June’s Notable Cases and Events in E-Discovery

1. Opinions from the U.S. District Court for the District of Kansas awarding a defendant expenses it incurred from implementing expansive technology assisted review

In a handful of recent decisions in Lawson v. Spirit AeroSystems, Inc., Case No. 18-1100-EFM-ADM (D. Kan.), Magistrate Judge Angel D. Mitchell, with affirmance from District Judge Eric F. Melgren, awarded a defendant approximately $850,000 in expenses it incurred from expansive technology assisted review (TAR) the plaintiff had insisted on and briefing in connection with seeking such expenses. 

In this breach of contract case, Plaintiff, the former CEO of Defendant, entered a noncompete agreement, which Defendant alleges he breached. Lawson v. Spirit AeroSystems, Inc., 2020 WL 3288058, at *1 (D. Kan. June 18, 2020). Because of the alleged breach, Defendant stopped paying Plaintiff amounts Defendant owed under Plaintiff’s retirement agreement. Plaintiff then filed suit and sought far-reaching discovery from Defendant. Id. at *3. 

The parties engaged in a lengthy and protracted discovery process, during which Plaintiff identified nearly 70 custodians and about 90 search terms. Defendant ultimately selected a handful of custodians and ran Plaintiffs’ search terms over their documents, which returned over 320,000 documents. After reviewing a sample set, Defendant determined that 85% of the documents were irrelevant. Magistrate Judge Mitchell weighed in on the parties’ discovery dispute and ordered the parties to, among other things, work together to formulate more effective search terms and to try to achieve a responsiveness hit rate of at least 85%.

After the parties’ attempts to narrow in on a set of documents with a higher responsiveness rate were unsuccessful, the parties decided to transition to conducting a TAR of the 320,000 documents. Id. at *6. The parties agreed that the initial review would be conducted by Defendant’s contract attorneys with a second-level review by attorneys at Defendant’s retained firm. After initial disagreement regarding the appropriate recall rate, Defendant agreed to continue its review until it reached an 80% recall rate. Id. at *7. Once Defendant reached 80%, Plaintiff moved to compel Defendant to review and potentially produce from the residual TAR set, a process Defendant estimated would cost $40,000. Magistrate Judge Mitchell denied the motion because Plaintiff refused to bear Defendant’s costs to review and produce the residual TAR documents, no authority supported what Plaintiff sought, and the additional review was not proportional to the needs of the case. Id. at *8. 

Thereafter, Defendant filed a motion requesting that Magistrate Judge Mitchell shift all costs and attorneys’ fees associated with the TAR to Plaintiff. Id. at *1. Defendant argued that it “spent months collecting, processing, hosting and searching millions of documents from custodians selected by [Plaintiff] and using search terms selected by [Plaintiff]; that this process cost hundreds of thousands of dollars and yielded only a small percentage of responsive documents ‘and far fewer relevant documents.’” Id. (emphasis in original). While Defendant was working to comply with Plaintiff’s demands regarding the TAR protocol, Defendant also conducted custodian interviews and gathered targeted files, which resulted in far more significant and fruitful productions on the main issues in the case. Defendant therefore requested that Magistrate Judge Mitchell shift the TAR expenses to Plaintiff, citing the proportionality standards in the Federal Rules of Civil Procedure. 

Plaintiff opposed the motion, arguing that cost-shifting was appropriate only for electronically stored information (ESI) that is not reasonably accessible, which was not the case here. Plaintiff also contended that Defendant had not cooperated, as Magistrate Judge Mitchell had ordered it to, in crafting search terms to reach an 85% responsiveness target, and that cost-shifting factors weighed in Plaintiff’s favor.

As an initial matter, Magistrate Judge Mitchell disagreed with Plaintiff’s argument that cost-shifting was appropriate only for ESI that is not reasonably accessible, stating it was “contrary to both the Federal Rules and the parties’ own agreed ESI protocol.” Id. at *10. While Magistrate Judge Mitchell acknowledged that “it is ordinarily presumed that the responding party bear the expense of complying with discovery requests,” Defendant’s motion was premised on Rule 26(c) of the Federal Rules of Civil Procedure, which is not limited to non-reasonably accessible discovery and was amended in 2015 “to make clear that the court may allocate discovery expenses for good cause in order to protect a party from undue burden or expense.” Id. at *8.

After considering each of the relevant proportionality factors courts look to in connection with Rule 26(c), Magistrate Judge Mitchell concluded that good cause warranted allocating the TAR expenses to Plaintiff to protect Defendant from undue burden and expense. Id. at *10. Magistrate Judge Mitchell explained that Defendant needed protection and relief from Plaintiff’s burdensome and costly discovery tactics and noted that she had repeatedly cautioned Plaintiff to better focus his requested electronic discovery or Magistrate Judge Mitchell would eventually shift costs. See Id. at *22. She also noted that Defendant had already shouldered its fair share of the expenses by cooperating with Plaintiff’s burdensome requests and by running a separate discovery process using traditional discovery means, which ended up producing many more responsive documents than the TAR protocol Plaintiff demanded. As Magistrate Judge Mitchell put it, because Plaintiff was “the party that wanted to proceed with the TAR process at a point in time when it was disproportional to the needs of the case, the court will allocate the TAR expenses to [Plaintiff] to protect [Defendant] from undue burden and expense.”

But Magistrate Judge Mitchell did not determine a specific dollar amount to allocate to Plaintiff and instead ordered Defendant to serve a notice by June 25, 2020, informing Plaintiff of the dollar amount Defendant was requesting. Magistrate Judge Mitchell also ordered the parties to confer in an attempt to reach agreement on the issue of expenses by July 2 and stated that if the parties could not reach an agreement, Defendant could file a motion seeking expenses by July 10.

Plaintiff appealed Magistrate Judge Mitchell’s order to the district court judge, Judge Eric F. Melgren. Lawson v. Spirit AeroSystems, Inc., 2020 WL 6939752, at *1 (D. Kan. Nov. 24, 2020). In his appeal, Plaintiff argued that “the Magistrate Judge’s order should be overturned because it was clearly erroneous and contrary to law.” Id. at *5. Specifically, Plaintiff contended that the TAR protocol was necessary in light of Defendant’s defense strategy and that Magistrate Judge Mitchell’s order was clearly erroneous because the TAR process “uncovered some useful evidence to support his theory of the case.” Id. at *6. On the first point, Judge Melgren stated that both parties have “a significant interest in discovery” and that up until Magistrate Judge Mitchell’s order, the main difference between the parties was that Defendant was shouldering the lion’s share of the costs of that discovery. Id. at *5. On Plaintiff’s second point, Judge Melgren stated that he was “not persuaded by [Plaintiff’s] ends justify the means argument.” Id. (internal quotation marks omitted). The mere fact that the TAR had led to production of some relevant documents did not, by itself, establish that it was proportional to the needs of the case. In the end, Judge Melgren affirmed Magistrate Judge Mitchell’s cost-shifting order, stating that “[t]he Magistrate Judge considered all the relevant facts and concluded that [Plaintiff’s] persistence in pursuing the costly, ineffective TAR was disproportional to the needs of the case” and that Plaintiff “failed to carry his burden to prove that the Magistrate Judge’s order was clearly erroneous or contrary to law.”

After the parties were unable to reach an agreement on expenses, Defendant filed a motion requesting expenses related to the TAR. Lawson v. Spirit AeroSystems, Inc., 2020 WL 6343292, at *1 (D. Kan. Oct. 29, 2020). Defendant sought $791,700.21 in expenses in connection with the TAR and $83,000 in costs and fees incurred conferring with Plaintiff and preparing the briefing associated with its motion for fees. Plaintiff objected to Defendant’s requests, arguing that “many of the expenses included in [defendant’s] calculation [were] unreasonable or outside the scope of the court’s order.” According to Plaintiff, Defendant’s reasonable TAR expenses were “no more than $330,000.”

Magistrate Judge Mitchell granted in part and denied in part Defendant’s application for expenses, ruling that in total, defendant “reasonably incurred $449,999.21 in expenses paid to [its e-discovery vendor] and $304,030.25 in attorney’s fees[.]” Id. at *19. Magistrate Judge Mitchell also provisionally granted Defendant’s request for expenses in connection with submitting its fee application but concluded that she could not “determine a specific dollar amount at this time.” Accordingly, she ordered the parties to engage in a similar protocol as before, and if, as before, the parties could not reach an agreement on the amount of these expenses, Defendant was to prepare a renewed application.

After the parties again failed to reach an agreement, Defendant followed Magistrate Judge Mitchell’s order and filed a renewed application. Lawson v. Spirit AeroSystems, Inc., 2020 WL 7700228, at *1 (D. Kan. Dec. 28, 2020). In its renewed application, Defendant sought, in total, $94,407.25 in expenses incurred in connection with its initial and renewed application. In response, Plaintiff argued that the expenses Defendant requested were “excessive, duplicative, and should be reduced by at least 50%.” Plaintiff also argued that he should not be responsible for any expenses incurred after Defendant filed its initial application. Recognizing that she had already provisionally granted Defendant’s request for expenses reasonably incurred in connection with preparing its original application, Magistrate Judge Mitchell granted Defendant’s renewed application in full, ruling that Defendant had reasonably incurred $94,407.25 in attorneys’ fees. Id. at *6.

2. A U.S. District Court for the Southern District of New York opinion sanctioning defendants for intentionally destroying encryption keys needed to access an accounting system used in a bribery scheme because the defendants should have known the encryption keys would be relevant to future litigation

In DoubleLine Capital LP v. Odebrecht Finance, Ltd., 2021 WL 1191527 (S.D.N.Y. March 30, 2021), U.S. Magistrate Judge Barbara Moses sanctioned several Defendants for intentionally destroying the encryption keys needed to access a “shadow” accounting system used in a bribery scheme, ruling that Defendants should have known the keys would be relevant to future litigation.

In this federal securities fraud action, Plaintiffs sought the sanction of a mandatory adverse inference instruction at trial against certain Defendants as a remedy for Defendants’ having intentionally destroyed the encryption keys needed to access the MyWebDay platform, an internal, “shadow” accounting system used to track the illicit bribe payments underlying the claims in the lawsuit. Id. at *2. After Defendants became aware of an investigation by Brazilian authorities into the suspected corruption and bribery scheme but before the litigations were filed, Defendants intentionally destroyed the physical encryption keys needed to access the system, which Plaintiffs contended contained crucial evidence regarding the scope and nature of the bribery scheme. Id. at *2-3.

Defendants did not deny they intentionally destroyed the encryption keys but argued that the motion for sanctions was premature because discovery had only just begun and that Plaintiffs could not demonstrate that “the information cannot be replaced or otherwise approximated through additional discovery.” Id. at *4. Defendants also argued that the relevant information from the MyWebDay system, namely the dollar amount of the bribes, could be addressed through a stipulation as to the amount of the bribes.

Magistrate Judge Moses began her analysis by describing the relevant standards for sanctions under Fed. R. Civ. P. 37(e). She noted that the sanctions permitted under Subsection (e)(1) of Rule 37, available upon a finding that the spoliation caused “prejudice to another party,” must be limited to “measures no greater than necessary to cure the prejudice.” Id. at 5. But to obtain the “particularly harsh” sanctions listed in Subsection (e)(2) — including adverse inference instructions and terminating sanctions — the court must first find that the party to be sanctioned acted with an “intent to deprive.” If such a finding is made, no separate showing of prejudice is required, because “the finding of intent [to deprive] ... support[s] ... an inference that the opposing party was prejudiced by the loss of information.” Id. (quoting Fed. R. Civ. P. 37(e) adv. comm. note to 2015 amendment).

Magistrate Judge Moses next addressed whether the encryption keys were destroyed after a duty of preservation arose, noting that “[t]he first element of the traditional spoliation test, which is also applicable to ESI under Rule 37(e), requires the moving party to demonstrate that the spoliating party had an obligation to preserve the evidence at the time it was destroyed.” Id. at *6 (internal quotations omitted). The obligation to preserve evidence arises when the party has notice that the evidence is relevant to litigation or when a party should have known that the evidence may be relevant to future litigation. Magistrate Judge Moses noted that Defendants could not have had specific notice of the current litigation at the time of the destruction of the encryption keys, because the litigation did not commence until a year later, but she nonetheless found that Defendants knew or should have known that the evidence they destroyed “may be relevant to future litigation.” In particular, she found that Defendants were aware of the investigations into the international bribery scheme and other related information before the destruction of the encryption keys and that those kinds of investigations tend to lead to civil lawsuits. As a result, she found that Defendants “were on notice that litigation was likely and that the information would be relevant to that litigation.”

Magistrate Judge Moses next addressed whether Plaintiffs were prejudiced by the loss of the encryption keys. Plaintiffs argued they were prejudiced because the unavailable MyWebDay data would reveal the full scope and extent of the scheme, which affected several elements of their claim, including materiality, scienter, reliance, and loss causation. Id. at *7. Defendants argued that the only relevant issue affected by the loss was the total volume of bribe payments at issue, which could be resolved by a stipulation that Defendants would not raise any claim or defense on the basis that the total amount of bribe payments was any amount less than $3.3 billion. Magistrate Judge Moses noted that Rule 37(e)(1) does not clearly define the meaning of the term “prejudice” but rather “leaves judges with discretion to determine how best to assess prejudice in particular cases.” Id. (citing Fed. R. Civ. P. 37(e)(1) adv. comm. note to 2015 amendments). Accordingly, she found that although Plaintiffs did not need the level of detail regarding the bribe payments they claimed to need, Plaintiffs had in fact been prejudiced because “the more granular and specific information that apparently resided in MyWebDay would be helpful to them in connection with establishing several elements of their securities fraud claims.”

Magistrate Judge Moses next found that the lost information could not be replaced through additional discovery. She noted that “[t]his is not a typical ESI spoliation case, in which the missing data — for example, emails improperly deleted from a party’s account — can be restored or replaced from another source, such as the accounts to or from which the missing emails were sent.” Id. at 8. She noted theoretical possibilities suggested by Defendants that employees may have emailed the secret MyWebDay information to one another unencrypted. or that another copy of the MyWebDay system may exist that is accessible without the destroyed physical encryption keys, but she discounted these and found that sanctions were warranted under Rule 37(e)(1) because “[t]here is ... every reason to believe that the information at issue here is permanently lost.”

Addressing the question of intent, Magistrate Judge Moses found that Plaintiffs could not demonstrate that the Rule 37(e)(2) sanctions they sought were appropriate because that required a finding that Defendants acted with the intent to deprive Plaintiffs of the information’s use in the litigation. She noted that the intent standard “is both stringent and specific: the intent contemplated by Rule 37 is not merely the intent to perform an act that destroys ESI but rather the intent to actually deprive another party of evidence.” Magistrate Judge Moses also noted that it was Plaintiffs’ burden to demonstrate that Defendants acted with the intent to deprive, not merely the intent to destroy. Ultimately, she found that Plaintiffs could not show that Defendants destroyed the physical encryption keys with the intent of depriving Plaintiffs in this litigation of that evidence, and that the more severe sanctions under Rule 37(e)(2), including the mandatory adverse inference instruction that Plaintiffs requested, were not warranted. Id at *8-9.

Magistrate Judge Moses therefore addressed the appropriate sanction under Rule 37(e)(1). Because “discovery is not yet complete and the precise scope of the issues that will be presented to the jury is not yet known,” she found the appropriate sanction was to allow Plaintiffs to present evidence and argument to the jury concerning Defendants’ intentional destruction of the physical encryption keys needed to access the MyWebDay system, and the jury would be permitted to consider that evidence, along with all the other evidence in the case, in making its decision. Id. at *9. She found this sanction (as distinguished from a mandatory or permissive adverse inference instruction) is permitted under Rule 37(e)(1) without any predicate finding of intent to deprive.

3. A U.S. District Court for the Western District of Virginia ruling that a defendant’s representatives’ continuous disregard for the court and its orders to provide the plaintiff with discovery warranted an adverse-inference jury instruction

In Sines v. Kessler, 2021 WL 1208924 (W.D. Va. March 30, 2021), Magistrate Judge Joel C. Hoppe found that one defendant’s representatives’ continuous disregard for the court and its orders to provide Plaintiff with relevant, discoverable materials warranted an adverse-inference jury instruction under Federal Rules of Civil Procedures 37(b)(2) and 37(e).

This civil rights action arose from the “Unite the Right” rallies held by a number of defendants in Charlottesville, Virginia, in August 2017. Plaintiffs, several local residents who were injured during the events, filed suit against the defendants, alleging that they “conspire[ed] to engage in violence against racial minorities and their supporters,” in violation of the Civil Rights Act of 1871 and related state laws. Id. at *5. Plaintiffs also alleged that one Defendant, a white nationalist group, had a key role in planning the rallies, primarily via invite-only online platforms its members used to communicate. Plaintiffs cited a number of messages from this platform, which referenced various acts of violence before, during, and after the Unite the Right events.

After Plaintiffs initially filed suit, Defendant’s officer and managing agent, Dillon Hopper, accepted service of process on its behalf. He then hired an attorney to represent Defendant. In response to Plaintiffs’ first set of requests, Hopper provided “vague, evasive, and materially incomplete” responses. Id. at *6. He also indicated that he had not taken steps to preserve potentially responsive materials. Hopper and Thomas Rousseau, one of Defendant’s other agents, continued similar discovery misconduct over the next 18 months, including disobeying a number of the court’s orders related to discovery. As a result, “the litigation slowed and everyone else’s costs piled up.” From the onset of the litigation, Hopper’s “consistent practice” was to “ignore outright the court’s orders or submit chaotically and defectively to them.” Id. (citations omitted). For these reasons, Magistrate Judge Hoppe had previously found that Hopper acted in bad faith and that Defendant’s “repeated and ongoing misconduct so far ha[d] caused significant procedural prejudice to Plaintiffs’ ability to resolve their claims in a just, speedy, and inexpensive manner.” Id. at *7. Following this finding, Hopper wrote a letter to the court assuring it that Defendant would work diligently to comply with the court’s orders. Believing that production was preferable to sanctions and awaiting deposition testimony of Defendant’s agents regarding preservation of evidence, Magistrate Judge Hoppe held off on imposing sanctions at that time.

During Defendant’s representatives’ depositions, they testified that, though they were Defendant’s leadership and social media managers during the relevant time period, they were unable to access any of the information in these accounts, including the usernames and passwords. Hopper claimed he had not understood the requests for documents he had received, though he did not indicate whether he had requested assistance from Defendant’s attorney. Furthermore, despite agreeing during his deposition to submit his phone to a vendor for processing, Hopper failed to do so for five months following his deposition. When he finally provided the phone, the vendor was unable to collect relevant ESI. During Rousseau’s deposition, he testified that he had not taken any steps to preserve information relevant to the litigation. Instead, he stated that he usually set up social media accounts so that messages would periodically delete automatically. Id. at *7–*8. 

Plaintiffs filed a motion for sanctions, requesting a jury instruction that Defendant intentionally withheld discoverable documents and allowing the jury to draw adverse inferences from that fact, including that agents of Defendant “chose to withhold such documents because [the agents were] aware that such documents contained evidence that [Defendant] conspired to plan racially-motivated violence” at the Unite the Right rallies. Id. at *1.

Magistrate Judge Hoppe first discussed relevant portions of Rules 26 through 37 of the Federal Rules of Civil Procedure. He noted that courts depend on “the good faith and diligence of counsel and the parties in abiding by these rules and conducting themselves and their judicial business honestly.” Id. at *2. When the parties fail to do so, the Rules allow the court to sanction a misbehaving party through Rules 37(b)(2) and 37(e).

Rule 37(b)(2) allows a district court to impose sanctions when a court order is in effect and that order has been violated. The Fourth Circuit’s framework for which sanction to impose requires a court to consider “(1) whether the non-complying party acted in bad faith, (2) the amount of prejudice that noncompliance caused the adversary, (3) the need for deterrence of the particular sort of non-compliance, and (4) whether less drastic sanctions would have been effective.” Id. at *3. Adverse-inference jury instruction sanctions require the court to find that the party acted willfully or in bad faith.

Rule 37(e) dictates the court’s ability to impose sanctions when a party has failed to preserve ESI. A spoliation sanction may be appropriate if (1) the ESI should have been preserved; (2) the ESI was lost; (3) the loss was caused by a party’s failure to take reasonable steps to preserve the ESI; and (4) the ESI cannot be restored or replaced through additional discovery. The first factor is met when the court finds the party should have reasonably anticipated litigation and should have known that the evidence at issue would be relevant to that litigation. If the movant establishes this, the court must then determine whether the movant has also established one of two options that permit imposing sanctions. First, if the court finds there was prejudice to another party from loss of the information, it may impose sanctions “no greater than necessary to cure the prejudice.” Id. (citation omitted). Second, if the court finds a party intended to deprive another party of information, the court may presume that the information would have been unfavorable to the party who lost it and may instruct the jury that it may or must so presume, or dismiss the case or enter a default judgment.

Based on these rules, Magistrate Judge Hoppe found that Defendant’s “continued disregard for th[is] Court and its orders to provide or permit discovery of relevant materials within their control, leaves no doubt that a permissive adverse-inference instruction against [Defendant] is warranted under either Rule 37(b)(2) or Rule 37(e).” Id. at *8. (internal quotation omitted). 

First, Defendant had a duty to preserve the requested ESI and other materials that were lost. The duty to preserve evidence arises as soon a party “reasonably should know that the evidence may be relevant to anticipated litigation.” Id. (quoting Silvestri v. Gen. Motors Corp., 271 F.3d 583, 591 (4th Cir. 2001)). Defendant’s duty arose, at the latest, when Hopper was served with the summons and complaint. Despite this, Hopper took “no special steps” to preserve relevant materials or to identify other key players with relevant information. Id. at *9. Rousseau continued to use automatic deletion features when he was able. According to their deposition testimony, neither Hopper nor Rousseau could recall how to log into the relevant accounts. Thus, these materials were lost and could not be restored.

Second, Defendant’s failure to produce these relevant materials harmed Plaintiff’s ability to prove their allegations. In attempting to prove a conspiracy, which is inherently difficult, Plaintiffs sought documents that would demonstrate Defendant’s agreements with co-conspirators to engage in racially motivated violence. Hopper and Rousseau’s failure to preserve and produce these types of materials “leaves an evidentiary gap justifying an appropriately tailored sanction against [Defendant].” Id. (internal quotation omitted).

Third, Defendant’s agents “acted with the intent to deprive [the plaintiffs] of the information’s use in th[is] litigation.” Id. (quoting Fed. R. Civ. P. 37(e)). Magistrate Judge Hoppe had first found that Hopper acted in bad faith over a year earlier. Following this finding, Hopper claimed that the discovery responses had slipped his mind and assured the court he would work diligently to respond to the requests. However, it turned out it had not slipped Hopper’s mind — in fact, Hopper had been discussing the litigation online. Hopper merely disregarded the lawsuit. Rousseau’s actions indicated that he was similarly resistant and indifferent to his obligations. “On this record, Hopper’s and Rousseau’s failure to preserve and produce materials that ‘naturally would have elucidated a fact at issue permits an inference’ that they knew giving such information to Plaintiffs ‘would have exposed facts unfavorable’ to [Defendant] or its case.” Id. (quoting Vodusek v. Bayliner Marine Corp., 71 F.3d 148, 156 (4th Cir. 1995)). 

Finally, “sanctions less severe than a permissive adverse-inference instruction” would not be effective. Id. at *10. Attorney’s fees for the procedural prejudice suffered by Plaintiffs had already been awarded. Hopper and Rousseau continued to fail to produce requested materials, and their reasons for being unable to do so were not always believable. Magistrate Judge Hoppe determined that Defendant “clearly is unwilling to obey this Court’s orders even under threat of severe Rule 37 sanctions.” Magistrate Judge Hoppe therefore granted plaintiffs’ request for a permissive adverse-inference instruction, subject to the presiding district judge’s approval.

4. A U.S. District Court for the Eastern District of California decision granting in part and denying in part a plaintiff’s request for precertification discovery of a full data set of electronic timekeeping and payroll information in native format, finding that producing the full set would be disproportional to the needs of the case and ordering the defendant employer to produce a sample of the data instead

In Tinnin v. Sutter Valley Medical Foundation, 2021 WL 1315435 (E.D. Cal. April 7, 2021), Magistrate Judge Erica P. Grosjean granted in part the Plaintiff’s request for precertification discovery of a full data set of electronic timekeeping and payroll information in native format, ordering the Defendant employer to produce a sample of the data.

In this Fair Labor Standards Act case, Plaintiff sought classwide discovery in support of her claim, which Defendant opposed in light of its pending motion to dismiss, stay, or strike the operative complaint. Id. at *1. The court had granted Plaintiff permission to file a motion to compel classwide discovery, and Plaintiff requested, among other things, a native electronic format production of all payroll records and timecards/sheets for the class. Id. at *2. Plaintiff claimed to need this information to compare it to non-native (i.e., paper) time records and ensure that there was overlap from the sampling of paper records and argued that a sampling of the native electronic data would not permit her to do this.

Defendant opposed production of this material primarily on the ground that production would be “extremely burdensome.” The court initially found Defendant’s burden argument insufficiently particularized by this assertion, but permitted Defendant to file a declaration supporting its objection. Defendant later filed a declaration from its director of payroll, who explained, among other things, that there were limitations on the number of entries that could be output in a single data file and would likely require production of more than 500 separate data files. Defendant argued that would increase the burden on it to produce the files because it would require a crew of people to monitor and oversee production of each file, consisting of approximately half a dozen employees or contractors at significant expense, and would take several months. Id. at *2-3. The declaration also noted that the payroll department was “stretched thin due to ‘extraordinary demands’ in dealing with pandemic issues and administering special COVID-related pay rules” and could not “easily spare a team of people to devote the many dozens of hours that would be needed to produce the class-wide discovery.” Id. at *3.

Magistrate Judge Grosjean began her analysis with a discussion of the standards for precertification discovery: “In determining whether to grant precertification class discovery, a court should consider its need, the time required, and the probability of discovery resolving any factual issue necessary for the determination.” In this context, she noted that the parties did not dispute the relevance of the requested discovery but only whether it would be unduly burdensome for Defendant to produce the full set of responsive payroll and timekeeping electronic records in native Excel format as Plaintiff requested. Id. at *4.

Magistrate Judge Grosjean stated that Defendant, as the party opposing discovery, had the burden of supporting its objections to the Plaintiff’s requests. She found that Defendant had not initially provided sufficient support for its assertion of undue burden by arguing that it would be “extremely burdensome” but had later provided sufficient evidence of burden in the form of the declaration by the payroll director. In particular, Magistrate Judge Grosjean found that Defendant had supported its argument that producing the records Plaintiff sought in the requested Excel format for a class of 7,400 employees would be time consuming and require significant manual labor to retrieve, format, process, and manage.

Magistrate Judge Grosjean then found that producing the requested data on a sampling basis was appropriate. She noted that courts have discretion in controlling the scope of precertification discovery and routinely order sampling of classwide records when production is unduly burdensome. Id. at 5 (citing cases). Magistrate Judge Grosjean found that “at this early stage of the proceedings, when a scheduling order has not yet issued and a motion to dismiss is pending, the broad discovery Plaintiff seeks is unduly burdensome and is not proportional to the needs of the case. Sampling is an appropriate method to relieve the burden on Defendant in light of the number of employees involved and the limitations of Defendant’s systems.”

Magistrate Judge Grosjean next addressed the format of production. Plaintiff requested the files in native Excel format, but Defendant argued that extracting the files in Excel format would result in “severe readability problems” and preferred to produce the files in PDF format (which Plaintiff found less useful). Ultimately, Magistrate Judge Grosjean decided that Plaintiff could select its preferred format in light of the advantages and disadvantages of the various production formats.

Finally, Magistrate Judge Grosjean addressed the appropriate sample size. Defendant had proposed a sample size of 10%, whereas Plaintiff had proposed sampling at a rate of 75%. In light of the circumstances of the case, and balancing the relevance of the information with the burden on Defendant, Magistrate Judge Grosjean found that a sample size of 20% was reasonable given the estimated putative class size of approximately 7,400 employees. Id. (surveying cases).