Hon. Ronald J. Hedges
There have been a number of federal decisions since December 1, 2015, which have addressed whether “steps” taken by parties which have “lost” electronically stored information (ESI) had been reasonable ones under Federal Rule of Civil Procedure 37(e). These decisions have sometimes found that the steps under review had, in fact, been unreasonable. Thus, we have numerous examples of what unreasonable steps a party (as well as a third-party or subpoenaed non-party) took to avoid the loss of ESI.

Unfortunately, we do not have any single compilation of what reasonable steps a party could take to avoid (1) the loss of ESI in the first instance and (2) the imposition of remedial measures (or sanctions) under Rule 37(e), although there are two excellent articles by a leading commentator on the 2015 amendments, Thomas Y. Allman, which point the way. T. Allman, ‘Reasonable Steps’: A New Role for a Familiar Concept, 14 DDEE 591 (2014); T.Y. Allman, A Second Look at ‘Reasonable Steps’: A New Role for a Familiar eDiscovery Concept, 15 DDEE 485 (2015); Sweet Seventeen?: The Latest Cases on Rule 37(e), 85 USLW 1059; T.Y. Allman, Amended Rule 37(e): The first Two Years of ‘Reasonable Steps,’ 20 DDEE 484 (2017).

Existing Guidance

Nonetheless, there are legislative histories and case law that can offer guidance — both positive and negative — on what reasonable steps might be. This paper offers some thoughts on the subject.

We should start at the beginning. The Preliminary Draft of Proposed Amendments to the Federal Rules of Bankruptcy and Civil Procedure [Reasonable Steps] (August 2013), proposed that Rule 37(e) be completely revised from its original 2006 text and include, among other things, “[f]actors to be considered in assessing a party’s conduct:”

The court should consider all relevant factors in determining whether a party failed to preserve discoverable information that should have been preserved in the anticipation or conduct of litigation, and whether the failure was willful or in bad faith. The factors include:

  • (A) the extent to which the party was on notice that litigation was likely and that the information would be discoverable;
  • (B) the reasonableness of the party’s efforts to preserve the information;
  • (C) whether the party received a request to preserve information, whether the request was clear and reasonable, and whether the person who made it and the party consulted in good faith about the scope of preservation;
  • (D) the proportionality of the preservation efforts to any anticipated or ongoing litigation; and
  • (E) whether the party timely sought the court’s guidance on any unresolved disputes about preserving discoverable information. [Proposed Rule 37(e)(1)(B)(2), Preliminary Draft at 316-17].

The Committee Note to this proposed subsection described it as follows:

“These factors guide the court when asked to adopt measures under Rule 37(e)(1)(A) due to loss of information or to impose sanctions under Rule 37(e)(1)(B). The listing of factors is not exclusive; other considerations may bear on these decisions, such as whether the information not retained reasonably appeared to be cumulative with materials that were retained. With regard to all these matters, the court’s focus should be on the reasonableness of the parties’ conduct.

The first factor is the extent to which the party was on notice that litigation was likely and that the information lost would be discoverable in that litigation. A variety of events may alert a party to the prospect of litigation. But often these events provide only limited information about that prospective litigation, so that the scope of discoverable information may remain uncertain.

The second factor focuses on what the party did to preserve information after the prospect of litigation arose. The party’s issuance of a litigation hold is often important on this point. But it is only one consideration, and no specific feature of the litigation hold — for example, a written rather than an oral hold notice — is dispositive. Instead, the scope and content of the party’s overall preservation efforts should be scrutinized.  One focus would be on the extent to which a party should appreciate that certain types of information might be discoverable in the litigation, and also what it knew, or should have known, about the likelihood of losing information if it did not take steps to preserve. The court should be sensitive to the party’s sophistication with regard to litigation in evaluating preservation efforts; some litigants, particularly individual litigants, may be less familiar with preservation obligations than other litigants who have considerable experience in litigation. Although the rule focuses on the common law obligation to preserve in the anticipation or conduct of litigation, courts may sometimes consider whether there was an independent requirement that the lost information be preserved. The court should be sensitive, however, to the fact that such independent preservation requirements may be addressed to a wide variety of concerns unrelated to the current litigation. The fact that some information was lost does not itself prove that the efforts to preserve were not reasonable.

The third factor looks to whether the party received a request to preserve information. Although such a request may bring home the need to preserve information, this factor is not meant to compel compliance with all such demands. To the contrary, reasonableness and good faith may not require any special preservation efforts despite the request. In addition, the proportionality concern means that a party need not honor an unreasonably broad preservation demand, but instead should make its own determination about what is appropriate preservation in light of what it knows about the litigation. The request itself, or communication with the person who made the request, may provide insights about what information should be preserved. One important matter may be whether the person making the preservation request is willing to engage in good faith consultation about the scope of the desired preservation.

The fourth factor emphasizes a central concern — proportionality. The focus should be on the information needs of the litigation at hand. That may be only a single case, or multiple cases. Rule 26(b)(1) is amended to make proportionality a central factor in determining the scope of discovery.  Rule 37(e)(2)(D) explains that this calculation should be made with regard to ‘any anticipated or ongoing litigation.’ Prospective litigants who call for preservation efforts by others (the third factor) should keep those proportionality principles in mind.

Making a proportionality determination often depends in part on specifics about various types of information involved, and the costs of various forms of preservation. The court should be sensitive to party resources; aggressive preservation efforts can be extremely costly, and parties (including governmental parties) may have limited resources to devote to those efforts. A party may act reasonably by choosing the least costly form of information preservation, if it is substantially as effective as more costly forms. It is important that counsel become familiar with their clients’ information systems and digital data -including social media — to address these issues. A party urging that preservation requests are disproportionate may need to provide specifics about these matters in order to enable meaningful discussion of the appropriate preservation regime.

Finally, the fifth factor looks to whether the party alleged to have failed to preserve as required sought guidance from the court if agreement could not be reached with the other parties. Until litigation commences, reference to the court may not be possible. In any event, this is not meant to encourage premature resort to the court; amendments to Rule 26(f)(3) direct the parties to address preservation in their discovery plan, and amendments to Rule 16(c)(3) invite provisions on this subject in the scheduling order. Ordinarily the parties’ arrangements are to be preferred to those imposed by the court. But if the parties cannot reach agreement, they should not forgo available opportunities to obtain prompt resolution of the differences from the court. [Committee Comment to Proposed Rule 37(e)(1)(B)(2), Preliminary Draft at 325-28].”

Although the amendment of Federal Rule 37(e) that was adopted effective December 1, 2015 does not address what might be reasonable steps, the Arizona Rules of Civil Procedure do. Analogies are suspect, especially when using the text of a State rule to interpret the meaning of a federal one. However, Arizona Rule 37(g), amended effective July 1, 2018, follows the text of Federal Rule 37(e). The Arizona rule includes subsection 37(g)(1)(C)(ii), which addresses “Reasonable Steps to Preserve:”

“Factors that a court should consider in determining whether a party took reasonable steps to preserve relevant electronically stored information include the nature of the issues raised in the action or anticipated action, the information’s probative value, the accessibility of the information, the difficulty in preserving the information, whether the information was lost as a result of the good-faith routine operation of an electronic information system or the good-faith and consistent application of a document retention policy, the timeliness of the party’s actions, and the relative burdens and costs of a preservation effort in light of the importance of the issues at stake, the parties’ resources and technical sophistication, and the amount in controversy.”

Perhaps these factors, together with those of the 2013 draft proposal to amend federal Rule 37(e), might inform courts and parties on what reasonable steps to avoid the loss of ESI might be.

Regardless, we do have case law on what might be reasonable or unreasonable. Take, as an example, GN Netcom, Inc. v. Plantronics, No. 12-1318-LPS, a civil action that was pending in the District of Delaware. The final decision at the trial level appears to have been filed January 3, 2018. Suffice it to say that sanctions under Rule 37(e) were imposed on the plaintiff for the loss of ESI after a corporate manager disposed of relevant and discoverable ESI and instructed several employees to do the same. This suggests a very reasonable step — avoid self-collection, at least when corporate personnel might be implicated in some scheme, and supervise preservation efforts by personnel when they are being instructed to preserve and self-collect.

Other lessons can be derived from Moody v. CSX Transportation, Inc., No. 07-CV-6398P (W.D.N.Y. Sept. 21, 2017). Moody involved the loss of ESI from an event recorder in a locomotive. ESI relevant to a personal injury accident had been downloaded from the recorder to a laptop. The laptop crashed a year or more after the downloading and the laptop was recycled or destroyed thereafter. Moreover, the lost ESI was supposed to have been uploaded to a central repository but could not be accessed. The court imposed an adverse inference instruction against the defendants to address the “evidentiary gap caused by defendants’ loss of such material evidence.” Among other things, the court found that “the defendants’ explanation for the loss of the data strains credulity” and that “[t]he proposition that a sophisticated railroad transportation corporation such as CSX could be involved in a serious accident in which an individual lost a limb and thereafter fail for four years to review critical data relating to how that accident occurred is unfathomable.” In other words, the defendants acted unreasonably in destroying or recycling the laptop and in failing to confirm that the ESI had been uploaded successfully pursuant to an established corporate procedure.

Moody and GN Netcom provide examples of what steps might be found to have been unreasonable. Other decisions could suggest other examples.

All of the above suggests what might be reasonable steps. These include:

  • Establish a process for preservation of ESI. This will help avoid ad hoc responses to a duty to preserve.
  • Document whatever process is used to preserve specific ESI.
  • Consider the benefit of written vs. oral litigation hold notices.
  • A ‘‘form’’ of litigation hold notice as well as a protocol to identify custodians of ESI for receipt of notices.
  • A procedure to identify custodians who may separate from employment so that any ESI in their possession can be preserved.
  • Issue litigation hold notices beyond the ‘‘four corners’’ of a party to third parties such as consultants, vendors, etc.
  • Include in agreements with third parties a provision requiring third parties to preserve ESI on request.
  • A means to monitor and modify holds.
  • A means to end holds in a timely and effective manner.

Having described what might be reasonable, it should be understood that any inquiry into whether a step was actually reasonable will be fact-sensitive. For example, take one of the above suggestions. Presumably, a written hold notice would always be reasonable. But what about an oral one? How might an oral hold notice ever be communicated in an effective (“reasonable”) way? The answer, as with answers to many legal questions, is “it depends.” An oral hold notice might be problematic. Nevertheless, it might be reasonable for a small entity. See Orbit One Communications, Inc. v. Numerex Corp., 271 F.R.D. 429, 441 (S.D.N.Y. 2010).

The elusive role of proportionality

Another problematic “step” might relate to the role of proportionality in making preservation decisions. We have sparse case law on this. An early discussion of proportionality and preservation appeared in Pippins v. KPMG LLP, 279 F.R.D. 245 (S.D.N.Y. 2012). There, the defendant argued that it would be disproportionate to require the preservation of over 2,500 hard drives. The district court rejected that argument:

“Preservation and production are necessarily interrelated. The application of the proportionality principle to preservation flows from the existence of that principle under the Federal Rules of Civil Procedure. Federal Rule of Civil Procedure 26(b)(2)(C)(iii) establishes a ‘proportionality’ test for discovery, and requires courts to limit the ‘frequency or extent of discovery’ where ‘the burden or expense of the proposed discovery outweighs its likely benefit considering the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the action, and the importance of the discovery in resolving the issues.’ Therefore, proportionality is necessarily a factor in determining a party’s preservation obligations. See [Orbit One Commc’ns, 271 F.R.D. at 436 n. 10] (‘Reasonableness

and proportionality are surely good guiding principles for a court that is considering imposing a preservation order. . . .’)

Because proportionality is a ‘highly elastic concept…[it] cannot be assumed to create a safe harbor for a party that is obligated to preserve evidence but is not operating under a court-imposed preservation order.’…As [US Magistrate] Judge Cott appropriately cautioned,

proportionality ‘may prove too amorphous to provide much comfort to a party deciding what files it may delete or backup tapes it may recycle’ before that party files a motion for a protective order seeking to have a court define its preservation obligations. … ‘Accordingly, ‘[u]ntil a more precise definition is created by rule,’ prudence favors [either] retaining all relevant materials,’ or swiftly moving for a protective order. However, proportionality is at the very least relevant to a decision on a motion for a protective order, even if not determinative of it. [279 F.R.D. at 255-56 (citations largely omitted)].’’

The district court concluded that the cost of preserving the hard drives in issue did not outweigh any benefit because the record did not support such an analysis. This ruling, made before amended Rule 37(e) became effective, demonstrates the fact-specific nature of any consideration of the ‘‘reasonableness’’ of proportionality in making decisions about the scope of preservation, although actual ‘‘loss’’ of ESI was not at issue in Pippins.

ML Healthcare Services, LLC v. Publix Super Markets, Inc., No. 15-13851 (11th Cir. Feb. 7, 2018), offers a rare appellate discussion of how proportionality can factor into 37(e). This action arose out of a slip-and-fall in a supermarket. Despite receipt of preservation demands from the plaintiff, the defendant, consistent with its policy, retained only one hour of video related to the slip-and-fall. The district court rejected the plaintiff’s argument that sanctions should have been imposed pursuant to Rule 37(e). The court of appeals affirmed:

“The district court did not abuse its discretion by concluding that Defendant’s failure to retain more video did not constitute bad faith or demonstrate an intent to deprive Plaintiff of evidence necessary to her case. Defendant immediately saved the most relevant portion of the video-the hour during which Plaintiff’s fall occurred, which covered the entire time Plaintiff was in the store-before any request for preservation or notice of litigation was provided. This initial preservation fulfills the request of Plaintiff’s first two preservation letters. As to Plaintiff’s subsequent preservation letters, the requests in those letters encompassed all video media from every camera at the store for a period of thirty five days-totaling 840 hours of video per camera, assuming the cameras run for 24 hours a day. Defendant might reasonably, and in good faith, have concluded that it did not have to comply with such a broad and far-reaching request.

Nor is there any other evidence of bad faith. First, there is no indication that Defendant destroyed the evidence in a manner inconsistent with its normal video retention policies. Second, this is not the sort of case where the unpreserved evidence clearly would have resolved a crucial issue in the case.  ***. Plaintiff hoped the video would contradict the testimony of Defendant’s employees that inspections were conducted at 5:00 and 6:00 a.m. on the morning of Plaintiff’s accident and that the aisle had been auto-scrubbed the night before the accident, with an employee following the machine to pick up excess liquid. Further, Plaintiff hoped the video would show that an employee’s stocking cart traversed the area of the fall, from which the jury might infer that the cart was leaking some sort of liquid, which dripped on the spot of the fall and remained on the floor until Plaintiff came into contact with it.  Yet, Plaintiff never narrowed her request to cover just those periods of time. Moreover, Plaintiff provides no reason to believe that the video would have actually shown these things such as conflicting testimony, inconsistent statements, or observations from other witnesses. Failure to preserve such speculative evidence does not raise the specter of bad faith in the same way that a failure to preserve evidence of a specific, crucial event in a case might. [slip op. at 31-32].”

Case law on proportionality as a “reasonable step” is indeed sparse. However, ML Healthcare Services is consistent with the interplay between proportionality and preservation noted in Pippins. It is also consistent with one of the goals of the 2015 amendment of Rule 26(b)(1) of the Federal Rules of Civil Procedure, which was to “crystalize[] the concept of reasonable limits on discovery through increased reliance on the common-sense concept of proportionality ***.”  2015 Year-End Report on the Federal Judiciary (Dec. 31, 2015).


How should this examination of what might constitute “reasonable steps” end? Everyone is familiar with the concept of reasonableness. “The word ‘reasonable’ is a paradigmatic example of a standard in the law, and its meaning is, if nothing else, vague.” B.C. Zipursky, “Reasonableness in and out of Negligence Law,” 163 U. Penn. L. Rev. 2131, 2133 (2015). For our purposes, unreasonable steps might be defined as negligent ones. See Leidig v. Buzzfeed, Inc., 16 Civ. 542 (VM) (S.D.N.Y. Dec. 19, 2017) (“While there is insufficient evidence that plaintiffs acted with an intent to deprive Buzzfeed of the [lost] evidence, the plaintiffs’ actions were certainly negligent.”) (footnote omitted).

Perhaps we cannot reduce what might be reasonable steps to a checklist. Or perhaps we can if we look to the definitions suggest by the early draft of the 2015 amendments, the Arizona rules, and the “good” and “bad” examples offered by case law.

*Reproduced with permission from Digital Discovery & e-Evidence, 18 DDEE 143, 3/1/18. Copyright  2018 by The Bureau of National Affairs, Inc.


Ronald J. Hedges is Senior Counsel with Dentons US LLP. He served as a United States Magistrate Judge in the District of New Jersey from 1986 to 2007, and has extensive experience in ediscovery, frequently speaking and writing as an expert on the subject. He has authored numerous publications, including Managing Discovery of Electronic Information, Third Edition.