Coping With E-Discovery: Practical Tips From An Experienced Litigator

Monday, November 26, 2012 - 10:42

The Editor interviews Christopher Cox, Partner, Weil Gotshal’s Silicon Valley office, where he leads the firm’s California complex commercial litigation practice.

Editor: Chris, tell us about your practice.

Cox: I’ve been practicing for 22 years and have been a litigator from the beginning. I joined Weil in 1997. I started as an associate, running large-scale document productions long before e-discovery and lived through the transition to processing and producing large volumes of electronically stored information.  I have experience managing multi-jurisdictional disputes, multi-district litigations, bankruptcy and adversary proceedings, internal investigations and litigations involving intellectual property, business torts, contracts, fiduciary duty, employee mobility and non-competes and unfair competition -- including in trial and on appeal. I initially focused on commercial litigation and then, because of the nature of our business in California, found myself doing a lot of patent litigation. In 2010, the firm asked me to help build our complex commercial litigation practice in California, which is my current focus, although I still keep one foot in the patent arena.

Editor: Did you want to mention any of your clients?

Cox: My recent and ongoing experience includes representing Lehman Brothers Holdings in its bankruptcy and related litigation; Panasonic, as plaintiff, in a shareholder derivative suit relating to the company’s minority stake in a joint venture to manufacture cardio-medical devices; and I recently represented Scott Kay, Inc. in copyright and trademark litigation brought by a competitor regarding jewelry designs. I’m currently working for Alliant Insurance Services in a multi-jurisdictional case involving mobility issues. I am working on some trade secret cases and patent cases for other clients. It is a fairly broad practice.

Editor: There is a great concern in the corporate counsel community about the cost of e-discovery. To what extent is the cost of e-discovery being abused as a tool to pressure a party into settlement?

Cox: The cost of e-discovery is not typically used as leverage for settlement in most commercial litigation between competitors where cost is a problem for both sides. It is definitely a problem in patent litigation initiated by non-practicing entities. In such cases, the burdens of e-discovery fall disproportionately on the defendants, especially corporate defendants. As a result, we’ve seen a trend in those cases toward judicially established e-discovery limits.

Editor: To what extent has the requirement of fact-based pleading in Twombly and Iqbal been successful in limiting e-discovery?

Cox: Twombly/Iqbal motions are powerful vehicles for limiting the scope of a case and thus the cost of e-discovery. It costs the client more up front on motion practice, but can pay off in limiting discovery costs by pruning away frivolous claims, which are often the most expensive. By forcing the plaintiff to describe the claims with better specificity, we’re able to narrow the scope of discovery.

Editor: Is that strategy equally effective at a state level?

Cox: Less so because different states have different pleading requirements. It is of little help in states with pure notice pleading standards. That said, although not precedential, the Supreme Court opinions will have influence with certain judges in those jurisdictions. In California state courts, there has always been a higher, fact-based pleading standard than pure notice pleading.   It is closer to the Twombly/Iqbal standard of requiring the pleading of ultimate facts necessary to establish each element of a claim.

Editor: Has any progress been made in limiting e-discovery at the federal level?

Cox: In September 2011, Chief Judge Rader of the Federal Circuit Court of Appeals unveiled a new model order for governance of e-discovery in patent cases. It was drafted by an e-discovery committee of the Federal Circuit Advisory Council. It places limits on many of the most expensive aspects of e-discovery. For example, there are limitations on when email productions can be requested.

The new model order also places presumptive limits on the number of custodians to be searched, the number of keyword search terms for each custodian and the relevant time frame for collection purposes. Obviously judges have discretion to alter those presumptive limits.  The Eastern District of Texas, which also handles a significant number of patent cases, adopted its own model rules that incorporate similar presumptive limits.

As the courts get more comfortable imposing such presumptive limits on the front end of patent cases, I believe we will see these limits extended to other types of litigation. From a corporate counsel’s perspective, there is no reason to wait for the FRCP to be amended to seek relief from burdensome and costly e-discovery. The marching orders given by corporate counsel to their outside counsel should be to take a look at these model rules being promulgated throughout the country to see if they can be adapted for their cases.

Editor: What is the impact of predictive coding in coping with the cost of e-discovery?

Cox: The impact of predictive coding in litigation is still somewhat uncertain. It has been approved in some courts and, as familiarity with the technology grows, we’ll see broader adoption of it. For those readers who are unfamiliar with the phrase, predictive coding involves having outside counsel review a subset of documents from a much larger volume of documents. He or she then codes those documents for responsiveness. Based on the coding of the sample documents, the remaining documents are categorized using an algorithm which assigns each document a percentage reflecting the likelihood that document has of being responsive and relevant.

In this way, predictive coding helps reduce the volume of documents that need human review. The important thing about predictive coding, however, is that it does not eliminate the need for review by humans. What it does is whittle away the larger volume that would otherwise have been reviewed by humans. You still need attorneys to review the remaining documents and make decisions as to relevance and responsiveness. Thus, predictive coding is not a panacea for eliminating cost, but it can provide significant cost savings for large-volume productions.

We see particularly significant savings from the use of predictive coding in internal investigations. Because these are by definition internal to a company, there is no need for negotiation with outside parties – as there is with litigation – over process and reliability. In an internal investigation, e-discovery is often the biggest expense, and time is usually of the essence. Under these circumstances, predictive coding allows us to find the most relevant documents quickly, efficiently and at a much lower cost.   

Editor: Where predictive coding is used, I understand that you would have the right to demand to see the documents that were used to train the machine. Some people have made the point that by disclosing those documents to the other side you’re giving away your case.

Cox: Obviously every case is different. However, since the costs and considerations like the one you mentioned affect both parties, such issues will usually be brought up during the meet and confer under Rule 16 and Rule 26. At that time, the parties should consider how e-discovery will be handled and discuss whether to use predictive coding. In considering the use of predictive coding, an element in their decision will be the extent to which the other side will become privy to information about their case.  Lawyers don’t like to disclose their work product, and for good reason.   

Editor: To what extent do you find that predictive coding has displaced keyword searches?

Cox: I would say that there’s a trend to use it more, but change takes time and people are still very comfortable with how they’ve been doing things for years. In my cases, keyword searching without predictive coding is still predominant. That said, when using keyword searching only, e-discovery costs can still be reduced by limiting the number of custodians, limiting keyword search terms and using such techniques as de-duping.  

Editor: Large companies are particularly concerned about the cost of overpreservation.

Cox: That’s a very legitimate concern.  If you are in litigation or reasonably anticipate it, you are going to have e-discovery costs, including the costs associated with preservation. That’s inevitable, but you can reduce those costs through proper planning and coordination among your outside counsel, your discovery vendors and your IT department – bearing in mind that penalties associated with spoliation can often dwarf the costs associated with preservation. When you are charged with spoliation, you can no longer show that what was destroyed was not relevant or responsive because it’s not there anymore. The value of a missing document becomes whatever the opposing counsel can dream up in their wildest fantasies.

It is therefore crucial that you have a defensible search and preservation plan so that if something does inadvertently happen, you can show the court that you were acting as a good corporate citizen and were taking your obligations seriously.

As soon as litigation or a reasonable anticipation of litigation appears on the horizon, corporate counsel and outside counsel need to sit down and determine the initial scope of potentially relevant evidence, which requires identifying witnesses, documents, subject matter, time periods and other relevant information. A defensible search should be conducted for relevant evidence that’s reasonably subject to a litigation hold. Careful implementation of such a program can permit a company to avoid preserving everything until it is told what to produce.

Editor: Can ADR provide relief from high e-discovery costs?

Cox I’m a big proponent of ADR because it can be the most cost-effective approach to dispute resolution if properly conducted.  

Typically, we use mediation.  If cost is a factor, the parties can elect court-supervised mediation. There are quite a few effective mediators who provide this public service.  Mediation allows deals to get done because one side is not the winner and one side is not the loser. Both parties walk away with something they can live with.  In contrast, nonbinding arbitration or early neutral evaluation can often be fruitless because whichever party loses will think the neutral simply got it wrong.  Those procedures make sense for the right kind of case, but mediation has a broader appeal.

Please email the interviewee at with questions about this interview.