Assessing the Impact of FRE 502 - Weil, Gotshal & Manges LLP

16  n  For The Defense  n  June 2010

n  David J. Lender is co-head of Weil, Gotshal & Manges LLP’s 160-attorney complex commercial litigation practice. He is a recognized expert on e-discovery and co-author of the leading treatise on the subject, Elec-tronic Discovery: Law and Practice. Eric C. Lyttle is a senior associate in Weil, Gotshal & Manges LLP’s liti-gation department. His practice focuses on the defense of products liability actions, mass tort matters, class actions and other complex litigation. In addition, he is a member of the firm’s E-Discovery Task Force and has counseled clients on a variety of electronic discovery- related issues.

When Privileged Information Slips Through

Assessing the Impact of FRE 502By David J. Lender

and Eric C. Lyttle

As courts more uniformly apply the various multifactor reasonableness tests to similar instances of inadvertent production, Rule 502’s cost-savings goal may become achievable.

Attorney review of client documents to avoid inadvertent production of privileged information and possible waiver has long been a staple of the discovery process. The ever-growing size of electronically stored information (ESI) in

E L E C T R O N I C D I S C O V E R Y

© 2010 DRI. All rights reserved.

For The Defense  n  June 2010  n  17

recent years, however, has caused privilege review to become an increasingly time- consuming and expensive task. Traditional page-by-page reviews now often require dozens of lawyers, with costs in the hun-dreds of thousands, if not millions, of dol-lars. In addition to these cost concerns, the sheer volume of ESI makes it more likely that a privileged document will be missed during a review and inadvertent produc-tion will occur. The increased usage of third-party vendors and the presence of metadata that may contain hidden attor-ney comments or edits in native files also increase the risks.

Federal Rule of Evidence 502 was enacted in response to these concerns. Her-alded as providing predictability and cost savings for all parties in litigation, the rule limits the consequences of inadvertent dis-closures of attorney- client communications and attorney work product. Rule 502(e) expressly allows parties to enter into agree-ments that the production of privileged information will not constitute a waiver of the privilege. When those agreements are incorporated into court orders, Rule 502(d) holds that they bind not just the signatories to the agreements but third parties as well. And Rule 502(a) specifically restricts sub-ject matter waivers to instances in which the production of privileged information was intentional and fairness dictates pro-ducing other related documents.

Rule 502(b) adopts a uniform standard that an inadvertent disclosure of privileged information does not waive the privilege if the holder of the privilege took “reasonable steps” to prevent disclosure and to rectify the error. The belief driving adoption of the rule was that consistent and forgiving rules on waiver would reduce the scope of costly privilege reviews prior to production.

But, as always, the devil is in the details. Two open issues stand out in particular. First, because Rule 502 adopted “reason-ableness” as the standard for evaluating a litigant’s efforts to preserve the privilege, courts applying Rule 502 have continued to follow the multifactor, reasonableness tests used by most courts prior to its enactment. Given the different ways that courts have applied these tests, the practical impact of Rule 502 has been unclear. Second, it is uncertain under which circumstances courts will view the use of technology and

keyword searching to constitute “reason-able” efforts to preserve the privilege under Rule 502. Currently, the case law offers more questions than answers, with recent decisions suggesting that litigants’ efforts to use technology for privilege review have come up short. But these cases also provide hope that, with appropriate quality assur-ance and quality controls in place, courts increasingly will view the use of these tools as “reasonable.” Resolving these open ques-tions will go a long way toward determin-ing the practical impact of Rule 502.

Rule 502(b) BackgroundPrior to Rule 502, the federal courts followed one of three different approaches when deal-ing with inadvertent disclosure of privi-leged information: (1)  the “strict waiver” approach, under which any disclosure, re-gardless of knowledge or intention, waived the privilege with regard to the disclosed material and, in some instances, to all other materials covering the same subject matter; (2) the “no waiver” approach, under which only an actual, knowing waiver operated as a waiver of the privilege; and (3) the “major-ity,” “intermediate” or “case-by-case” rule, which used a multifactor, balancing test to make a case-by-case determination. As sug-gested by its name, the majority of courts ap-plied the case-by-case rule.

Congress took the middle road, too: Fed-eral Rule of Evidence 502 was an effort to codify the majority rule, “intermediate” test. Indeed, under Rule 502(b), inadver-tent disclosures that occur in federal court cases will not result in a waiver of privi-lege or work product protection unless the disclosing party failed to take “reasonable steps” to both avoid the disclosure in the first instance and rectify it once discov-ered. Importantly, this inadvertent disclo-sure rule bars courts from finding waivers in federal and state proceedings.

What Are “Reasonable” Efforts to Preserve the Privilege?For all its attempts to standardize the law of waiver, Federal Rule of Evidence 502(b) leaves open two big questions: what con-stitutes “reasonable steps” to avoid an inadvertent disclosure in the first place, and what constitutes “reasonable steps” to rectify it afterward? The Explanatory Note on Rule 502 suggests several factors

that a court may consider in determin-ing whether a party has taken the requisite “reasonable steps,” none of which is dispos-itive. In addition to the “overriding issues of fairness,” these factors include:• The reasonableness of the precautionary

measures• The promptness of the response to rec-

tify the error

• The scope of discovery• The extent of the disclosure• The number of documents to be

reviewed; and• The time constraints for production.

These factors closely mirror the tradi-tional multifactor tests that most federal courts had used to evaluate inadvertent pro-duction prior to Rule 502. Indeed, courts ap-plying Rule 502 have continued to rely on and use the same pre-502 multifactor tests. See, e.g., Peterson v. Bernardi, 262 F.R.D. 424, 428–29 (D.N.J. 2009). This means that pre-Rule 502 cases can prove quite instruc-tive in determining how courts will ap-ply Rule 502. But it also means that while Rule 502 provides a consistent standard for waiver, its practical effect is unclear because it remains subject to the different ways that courts apply multifactor “reasonableness” tests. Several recent cases, all applying sim-ilar multifactor tests for reasonableness, de-cided both before and after the enactment of Rule 502, highlight some of the potential inconsistencies that remain.

Examples of Courts Finding WaiverIn SEC v. Badian, No. 06 Civ. 2621 (LTS-DFE), 2009 WL 222783 (S.D.N.Y. Jan. 26,

While Rule 502 provides

a consistent standard for

waiver, its practical effect is

unclear because it remains

subject to the different ways

that courts apply multifactor

“reasonableness” tests.

18  n  For The Defense  n  June 2010

E L E C T R O N I C D I S C O V E R Y

2009), Rhino Investors, Inc., a nonparty, sought to retrieve 260 documents that it claimed it had inadvertently produced in 2003 in the midst of an SEC investigation. The Badian court applied the four- factor test specified by the Southern District of New York in Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., Inc., 104 F.R.D. 103, 105 (S.D.N.Y. 1985), to ultimately conclude

that Rhino had waived the privilege. Those factors were: (1) the reasonableness of the precautions to prevent inadvertent disclo-sure; (2) the time taken to rectify the error; (3) the extent of the disclosure; and (4) the overarching issue of fairness.

Discussing the first factor, the Badian court emphasized the lack of evidence showing that Rhino took precautions to prevent the production of privileged mate-rials. Rhino’s unsupported assertions that a review had taken place, absent evidence or an explanation of that review in the form of affidavits from counsel or from others, did not satisfy the court. Without this evidence, the court determined that it had “no basis” to conclude any precautions had been taken at all, let alone that they had been reason-able. Badian, 2009 WL 222783, at *3.

Rhino’s efforts to rectify the error like-wise were deemed insufficient. The court first noted the rule in the Southern District of New York that the clock for waiver begins to run once the producing party realizes it has produced privileged materials. See Ara-mony v. United Way of Am., 969 F. Supp. 226, 237 (S.D.N.Y. 1997) (“The period after the producing party realizes that privileged information has been disclosed is the rel-

evant period for measuring whether the privilege has been waived.”). However, the court determined that Rhino “realized” that it had produced privileged material at the time that it produced its documents because it was aware that it had produced a set of documents that likely contained privileged material without undertaking a sufficient effort to identify and assert the applicable privileges. Thus, the court held that Rhino delayed seeking return of the documents for five years. Badian, 2009 WL 222783, at *4–5.

Evaluating the third factor, the court noted that Rhino originally claimed that as much as five percent of the production was privileged. Even after reducing its privilege claim to 260 of 67,000 documents, four-tenths of a percent, the court concluded that this still was a “significant” number of documents. Id. at *4.

The “fairness” factor did not save the day for Rhino either. Here, the court con-cluded that it saw no fairness in preclud-ing the SEC from using the inadvertently produced documents when Rhino did not act with care to guard the privilege in the first place. The court declined, however, to extend the waiver beyond the documents actually produced to the SEC. Id. at *6–7.

In Relion, Inc. v. Hydra Fuel Cell Corp., No. CV06-607-HU, 2008 WL 5122828 (D. Or. Dec. 4, 2008), the court articulated the Rule 502(b) standard as follows: “the court will find the privilege preserved if the privilege holder has made efforts ‘rea-sonably designed’ to protect and preserve the privilege; conversely, the court deems the privilege waived if the privilege holder fails to pursue all reasonable means of pre-serving the confidentiality of the privi-leged matter.” Id. at *2. The review at issue in this case involved a paper review taking up “40 feet” of shelf space. After the plain-tiff’s counsel first reviewed the documents for privilege, defense counsel was allowed to inspect the documents and select the documents that it wanted copied. Two e-mails were inadvertently included in the production. There was conflicting evi-dence on exactly when the plaintiff’s coun-sel learned of the production, but, in any event, the plaintiff did not assert inad-vertent waiver until four months after it had produced the e-mails. The court ulti-mately concluded that waiver had occurred

because the plaintiff had reviewed the doc-uments prior to inspection by opposing counsel, and then had two additional, sub-sequent opportunities to review the doc-uments selected by the defendants, when sets of copies were provided to the defend-ants in both hard-copy and electronic, text- searchable formats. Based on these facts, the court concluded that the plaintiff “did not pursue all reasonable means of preserv-ing the confidentiality… and therefore that the privilege was waived.” Id. at *3.

Although not specifically a Rule 502 case, Clarke v. J.P. Morgan Chase & Co., No. 08 Civ. 02400 (CM) (DF), 2009 WL 970940 (S.D.N.Y. Apr. 10, 2009), reveals the impor-tance of demanding immediate return of privileged documents to avoid a waiver claim. In that case, the defendant had cir-culated the e-mail in question to poten-tial plaintiffs in the course of its business. Because the document was not inadver-tently produced in litigation, Rule 502 did not technically apply. Nevertheless, the court recognized that the same factors rele-vant to a Rule 502 waiver analysis were gen-erally weighed by courts in the context of extrajudicial disclosures. Id. at *5.

Applying these factors, the court held that any privilege attached to the e-mail had been waived. The court stressed that the e-mail in question was the very first docu-ment produced by the plaintiffs in a small group of documents of only 532 pages. And, even after the defendant’s counsel ques-tioned a plaintiff about the e-mail, he made no assertion of privilege and “said nothing about the matter for another week.” Id. at *6. In essence, the defendant waited more than two months after the e-mail had been produced to assert a claim for privilege. The court noted that the cases that found no waiver involved producing parties that had “immediately” asserted the privilege once they discovered the inadvertent dis-closure, and that two months was too long to wait. Id.; see also Sitterson v. Evergreen Sch. Dist. No. 114, 196 P.3d 735, 742 (Wash. Ct. App. 2008) (citing Rule 502 and finding waiver where counsel for the district of-fered no evidence that it had taken precau-tions to prevent disclosure, the district had waited three years after production and just ten days before trial to seek return, and the district’s production was small, constitut-ing only 439 documents).

Cases that found

no waiver involved

producing parties that

had “immediately”

asserted the privilege

once they discovered the

inadvertent disclosure.

For The Defense  n  June 2010  n  19

Examples of Courts Finding No WaiverThe approach of the court in Alcon Manu-facturing, Ltd. v. Apotex, Inc., No. 1:06-cv-1642-RLY-TAB, 2008 U.S. Dist. Lexis 96630 (S.D. Ind. Nov. 26, 2008), to the question of waiver seems more consistent with the leg-islative intent of Rule 502 than that of some other courts. In that case, the court ordered the return or destruction of an inadver-tently produced document that had been marked as an exhibit in two depositions before the producing party realized it was privileged and asked for its return pursu-ant to the “claw-back” agreement in the parties’ protective order. In enforcing the agreement, the court noted that every last effort was not necessary to preserve the privilege under Rule 502:

Perhaps the situation at hand could have been avoided had Plaintiffs’ counsel meticulously double or triple- checked all disclosures against the privilege log prior to any disclosures. However, this type of expensive, painstaking review is precisely what new Evidence Rule 502 and the protective order in this case were designed to avoid.

Id. at *18. Importantly, however, the pro-ducing party in this case had the benefit of a “claw-back” agreement for privileged documents, something that Rule 502(e) encourages. Absent this type of an agree-ment, the producing party may not have been as fortunate under the circumstances.

In Laethem Equipment Co. v. Deere and Co., No. 2:05-CV-10113, 2008 WL 4997932 (E.D. Mich. Nov. 21, 2008), the plaintiffs learned that they had included privileged materials in their two production disks when defendant’s counsel attempted to use the privileged materials during a dep-osition. The plaintiffs then took several affirmative steps to preserve the privilege, including: (1) objecting to the use of the privileged materials; (2) following up the same day with a letter demanding their return; (3) lodging the same objection and demand their return at subsequent depo-sitions; and (4) three weeks after learning of the inadvertent disclosure, securing an order from the court compelling their return. In a subsequent motion to compel their production filed by the defendant, the court analyzed the alleged waiver under Rule 502(b). The court first found that there was no indication in the record or argu-

ments by counsel that the disclosure was anything other than inadvertent. The court next found that the plaintiffs took “reason-able steps” to avoid disclosure. Id. at *9. Unfortunately, the court did not provide much explanation of its analysis of this fac-tor, or even identify which test for “reason-ableness” it had used. But the court did note two particular facts that seemed to bear on the “reasonableness” issue. First, the plain-tiffs produced the two disks outside of the “inspect and copy” procedure established by the parties, which would have given the plaintiffs the opportunity to conduct a privilege review before turning them over to the defendants. And second, the vol-ume of disclosed, privileged data was small compared to the total volume of data pro-duced in the case. Finally, the court found that the plaintiffs acted diligently to rectify the disclosure once they had learned of the inadvertent production. Id.

Identifying the Specific Areas of UncertaintyStepping back from these cases, a few specific areas of uncertainty and conflict emerge:

First, when does the clock for determining waiver begin to run? Does it begin to run when documents are produced or when the producing party realizes that privileged materials have been produced? Most cases analyzing this issue under Rule 502(b) have held that the clock begins to run when the producing party first learns that privileged materials have been produced, but some exceptions persist. See Victor Stanley, Inc. v. Creative Pipe, Inc., 250 F.R.D. 251, 263 (D. Md. 2008). And what should we make of Badian? Badian recognized the rule in the Southern District of New York that the waiver clock begins running from the time that the party “realizes” the inadvertent production has occurred. But the Badian court ultimately concluded that the pro-ducing party should have “realized” that it produced privileged material when it delivered them because of the nature of the material and its failure to perform an effec-tive privilege review. The use of an objective standard, as represented by Badian, rather than a subjective standard, could effec-tively gut the protections afforded by the lenient view of when the clock for waiver begins to run.

Second, what number or percentage of in-advertent disclosures will a court consider significant to find that the privilege has been waived? In Rhoads Industries, Inc. v. Build-ing Materials Corp. of America, 254 F.R.D. 216 (E.D. Pa. 2008), 812 out of 78,000 doc-uments were inadvertently produced, but the court, while critical of the party that inadvertently produced the documents,

ultimately did not find waiver. The court in Victor Stanley Inc. v. Creative Pipe Inc., 250 F.R.D. 251 (D. Md. 2008), in compar-ison, found that 165 inadvertently pro-duced documents was a sufficiently large number, among other factors, to warrant waiver. And the Badian court found that 260 out of 67,000 documents, four-tenths of a percent, was a “significant” number of documents to support a waiver. Another unanswered question is how courts should measure the quantity of inadvertently pro-duced documents. Is it the total number of privileged documents produced, the total number of privileged pages produced, or the relative percentage of privileged docu-ments produced compared to the total pro-duction? This question and the absence of a definitive answer are especially important given the type of large scale document pro-ductions that Rule 502 was intended to ad-dress. The sheer volume of ESI may lead a party to inadvertently produce a high num-ber of documents, while as a percentage of the total production the percentage remains low. Ultimately, we expect courts to focus both on relative and absolute numbers in assessing the reasonableness of a party’s efforts to avoid inadvertent production. On

Ultimately, we expect

courts to focus both on

relative and absolute

numbers in assessing

the reasonableness of a

party’s efforts to avoid

inadvertent production.

20  n  For The Defense  n  June 2010

E L E C T R O N I C D I S C O V E R Y

the one hand, if evidentiary support shows that a party took reasonable steps to avoid inadvertent disclosure, a court may find the absolute number of inadvertently produced documents less significant. On the other hand, if a party inadvertently produces a great number of privileged documents, it could undermine the party’s argument that it took reasonable steps to avoid disclosure.

Third, how long is too long to wait to de-mand return? The case law clearly indicates that a party must immediately demand re-turn. And “immediate” apparently means immediate. Four months in Relion and two months in Clarke were both deemed too long. And in another recent case a court, while finding that the party had not waived the privilege, stated that waiting even six days was too long to qualify as an imme-diate demand for return. See Brookdale Univ. Hosp. & Med. Ctr. v. Health Ins. Plan of Greater N.Y., No. 07 CV 1471 (RRM)(LB), 2009 WL 393644, at *3 (E.D.N.Y. Feb. 13, 2009) (finding no waiver but holding that waiting six days before responding to HIP’s letter, which notified Brookdale of the in-advertent disclosure, and waiting another nine days before notifying HIP of the spe-cific documents that were inadvertently produced was not prompt).

Fourth, what amount of evidence does a party need to meet its burden of estab-lishing reasonableness? A party typically meets the evidentiary burden with affida-vits from counsel or others involved in the discovery process that establishes the steps

undertaken to preserve the privilege and why those steps were reasonable under the circumstances. In dealing with traditional document- by- document reviews, a party generally should not have difficulty meet-ing this burden. For example, in Amer-ican Coal Sales Co. v. Nova Scotia Power Inc., No. 2:06-cv-94, 2009 WL 467576 (S.D. Ohio Feb. 23, 2009), the court found that the plaintiff had employed reasonable steps to avoid inadvertent disclosure “based on the evidence that two attorneys reviewed all documents before they were produced.” Id. at *18. Similarly, in Heriot v. Byrne, 257 F.R.D. 645 (N.D. Ill. 2009), the court relied on the fact that the documents were reviewed by paralegals and other non- lawyers before they were turned over to a vendor who committed the error. Id. at 660.

A Heightened Reasonableness Burden for ESI and TechnologyOne core question that will greatly deter-mine how much cost savings companies can expect from Rule 502 is the extent to which courts will consider the use of technology “reasonable,” in particular the use of keyword searching and linguistic tools. The Explanatory Note on Rule 502 explicitly states: “Depending on the cir-cumstances, a party that uses advanced analytical software applications and lin-guistic tools in screening for privilege and work product may be found to have taken ‘reasonable steps’ to prevent inadvertent disclosure.” But even though the Explan-atory Note on Rule 502 states that using technology may meet the “reasonableness” standard, it does not specify how a party should use it. Two cases, however, provide some insight on what a party will need to do to meet this reasonableness standard.

Victor Stanley, Inc. v. Creative Pipe, Inc.In Victor Stanley, Inc. v. Creative Pipe, Inc., 250 F.R.D. 251, 263 (D. Md. 2008), a case decided before Rule 502 was enacted, 165 documents slipped through a privilege review based on keyword searches and a “title page” analysis for non- searchable items. Chief Magistrate Judge Paul Grimm of the District of Maryland—one of the judges principally responsible for drafting Rule 502—used the five- factor test outlined in McCafferty’s, Inc. v. Bank of Glen Bur-nie, 179 F.R.D. 163, 167 (D. Md. 1998), to

evaluate the reasonableness of defendants’ efforts to protect the privilege. The factors evaluated by Judge Grimm were: (1)  the reasonableness of precautions that the pro-ducing party had implemented to pre-vent inadvertent disclosure in view of the extent of the document production; (2) the number of inadvertently disclosed materi-als; (3) the extent of the disclosure; (4) any delay in initiating and the measures that the producing party employed to rectify the disclosure; and (5) whether the “overriding interests of justice” would or would not be served by relieving the producing party of its errors. Victor Stanley, 250 F.R.D. at 259. The court ultimately found that the defend-ants had waived the privilege because they did not show that the document searching and reviewing methodology they employed were “reasonable.” Id. at 262–63.

The court viewed the first factor as weigh-ing the most strongly in favor of a waiver. The court noted a number of problems with the defendants’ efforts to preserve the priv-ilege, including that they had failed to de-scribe: the key words selected; the rationale for selecting them; the qualifications of the individuals who had selected the key words and designed the search methodology; how the search was conducted (individual key words, Boolean, Bayesian classifiers, “fuzzy” searches, “clustering,” or “concept and categorization”); the quality controls that they had employed to assess reliabil-ity and accuracy prior to production; and whether they had performed sampling to comfortably determine that the documents produced through the particular methodol-ogy were not privileged. Id. at 256, 259–60. In short, the court found that the defend-ants had failed to meet the necessary bur-den of explaining what they had done and why it was sufficient. What the defendants did describe, the court found “regrettably vague.” Id. at 256.

The court quickly disposed of the other factors. In discussing the second and third factors, the court noted that the defend-ants had produced 165 documents that they asserted were privileged, which the court considered a significant amount because it was not simply “an instance of a single document slipping through the cracks.” Id. at 263. Addressing the fourth factor, the court specified that the defendants had not acted promptly to rectify the error, noting

When using keyword

searching for privilege

review, proper quality-

assurance testing, or

the lack thereof, is a

factor in considering

whether precautions

have been reasonable.

For The Defense n June 2010 n 21

a “one-week period between production by the defendants and the time of the discov-ery by the plaintiff of the disclosures—a period during which the defendants failed to discover the disclosure.” Id. Finally, writ-ing about the fifth factor, the interests of justice, the court noted that “every waiver of the attorney- client privilege produces unfortunate consequences” before con-cluding that the only “injustice” here had been committed by defendants to them-selves. Id. It bears mentioning, however, that the defendants had abandoned earlier efforts to secure a ‘claw-back’ agreement, a fact which may have encouraged the court to evaluate the defendants’ efforts with an especially critical eye.

Rhoads Industries, Inc. v. Building Materials Corp. of AmericaIn Rhoads Industries, Inc. v. Building Mate-rials Corp. of America, 254 F.R.D. 216 (E.D. Pa. 2008), the court applied Rule 502 to a set of facts similar to the facts in Victor Stan-ley but reached a more charitable conclu-sion for the producing party. The plaintiff in Rhoads inadvertently included approx-imately 800 privileged documents among 78,000 e-mail files produced in response to the defendants’ discovery requests. After a defendant notified the plaintiff that certain documents appeared to be privileged, the plaintiff responded that the production was inadvertent and did not waive any privi-lege, and the plaintiff also requested that the documents be sequestered. The parties agreed that the documents were privileged and that the production had been inadver-tent, so the only issue left for the court was whether the plaintiff had taken reasonable steps to preserve the privilege and rectify the inadvertent production. The court first decided that Rule 502 was applicable and then turned to the same five-factor test used by Judge Grimm in Victor Stanley.

Applying the first four factors, the court found that the plaintiff’s efforts to avoid in-advertent disclosure had not been reason-able. Focusing on the first factor, the court noted that, while the plaintiff had retained an information technology consultant and purchased a software program designed to identify responsive e-mails and screen for privilege, the specific search terms used were inadequate, particularly because the names of all attorneys had not been used.

The court also faulted the plaintiff for lim-iting its search for privileged documents to e-mail address lines rather than also including the e-mail body. By failing to search the e-mail body, any potentially privileged e-mails that were subsequently forwarded would not have been captured by the plaintiff’s search. Echoing Victor Stanley, the Rhoads court noted that, when using keyword searching for privilege re-view, proper quality- assurance testing, or the lack thereof, is a factor in consider-ing whether precautions have been rea-sonable. In this case, the plaintiff did not test the reliability or comprehensiveness of the keyword search. A manual review of some of the mailboxes by plaintiff after the electronic search, but prior to produc-tion, had revealed the existence of addi-tional privileged materials and should have alerted the plaintiff to the shortcomings of its search methodology. And the plaintiff’s only quality- assurance testing of search terms consisted of running the exact same search again. Id. at 224.

On the second factor, the court noted that 812 inadvertently produced docu-ments out of a total of 78,000 was a large, absolute number, regardless of the rela-tive percentage. The court found that the record offered insufficient evidence on the third factor either to support or deny find-ing a waiver. But, on the fourth factor, the court concluded that the plaintiff had failed to prepare for and devote significant resources to the segregation and review of privileged documents sufficiently far in advance of a production that it knew would be large. Indeed, the court offered language sure to catch the attention of litigants: “An understandable desire to minimize costs of litigation and to be frugal in spending a client’s money cannot be an after-the-fact excuse for failed screening of privileged documents….” Id. at 224–25, 227.

For the court, however, not all of these factors were created equal. Indeed, despite finding that four of the five factors favored finding a waiver, the court ultimately found that the fifth factor, the interests of justice, trumped the others, and ruled against a finding of waiver. In reaching its deci-sion, the court noted that the “[l]oss of the attorney- client privilege in a high-stakes, hard-fought litigation is a severe sanc-tion” that could cause serious prejudice.

Id. Denying the defendants access to the documents, on the other hand, would not prejudice them because they had neither a right to, nor an expectation of access to, any of the plaintiff’s privileged communi-cations. Id.

Although Victor Stanley and Rhoades did not sanction the parties’ use of key words, they do provide some helpful guidance on how a party can try to meet the reasonable-ness test when conducting a technology- based privilege review. The key points are that it is important that a party (1) design a justifiable search that it reasonably an-ticipates will isolate privileged documents, and (2) perform sufficient quality control and assurance testing before production, for example through sampling, to ensure that the search was reasonable and likely to identify privileged documents. Indeed, a party may need to offer expert verifica-tion in certain courts, given the two deci-sions by Judge Facciola of the United States District Court of the District of Columbia, United States v. O’Keefe, 537 F. Supp. 2d 14, 23–24 (D.D.C. 2008), and Equity Analyt-ics, LLC v. Lundin, 248 F.R.D. 331 (D.D.C. 2008), both of which held that evidence re-garding the adequacy of electronic keyword searches may have to meet the criteria for expert evidence under Federal Rule of Ev-idence 702. Regardless of whether a court requires expert evidence, litigants wishing to use technology- based privilege reviews will want to heed Judge Grimm’s general admonition in Victor Stanley:

Use of search and information retrieval methodology, for the purpose of iden-tifying and withholding privileged or work- product protected information from production, requires the utmost care in selecting methodology that is appropriate for the task… [and] requires careful advance planning by persons qualified to design effective search methodology. The implementation of the methodology selected should be tested for quality assurance; and the party selecting the methodology must be prepared to explain the rationale for the method chosen to the court, demonstrate that it is appropriate for the task, and show that it was properly implemented.

250 F.R.D. at 262.FRE 502�, continued on page 70

70  n  For The Defense  n  June 2010

ConclusionIt is axiomatic that predictability breeds security. Thus, as courts more uniformly apply the various multifactor reasonable-ness tests to similar instances of inadver-tent productions, Rule 502’s cost- savings goals may become achievable. In particu-

lar, creating a consistent set of standards regarding when the waiver clock begins to run, the number or percentage of inadver-tent disclosures considered unreasonable, and the requisite type of evidence sufficient to establish reasonableness will help estab-lish consistency that litigants can rely on. Equally important, achieving cost savings

FRE 502�, from page 21 under Rule 502 will demand e- discovery technology that provides the quality assur-ance and quality control that make relying on it “reasonable.” As the technology and litigants’ use of it improves, we may find the admirable goals of Rule 502 within reach.