, then it does not need one. It is probably cheaper to just hire outside counsel and vendors to handle e-discovery cases when they occasionally arise. But for everyone else, especially in these times of cost-cutting budgets, it is a necessity. That is because, as my friend Ed Foster likes to say, it can save your company boatloads of money! It saves on outside counsel fees and e-discovery vendor costs. It also saves on inflated settlements to avoid the expense, hassle and risks of e-discovery. Yes, a corporate e-discovery team does cost some money to setup, but as Benjamin Franklin said: “An ounce of prevention is worth a pound of cure.”

Today millions of British pounds and U.S. dollars are being spent on e-discovery cures. This expense is a harsh reality of litigation. So too are the risks, mistakes and sanctions that frequently occur when a company puts its fate in the hands of ill-prepared legal counsel or profit-motivated vendors. Even in the best of circumstances, it is all too easy for a medium or large size company to burn through a few million dollars in e-discovery expenses in just a couple of months. See Kentucky Speedway, LLC v. NASCAR, 2006 U.S. Dist. LEXIS 92028 (E. D. Ky. Dec. 18, 2006). The solution is for a company to take control of its information and its e-discovery activities, and not ignore the problem or over-delegate it to others. This means forming an internal e-discovery team.

But don’t just take Ed Foster’s word for it. Ask the people and companies who have already formed e-discovery teams. For instance, ask Jeff Ghielmetti of Cisco Systems, who has more experience at this than anybody. Jeff often tells the story of how Cisco established the first internal corporate e-discovery team after the stock market crash of 2000-2001. At that time, Cisco was hit with a flood of litigation, often involving millions of pages of electronic documents. One of the first cases came with a $23,500,000 bill for e-discovery! Cisco could not continue at that kind of burn rate, so out of necessity they decided to try something new, and go in-house. For help they turned to Jeff, a Cisco engineer, not a lawyer, but he had the full backing of Cisco’s forward-looking (a/k/a “desperate”) legal department.

Under Jeff Ghielmetti’s direction, Cisco turned away from the traditional model of hiring law firms and e-discovery vendors to manage their data preservation, collection and analysis, and they started to do it themselves. Cisco set up the first internal, multi-disciplinary corporate e-discovery team. It also developed its own proprietary software and computer systems to help the team perform its tasks. Jeff reports that Cisco’s program has been a huge success, not only in cutting costs, but also in better management of risks. According to Cisco’s case study, “Cisco has been able to reduce its discovery costs by approximately 97 percent-for an overall reduction in litigation expenses of 64 percent.” Those are pretty impressive numbers. For detailed information on the Cisco team and the centralized Storage Area Network (SAN) it developed for e-discovery, see How Cisco IT Uses SAN to Automate the Legal Discovery Process.

If you don’t want to take both Ed and Jeff’s word for it, then you could also ask Laura Kibbe. She is the young attorney who helped Pfizer form its first e-discovery team in 2005. As a result of her work, Laura was named one of four of Corporate Counsel Magazine’s trailblazers in 2006. The magazine article on Laura summed up the benefits of her work:

Why pay millions for an in-house e-discovery system when, as most companies do, you can outsource the work? Simple — farming it out to law firms and consultants, all of whom really want the job, can cost even more. Especially when you are an oft-sued drug company like Pfizer. In just one year, Kibbe says, the new system saved “multiple millions of dollars,” and served up several side benefits. She gave Pfizer a “repeatable process” that can find and extract information quickly.

Pfizer GC Allen Waxman is proud of what Kibbe has accomplished in just under two years. Her response team, Waxman says, has reaped “substantial efficiencies and added effectiveness.” . . .

Probably the best thing Kibbe brings to the tech table is the fact that she’s a lawyer who has to use the system. That means that she doesn’t need to rely on outside firms, who sometimes make promises to produce data that their client can’t keep. This way, Kibbe can make sure that what Pfizer tells one court it can or can’t do remains consistent in all courts.

You could also ask Kevin Esposito, the Pfizer attorney who is the current director of its e-discovery team. I have written about him before as one of the lead speakers at the ABA 1st Annual National Institute on E-Discovery. He is adamant that the key to significant costs savings in the long run is to spend the time and money up front to develop an effective e-discovery team. He states this requires careful mapping of all a company’s data, and establishment of sound processes and procedures for preservation, identification and collection.

You could also ask Jonathan Eisenberg of Merrill Lynch. He co-heads its Global Litigation department and supervises its e-discovery team. Jon realized that Merrill Lynch should begin to take e-discovery in-house when he saw how many of his experienced outside counsel were becoming ineffective in securities cases because of the complexities of e-discovery. The attorneys were experts in securities law and litigation, but not in e-discovery. Merrill Lynch responded and now has a well-developed team that performs most of the e-discovery work itself, instead of relying on its outside counsel or vendors.

Jonathan describes an eight-step process that Merrill Lynch developed to manage e-discovery:

1. Established a full time in-house interdisciplinary team of legal and IT professionals dedicated to electronic discovery, i.e., formed and funded an e-discovery team.

2. Established detailed litigation hold procedures that were later refined and improved over hundreds of cases. Jonathan describes this key process as having seven steps: a) ID the custodians; b) inventory and map the data sources, which takes many months the first time through; c) send written notices and reminders to all impacted employees; d) monitor and enforce compliance with the lit-hold; e) interview the key players; f) collect the discoverable information; g) export data for production to outside counsel. To that list, I would add the imposition of automated software holds and suspension of usual document retention procedures.

3. Rewrote and implemented an aggressive document destruction policy. In my experience, this is a very challenging and time-consuming process, especially if a company does not already have an effective electronic document retention policy in place.

4. Purchased and refined custom search software to quickly locate electronically stored information across all of their systems that might be relevant to a particular case.

5. Developed a system and software tools to process all ESI and filter out all duplicates and other obviously non-responsive data before export to vendors for further processing and analysis.

6. Purchase and use what Jonathan calls Encase-type software tools to image and search for non-email ESI.

7. Located and trained a pool of reliable contract lawyers to review pre-productions.

8. Trained several of their IT technicians to be forensic experts.

If Ed, Jeff, Laura and Jonathan’s references were not enough, you could also ask James Wright. He is the project manager who leads Halliburton’s team. Jim has a power-point online that lists some of the key points on e-discovery teams. The presentation, which I have heard and previously written about, begins by summarizing some of the benefits of taking e-discovery in-house.

Why should Corporations participate in E-Discovery?
1. Corporations own all the risk of litigation and pay all the costs.
2. It’s just not a paper world anymore: data volumes changed everything.
3. Corporations know their E-data better.
4. E-Discovery mgmt is inconsistent among law firms.
5. Law firms risks & incentives differ from corps: law firms have a more conservative methodology towards discovery; a de-facto disincentive exists for data reduction.

The fifth point made by Halliburton’s team captain is two-fold. First, law firms are naturally more cautious is making decisions and will always err on the side of including more data in the preservation, collection and review set. Secondly, there is a “disincentive,” which Jim in person bluntly calls a conflict of interest, for law firms where attorneys are paid by the hour to review documents they decide may be responsive. He implies it’s like hiring the fox to guard the hen house. Vendors are of course in the same position, since most charge by the megabyte of processed data.

And if all that were not enough, you could also ask Patrick Oot, the young but senior counsel at Verizon who leads its new e-discovery team. Patrick says that:

Managing electronic discovery is probably a number-one initiative at all legal departments. Depending on the volume of litigation an organization has, this can be one of largest line items in your [legal] budget.

Patrick reports that his team approach is keeping vendor prices very competitive and Verizon’s overall e-discovery costs are now under control. Still, you need good vendors and outside counsel to represent you and handle much of the work. Patrick complains about staffing and observes that it is difficult to find experienced senior e-discovery counsel at many law firms. He urges law firms to invest in their own e-discovery teams so that they can better assist their clients’ teams.

The culture at many law firms dictates electronic discovery counsel and director positions as nonpartner track positions. Although firm culture is shifting, many firms fail to place leadership emphasis with true decision-making power on this crucial position.

It takes months to find a solid project management candidate. Firms need to re-educate their teams.

As an e-discovery attorney in a law firm myself, a partner by the way, I could not agree more.

All of the outside consultants with real expertise in the area agree with the need for e-discovery teams, at least the ones not employed by some of the smaller e-discovery vendors. In fact, Ben Hawksworth of Ernst & Young calls the internal corporate team approach the new “holy grail” of e-discovery.

Still not convinced? Then it is unlikely that more references will persuade. To be honest, most companies are not moved by reason to form an e-discovery team, they are moved by experience, bad experience. Typically, a team is not formed until after a bad case. By that time, boatloads of money have already been spent for fees, costs, and a settlement with the plaintiff. The size of the loss in one case is typically far more than the total cost of team formation. Remember, Cisco was motivated to start the first e-discovery team by a $23,5000,000 bill. This tendency of large organizations to ignore the advice of Ben Franklin makes the plaintiff’s bar quite happy, and more Zubulake type judgments happen every day. I hope that does not happen to you, but without an effective team in place, it is very likely.

It is just like Robert D. Owen, the co-chair of Fulbright & Jaworski’s e-discovery group, likes to say: instead of taking the time to get prepared, many companies have decided to play “the e-discovery lottery. They have decided to take the chance that they won’t be hit. It’s a gamble.” I agree completely with Bob. Many companies are gambling a boatload of money they won’t be the next Morgan Stanley or Qualcomm. It’s a bad bet. The only thing uncertain is the size of the boat and when it will sink.