Preventing e-Discovery Headaches

By Keith Ecker

The rapid growth of electronically stored information (ESI) has greatly altered the litigation landscape. In December 2006 the U.S. Judicial Conference responded to the changing times by amending the Federal Rules of Civil Procedure, codifying the rules surrounding electronic discovery in federal court, but also creating new burdens for both plaintiffs and defense counsel.

To help ease some of these burdens, the federal rules require a meet-and-confer conference, under rule 26(f). Under the amended rules, a responding party must generally discuss preserving and producing ESI no later than 90 to 120 days following service of the complaint. At that time the responding party must identify all sources of ESI--including backup tapes, cell phones, laptops, BlackBerries, instant-messaging systems, and home computers--in connection with its disclosures, as well as identify what categories of information it need not produce because of undue burden or cost.

"[I]t is important for the attorneys to collaborate in the early stages of a case when discovery is being planned," says Gregory B. Wood, a partner in the Los Angeles office of Fulbright & Jaworski. "Parties must sit down and create a discovery plan, such as [database search] keywords being defined and protocols being set, so that they can move forward effectively."

The problem is that sometimes the meet-and-confer conference doesn't go as smoothly as planned. Although, ideally, both parties will come to the meeting fully prepared and ready to compromise on discovery protocols, this isn't always the case.

Benjamin Galdston, an associate in the San Diego office of Bernstein Litowitz Berger & Grossmann, found himself in just such a situation when litigating a securities class action against a company that had a complex and highly customized enterprise resource planning system. The system tied together all facets of the company's operations, from finance to accounting to human resources. Galdston made a routine request for the defense to produce information from this system in "native" format--the form of data used in the ordinary course of business.

"What we got was the proverbial data dump," Galdston says. "The information had no structure to it at all--which, I suspect, the defense knew would be an issue. The lesson learned was to be careful what you wish for, because you might just get it."

Galdston realized he should have had a better understanding in advance of the defense's information management systems. However, he moved fast to correct the situation, returning to the table with opposing counsel to find an effective solution. After extensive negotiations, the defense agreed to set up secure terminals where plaintiffs' counsel could access and search the proprietary system.

Being inundated with mass amounts of ESI is a major risk faced by a requesting party. But producing parties have their challenges as well. Ronald W. Stevens, a partner at the Los Angeles office of K&L Gates, had to negotiate extensively with opposing counsel when his client--the defendant, in this case--was asked to collect and produce relevant information from a large body of documents.

Stevens's client was a large, multinational firm with offices in South America, Europe, and the United States. The case involved accounting issues at several different operations, each of which had a number of major construction projects with contracts worth tens of millions of dollars.

"For just one of the projects, the hard-copy documents from the job site alone filled an entire shipping container," Stevens says. "And those weren't all the hard-copy documents by a long shot related to that particular project. It just represented the order of magnitude of the production, both hard copies and ESI."

Stevens estimated there were potentially hundreds of millions--if not billions--of electronic documents that would have to be collected, processed, and reviewed for responsiveness and privilege to fulfill the request. The undertaking was enormous, and it posed a gigantic cost to his client.

"In a matter like this, where you're talking about expenditures of such large sums, the decision of how to proceed is always the client's," Stevens says. "These are not decisions lawyers should make on their own. The risk and cost are all borne by the client."

Stevens laid out a set of options for his client, and they made the decision together. "I could have just handed over [to opposing counsel] all the shipping containers for the project, given them access to a warehouse of documents, inundated them with disks of ESI, and said, 'See you in twelve years,' " Stevens says. "But by burying them under a ton of material, you're also dumping everything on them that is privileged."

Instead, Stevens sat down with opposing counsel and struck an agreement under rules 26(f) and 16(b), which encourage parties and the court to consider "quick peeks" at information and "clawbacks" of any privileged material included. Opposing counsel agreed to sample some of the data to narrow the pool of potentially relevant information. Both parties then came up with a preliminary list of search terms, which Stevens's team ran against an identified group of computers, specifically those most likely to produce relevant information.

"By agreeing to this method, we were both putting ourselves in a position where, if a dispute ultimately developed, we could both say that we made reasonable efforts to control the scope of discovery," Stevens says.

Stevens's case serves as a positive example of constructive negotiations between sides in litigation. "If you have a good process and you have your ducks in a row, you have several advantages," says John Patzakis, chief strategy officer at Guidance Software in Pasadena. "You can control costs, produce data more efficiently and effectively, and keep control of the process."

Sometimes, however, parties aren't so cooperative. When this happens, the court may choose to levy serious punishments. For example, in a recent federal case in Maryland, Magistrate Judge Paul W. Grimm refused to allow the producing party to reclaim 165 privileged documents that were inadvertently produced as a result of sloppy discovery (Victor Stanley, Inc. v. Creative Pipe, Inc., 250 F. R. D. 251 (D. Md. 2008)). Although the two parties engaged each other in a rule 26(f) conference, the defendant had abandoned the agreed-upon discovery plan and conducted its own keyword search without consulting the court or the plaintiff.

Wood of Fulbright & Jaworski recalls a patent-infringement case that went awry because his own client failed to cooperate with him. The case involved a significant amount of financial information. Not wanting to undertake the burden of producing copious amounts of data, Wood's client chose to selectively produce information it deemed relevant to the case.

"Without actually telling me, the client eliminated documents from production by cutting out information it deemed irrelevant," Wood says. "This came a day before we had a hearing before a magistrate judge. The hearing was difficult for me because I had to explain to the court that this was something that was unacceptable, and not condoned by me."

The court ultimately issued substantial sanctions against Wood's client and compelled it to produce its entire database of financial information; Wood subsequently withdrew from representation. Fortunately, the court did not sanction Wood or his firm for his client's actions--a fate that befell attorneys representing Qualcomm in the now infamous ruling last year by Magistrate Judge Barbara L. Major in San Diego (Qualcomm, Inc. v. Broadcom Corp., 2008 WL 66932 (S.D. Cal. 2008)).

Horror stories such as Wood's may be avoided by ensuring that parties are properly prepared before the meet-and-confer session. One way to accomplish this is to create a data map of the client's information technology infrastructure. Browning E. Marean III, a partner at DLA Piper and cochair of its e-Discovery Readiness and Response Group, helps his clients construct data maps before they sit down with opposing counsel.

"Data maps allow you to determine what information within the organization might not be reasonably accessible, such as backup tapes," says Marean. By coming to the meet-and-confer conference with evidence of what is and isn't reasonably accessible, the producing party will have more success arguing to both the requesting party and the court that certain information is unduly burdensome or costly to produce.

Another preparation strategy for the meet-and-confer conference is early case assessment. By deploying certain technologies, including content clustering and keyword searching, parties can get a snapshot of the type of data they possess. Marean's firm may employ early case assessment to help reduce his clients' litigation costs and carve out a strategy for the 26(f) conference. DLA Piper uses a tool equipped with advanced concept search and clustering technologies to cull and analyze large bodies of data. Once Marean has a glimpse of what his client has in hand, he can move forward with creating a discovery plan.

"At the front end of a case, you need to put together a discovery plan that focuses on electronic information," Marean says. "Make sure that the plan is comprehensive, covering everything from how you are going to identify the potentially responsive data to how you will produce it to the requesting party."

Finally, it's not just plaintiffs and defendants that participate in e-discovery negotiations--the court can intervene as well. If parties are unable to cooperate, the court may lay down its own ground rules for how the discovery process should proceed. Magistrate Judge Grimm in Maryland did just that in another recent case, Mancia v. Mayflower Textile Servs. Co. (253 F. R. D. 354 (D. Md. 2008)). Plaintiffs in a wage-and-hour class action had made an overly broad discovery request, and the defendant gave vague, boilerplate responses for why it should be denied. In his opinion, Judge Grimm admonished both parties for their inability to effectively engage each other during the rule 26(f) conference, writing that their actions were "hindering the adjudication process ... and violating [their] loyalty to the 'procedures and institutions' the adversary system is intended to serve." (253 F. R. D. at 362.)

Besides encouraging parties to cooperate, a court can also order the requesting party to share in some of the costs of production, a practice known as cost shifting. If a discovery request includes information that the defense claims is not reasonably accessible, the court will evaluate the potential relevance of the information and may decide to order production under the stipulation that the requesting party bear some of the financial burden.

"Cost shifting is nice to talk about, but it's still pretty hard to come by," says Marean. "It's deeply rooted in the courts that the producing party pays for the cost of production."

Looking ahead, as the federal courts become more familiar with e-discovery, there may be more attempts to strike a balance between parties through data mapping, sampling, cost shifting, and other tactics. For now, though, the bulk of the responsibility for simplifying the e-discovery process remains with the litigation opponents. Counsel for both parties can reduce the cost of e-discovery by coming to the table with a clear understanding of what information the requesting party wants, and what information the responding party has in its possession.

As attorney Wood concludes, "Collaboration between counsel is the linchpin to economically reasonable litigation, where cases are decided not out of threat of driving up discovery costs but based on their merits."

Keith Ecker is a Chicago-based freelance legal and technology writer.

---