At Evidence Exchange we aim to provide helpful guidance to our clients regarding complex litigation technology and discovery strategy management issues. One of the keys to successful discovery in complex cases is knowing how and when to deploy the appropriate search tools. Lawyers need to select and use software that not only retrieves the relevant documents, but also performs their tasks in a defensible and cost-efficient manner.
It is critical for parties to consult with their e-discovery experts and engage in project planning that takes into consideration issues such as case strategy, cost sensitivity and the nature of the data set. A good solid search strategy will allow lawyers to gain full knowledge of the relevant facts that are necessary in order to develop a realistic appraisal of a given case. And, selecting the right search tool is one major step along the path to reach that important goal. In this issue of The Metropolitan Corporate Counsel, we interview Ross Gotler of Paul Weiss.
Ross is Practice Support & E-Discovery Counsel at Paul, Weiss, Rifkind, Wharton & Garrison LLP and oversees the firm’s Discovery Services and Practice Support teams. He advises the firm’s lawyers and clients on technology and practice related to e-discovery in a wide variety of complex litigations and regulatory investigations, with clients ranging from leading financial services institutions to global trading companies. Ross is a member of the Sedona Conference Working Groups on Electronic Document Retention and Productions and on International Electronic Information Management, Discovery and Disclosure. He is a board member of the E-Discovery, Compliance, and Litigation Support Management Association (ECALSM) and has participated in the National Institutes of Standard and Technology’s Text Retrieval Conference (TREK) Legal Track.
Tom Svoboda is Managing Director at Evidence Exchange where he leads the firm’s client education and peer-to-peer collaborative ventures. He also serves as a strategic advisor to the company and its clients. Prior to joining Evidence Exchange, he held executive management positions at Cravath, Swaine & Moore LLP and IBM. Tom is a member of the Sedona Conference Working Group on Electronic Document Retention and Production.
Svoboda: While there are numerous search tools and methods available to litigators, what advice can you give us on selecting the proper tools?
Gotler: In e-discovery practice, it’s important to remember the old maxim “always use the right tool for the job.” Your email program may be great for email, but it may not be the best system for preserving, collecting and reviewing electronically stored information (“ESI”). Outside counsel, in-house e-discovery professionals, and outside consultants – who ideally aren’t selling a particular system – can be very helpful in recommending the proper search tools and methods for each matter.
While cost and proportionality issues should always be considered, the litigation hold and document preservation phase of an e-discovery project is not the time to cut corners. Investing in defensible tools and processes at this phase, such as having an outside consultant or trained in-house technology professional conduct preservation and collection activities, can pay dividends by helping to protect a party against later claims of spoliation.
For example, I've been involved in matters where a party wanted to rely on having employees self-identify and archive email messages as their only means of preservation. In many of these situations, it was more prudent to train in-house technology professionals on how to capture full email files using a proven, reliable collection tool or to engage an outside consultant to conduct such collection. This can result in a more defensible preservation and collection process that can better withstand future scrutiny.
For review and production, there is no one-size-fits-all search tool or process. It’s important for parties to consult with their e-discovery experts and engage in project planning that takes into consideration issues such as case strategy, cost sensitivity and the nature of the data set, and that results in a plan for what tool to use and how to use it. For example, many companies and their outside counsel may be considering using predictive coding as part of a document review, as opposed to traditional linear review. If the set of ESI to be reviewed likely contains a substantial amount of irrelevant material, predictive coding may be an effective way to expedite document review and lower its cost. But when the vast majority of the ESI review set is likely to be relevant or an agreement exists to produce all documents that contain specific keywords, a traditional linear review may be more appropriate.
Svoboda: The most common and widely used search technology uses keywords – a list of words or word combinations generated for the current understanding of the facts in a case. Can you tell us what your experience has been in using this method?
Gotler: Keywords can be effective if used properly and in the right context. Judge Andrew Peck put it well in his William A. Gross decision when he discussed the need for “careful thought, quality control, testing, and cooperation with opposing counsel” in designing keywords. The discussion about using keywords, though, is very different depending on the phase of the e-discovery process.
Some litigants may use keywords at the beginning of the e-discovery process to preserve potentially relevant information. For example, as part of implementing a litigation hold, a party might filter live email boxes with keywords and preserve the results – and nothing else. While this may be acceptable in some circumstances, it introduces a high level of risk, since the messages that are not keyword hits are not preserved. One way to mitigate this risk is to disclose to opposing counsel the preservation plan and to agree on the set of keywords used. Since the defensibility of the preservation effort is dependent on the quality of the original search, a party should carefully consider if and how it should use keywords for preservation.
Keywords are frequently used by parties to filter the set of collected ESI. When following such a process, any documents that contain a keyword (and any document "family members") are usually placed in a set for review, while the rest of the documents are set aside and not reviewed. When developing keywords to filter a larger set of ESI for review and production, there are a number of considerations for general counsel and their outside law firms. First, is such keyword filtering the best course for the matter, or should other techniques – such as predictive coding, concept clustering and metadata analysis (such as sender and recipient or key phrases in the ESI set) – also be utilized?
Next, invest effort up front to ensure the keywords are as effective as possible at "recall" – accurately identifying for further review the complete set of documents in the full ESI collection that are responsive. Accomplish this through research such as interviews with custodians about what language they may have used to discuss a matter and through a review of some representative documents to see such language firsthand. Such research may help identify critical information like code words, abbreviations or project names that otherwise may not have been included in the keyword list.
Additionally, consult with your e-discovery experts on mechanics of keyword creation, such as how to create wildcard, stemming and proximity searches and how to create simple and complex Boolean queries. Different applications have different searching capabilities – a search that works on one system may require a completely different syntax on another system. Avoid over-inclusive keywords, which could result in a large amount of irrelevant documents being added to the review set. Test the keywords, preview the quantity and content of the results and make changes to the keywords as necessary. Quality check the search both before and after it is executed – ensure that search queries are accurately constructed and that you are receiving the results that you expect to receive. Whenever appropriate, parties should collaborate on keywords to be used for filtering, as this may avert future e-discovery management disputes, but before agreeing to a set of keywords, test the keywords to be sure of the document review burden that will result.
Svoboda: What’s the best approach for dealing with smaller document collections or for looking for the proverbial “smoking gun”?
Gotler: It's true that not every litigation is complex or involves millions of documents. In some cases, especially with small sets of ESI, it may make sense to put everything into a review system and conduct a full document-by-document linear review. I would advise parties, though, always to give careful consideration to how technology may be used to expedite and improve the quality of the e-discovery process. For example, if a party wants to review all email sent by one person to certain others during a specific time period, there are document review tools that allow parties to apply metadata filters to the ESI set to show such documents in a matter of seconds. Identifying documents using this method, as opposed to relying on linear review, may result in a faster, less expensive review. And quality and precision of the review may increase since the reviewers are only looking at a potentially highly relevant subset of the ESI.
In a recent expedited investigation, the goal was to conduct a thorough review of key documents within a given time period while also controlling costs and ensuring the highest quality and precision possible. Instead of a full linear review of every custodian's ESI, which was not what the requesting party wanted, we used advanced techniques to identify email conversations from key custodians, and then we reviewed those conversations. Our result was a comprehensive review of the documents at interest in a fraction of the time and cost of a linear review and, importantly, a satisfied requesting party.
Published March 22, 2012.
