MEMORANDUM AND ORDER
Before the Court is Defendants’ oral motion for Plaintiffs to pay all or part of the costs relating to the discovery of Defendants’ electronic mail. Plaintiffs filed a Brief Addressing Cost^Shifting for Discovery of Electronic Mail (Doc. 58), opposing the shifting of any cost burden from Defendants to Plaintiffs. Defendants filed a Memorandum Addressing Cosi^Shifting for Discovery of Electronic Mail (Doc. 63), in support of its oral motion, and Plaintiffs filed an additional Reply Memorandum Regarding Shifting Costs of Discovery of Electronic Evidence (Doc. 66). After discussion about this matter at a scheduling conference, Plaintiffs filed a Sup*632plemental Memorandum (Doe. 117), and Defendants also filed a Supplemental Memorandum, with an attached Affidavit of Kevin Norman (hereafter Norman Aff.), and excerpts from several deposition transcripts. (Doc. 121).
BACKGROUND
Plaintiffs, female officers of the City of Wichita’s (“City”) police department, filed suit against the City, the Police Department, and Chief Norman Williams in his individual and official capacities, alleging a multitude of violations of federal and state law. Plaintiffs allege that, while working for the City, they were subject to, among other things, sexual harassment, hostile work environment, gender discrimination, and violations of their equal protection and due process rights in violation of various federal and state laws. See generally, Plaintiffs’ Amended Complaint. (Doe. 53).
The case was filed as a class action, but the District Court denied Plaintiffs’ motion to certify a class. (Doc. 86). The undersigned magistrate judge later denied Plaintiffs’ motion to amend their complaint in order to reassert class claims. (Doc. 118). As a result, this ease now involves only the claims of the four named plaintiffs.
During discovery, Plaintiffs requested information relating to e-mails sent to or by other officers within the Wichita Police Department. Through negotiations, the parties narrowed that request to copies of e-mails from 117 different supervising officers to the extent that such e-mails exist on a back-up tape of July 23, 2004.
Defendants store e-mail information only on (1) their active user e-mail files and (2) on back-up tapes. Plaintiffs assert that a search of current active user files is inadequate because there is no way to obtain information regarding past deleted e-mails. Thus, the back-up tape is the only real source of information regarding e-mails in existence on July 23, 2004, but that may have been deleted thereafter.
The City keeps back-up tapes for disaster recovery purposes only, and not to retrieve information. Thus, in order to conduct a word search of the e-mails contained on a given back-up tape, the City would have to restore the back-up tape to an e-mail server. Because of recent hardware upgrades, the City has an Exchange 5.5 server available for such a restoration. However, despite such hardware availability, the City claims that new software and labor costs would be substantial. The current dispute revolves around who should pay the costs of restoring the back-up tapes and retrieving the requested information from the July 23, 2004, backup tape.1
Kevin Norman is the person who has supervised and monitored work of the City’s IT/IS Department in this case. In his most recent affidavit (Doc. 121, Norman Aff.), Norman reviews the actions the City has taken to date to comply with Plaintiffs’ demands for discovery of electronic data and specifically e-mail.2 There are two ways to search the e-mail files of the 117 supervisors for the specific words Plaintiffs’ have requested: (1) By use of a software program known as Discovery Attender which will search the entire back-up tape at once after it is restored and loaded on the Exchange 5.5 server; and (2) a “manual” search of the copies of 117 e-mail accounts of these supervisors which the City has copied as .pst files. The manual method requires that each .pst file be loaded and then searched using the Microsoft Outlook search feature by searching the subject and message bodies of all emails. Norman notes that under the manual system, each .pst file would have to be searched multiple times depending on the number of keywords desired, then any identified e-mail message would have to be opened *633and copied to a separate folder. (Doc. 121, Norman Aff. Hit 5, 7).
Norman sets out the costs of the proposed search using the two alternative methods. To use the Discovery Attender software program would require that this software be purchased at an estimated cost of $2,224.953 which includes the cost of shipping, handling and minimal support. The software would then be loaded by City IT/IS staff on the server and they would run the search of the back-up tape using the words requested by Plaintiffs. To search the 117 e-mail boxes on the back-up tape for the seventeen (17) words identified by Plaintiffs would take an estimated 8 hours at a rate of $50 per hour, bringing the total estimated cost to $2,624.95. (Doe. 121, Norman Aff. If 6).
To do the search “manually,” i.e., by having the City’s IT/IS staff search the .pst files individually using Microsoft Outlook, is estimated to take 20 minutes for each of the 117 mailboxes for supervisors which, at an hourly rate of $50, would be approximately $1,950. This process is labor-intensive and more subject to human error, and the time (and therefore cost) could be greater in the event numerous documents are located and need to be loaded to other files. (Doc. 121, Norman Aff. 1Í 7-8).
Plaintiffs argue that the cost of this electronic discovery should be borne by Defendants as the producing party, and that requiring the City to expend these costs would not constitute an “undue burden” on the City. The City seeks a protective order which would require that the costs of this search should be borne, at least in part, by Plaintiffs. The City notes that it has no future use for the Discovery Attender software, it would not buy that software absent an order of the Court, there is no evidence that the search of the back-up tape will even find any of the words identified by Plaintiffs for use in a search, and the burden of compliance with Plaintiffs’ requested discovery would be an undue burden on Defendants.
The Court has withheld ruling on this issue in the hope that the parties would be able to reach some compromise concerning e-mail production that would satisfy both sides. During the time since this issue was first raised, the parties have apparently explored various ways to accomplish a search of the specific back-up tape in issue, but no agreement has been reached. Plaintiffs have proposed alternative methods of searching which would have Plaintiffs’ expert conduct the search, but the City has refused these alternatives since the electronic data to be search contains highly confidential information.
DISCUSSION
While there is a presumption that the responding party will bear its own costs of production, the Federal Rules permit the Court to grant a protective order “to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense.” Fed.R.Civ.P. 26(c). Trial courts have discretion in determining when a protective order is appropriate. See Boughton v. Cotter Corp., 65 F.3d 823, 828 (10th Cir. 1995) (stating that a trial judge’s grant of a protective order will be reviewed for abuse of discretion). Additionally, “the party seeking a protective order has the burden to demonstrate good cause.” Horizon Holdings, L.L.C. v. Genmar Holdings, Inc., 209 F.R.D. 208, 214 (D.Kan.2002) (citation omitted). The moving party must make “a particular and specific demonstration of fact, as distinguished from stereotyped and conclusory statements.” Gulf Oil Co. v. Bernard, 452 U.S. 89, 102 n. 16, 101 S.Ct. 2193, 2201, 68 L.Ed.2d 693 (1981) (quotation omitted).
The current version of the Federal Rules of Civil Procedure provide some general guidance in determining what constitutes good cause for granting a protective order. Fed.R.Civ.P. 26(b)(2) provides that
[t]he frequency or extent of use of the discovery methods otherwise permitted under these rules and by any local rule shall be limited by the court of it determines that ... (iii) the burden or expense of the proposed discovery outweighs its likely benefit, taking into account the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the litigation, and the importance of the proposed discovery in resolving the issues.
*634Commentators have opined that this rule “should be used to discourage costly, speculative, duplicative, or unduly burdensome discovery of computer data and systems.” Manual for Complex Litigation (4th) § 11.446.
An amendment to Fed.R.Civ.P. 26(b)(2) will become effective on December 1, 2006, and this amendment was “designed to address issues raised by difficulties in locating, retrieving, and providing discovery of some electronically stored information.” Advisory Committee Note on Amendment to Rule 26(b)(2). Pursuant to amended Rule 26(b)(2)(B),
[a] party need not provide discovery of electronically stored information from sources that the party identifies as not reasonably accessible because of undue burden or cost. On motion to compel discovery or for a protective order, the party from whom discovery is sought must show that the information is not reasonably accessible because of undue burden or cost. If that showing is made, the court may nonetheless order discovery from such sources if the requesting party shows good cause, considering the limitations of Rule 26(b)(2)(C). The court may specify conditions for the discovery.
Therefore, the burden of establishing that the information is not reasonably accessible because of undue burden or cost is on the party responding to the discovery request, in this case the City.
Once it is shown that the source of electronically stored information is not reasonably accessible, the requesting party, here Plaintiffs, may still obtain discovery by showing good cause, considering the limitations of amended Rule 26(b)(2)(C), which balance the costs and potential benefits of the proposed discovery. The amended Rule 26(b)(2)(C)— which is substantially the same as present Rule 26(b)(2)(iii)—provides that
[t]he frequency or extent of use of the discovery methods otherwise permitted under these rules and by any local rule shall be limited by the court if it determines that: ... (iii) the burden or expense of the proposed discovery outweighs its likely benefit, taking into account the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the litigation, and the importance of the proposed discovery in resolving the issues. The court may act upon its own initiative after reasonable notice or pursuant to a motion under Rule 26(c).
A. Defendant’s Choice of Storage Media
As a preliminary matter, Plaintiffs argue that the City’s choice of data storage systems and Plaintiffs’ lack of control over such choice require the Court to deny the motion. Plaintiffs cite In re Brand Name Prescription Drugs § Antitrust Litigation for the proposition that Defendants should not be able to shift the costs of production where the costs are the result of the back-up system chosen and implemented by Defendants. Case No. 94-897, 1995 WL 360526, at *1-3 (N.D.Ill.1995). The Court rejects this bright-line argument.
While courts have, in the past, rejected requests for cost-shifting where the requesting party chose a difficult-to-access storage media, most courts are now moving toward a more sophisticated approach of considering various factors in deciding whether to shift the costs of electronic discovery. Paul M. Robertson & Lawrence T. Stanley, Jr., Electronic Discovery: a Survey of Relevant Law and Recent Developments, Committee on Pretrial Practice and Discovery, vol. 13, No. 1 at 11 (Winter 2005).
The Court is not aware of any decision since Zubulake v. UBS Warburg LLC, 217 F.R.D. 309 (S.D.N.Y.2003) (Zubulake I), that unequivocally prohibits cost-shifting where the Defendant voluntarily utilized storage technology that is difficult to access or restore. Furthermore, the Court has discretion in determining whether to shift the costs of discovery. See Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 358, 98 S.Ct. 2380, 57 L.Ed.2d 253 (1978) (stating, in dicta, that district courts may, in their discretion, condition discovery on the requesting party’s payment of costs); 8 Charles Alan Wright, Arthur R. Miller & Richard L. Marcus, Federal Practice and Procedure, § 2038 (2d ed.1994) (stating that the “court may order the party seeking discovery to pay the expenses caused thereby though it is not required to do so”) (citations omitted).
*635The Court is cognizant of the possibility that potential litigants might intentionally retain stored data in an inaccessible media in order to deter discovery in litigation.4 Where, as here, the storage media is reasonably related to the purposes for which the information is retained, the court will not automatically require that party to bear the costs of retrieving the information. Cf. Oppenheimer Fund, 437 U.S. at 359, 98 S.Ct. 2380 (noting that a defendant may be required to bear the costs of information retrieval when it would perform such retrieval in the ordinary course of its business).
Defendants state, and Plaintiffs do not challenge, that the City’s back-up e-mail information was kept for the sole purpose of disaster restoration. The Advisory Committee Note to the Proposed Amendment to Rule 26(b)(2) acknowledges that certain information “can be located, retrieved, or reviewed only with very substantial effort or expense.” As an example, the Committee identifies information “stored solely for disaster-recovery purposes”—like the information at issue. It is, therefore, apparent that Defendants’ storage choice was not only common, but anticipated by those drafting the amendments to the Federal Rules of Civil Procedure.5 The Court thus finds that backup tapes were a reasonable storage method for such purpose, and assigns no fault to Defendants for utilizing such method.
B. Cost Threshold
Plaintiffs also argue that the low cost of compliance in this case (approximately $2624.95),6 in relation to compliance costs for electronic discovery in other cases, makes any cost-shifting analysis totally unnecessary. Again, the Court disagrees with this bright-line argument.
In an early case, the Supreme Court stated that there may be circumstances where costs should not be shifted because the expenses involved are “so insubstantial as not to warrant the effort required to calculate it and shift it to the [requesting party].” Oppenheimer Fund, 437 U.S. at 359, 98 S.Ct. 2380 (considering cost-shifting in the context of a disclosure of information for the purpose of class notification under Fed.R.Civ.P. 23). The Oppenheimer court cited, as an example of such insubstantial costs, the expense of making files available for inspection. Id. (citing In re Nissan Motor Corp. Antitrust Litigation, 552 F.2d 1088 (5th Cir.1977)). After Oppenheimer, and early in the history of electronic discovery, some courts did focus on whether the cost of compliance was “excessive or inordinate,” but generally refused to set out an ironclad formula which would determine what was “undue” within the meaning of the federal rules. See e.g., Bills v. Kennecott Corp., 108 F.R.D. 459, 463-64 (D.Utah 1985) (refusing to require payment of $5,411.25 by plaintiffs-employees to obtain computer records from their employer). Likewise, courts7, commentators8, and the *636drafters9 of the Federal Rules of Civil Procedure have repeatedly stated that the rules are made intentionally flexible in order to allow courts to consider the relevant factors of each case. “[E]ven a very slight inconvenience may be unreasonable if there is no occasion for the inquiry and it cannot benefit the party making it.” 8 Charles Alan Wright, Arthur R. Miller & Richard L. Marcus, Federal Practice and Procedure, § 2036 (2d ed.1994). Thus, the Court finds that an analysis is necessary to determine whether the cost and expense of complying with the discovery request would be unduly burdensome to Defendants under the facts of this case.
C. Cost-Shifting Factors
The Court has found no Tenth Circuit case dealing with cost-shifting of electronic discovery between parties.10 The current leading case on cost-shifting is Zubulake I, 217 F.R.D. 309 (S.D.N.Y.2003), which has been cited by both parties. See also Zubulake v. UBS Warburg LLC, 216 F.R.D. 280 (S.D.N.Y.2003) (Zubulake III).
In Zubulake I, the court considered whether the plaintiff should bear the cost of complying with her discovery requests relating to the production of e-mails stored on back-up tapes. The defendant estimated that production would cost approximately $175,000.00. Id. at 316. Zubulake I established a two-part test for determining when cost-shifting is appropriate for electronic discovery requests. The first inquiry is whether the requested discovery is unduly burdensome. Id. at 318. The court determined that “whether production of documents is unduly burdensome or expensive turns primarily on whether it is kept in an accessible or inaccessible format.”11 Id. After finding that some of the information was contained on inaccessible back-up tapes and, therefore, unduly burdensome to produce, the court applied a multi-factor test to determine whether cost-shifting was appropriate.
Zubulake I modified the test applied in Rowe Entertainment, Inc. v. The William Moms Agency, Inc., 205 F.R.D. 421 (S.D.N.Y.2002) to create its own test. The Court established seven factors to be considered in descending order of importance: (1) the extent to which the request is specifically tailored to discover relevant information; (2) the availability of such information from other sources; (3) the total cost of production compared to the amount in controversy; (4) the total cost of production compared to the resources available to each party; (5) the relative ability of each party to control costs and its incentive to do so; (6) the importance of the issues at stake in the litigation; and (7) the relative benefits to the parties of obtaining the information. Zubulake I, 217 F.R.D. at 322. The court noted that the first two factors (comprising the “marginal utility test”) were the most important, the second three factors were the next most important, and the final two factors would rarely make any difference, but could be very important in the few cases to which they would apply. Id. at 323.
The analysis in Zubulake I (and the application of those factors in Zubulake III) must be considered in connection with the later amendment to Rule 26(b)(2) which will *637become effective on December 1, 2006. The proposed rule requires a balancing of the costs and potential benefits of discovery, and the Advisory Committee identified several items to be considered, including:
(1) the specificity of the discovery request;
(2) the quantity of information available from other and more easily accessed sources;
(3) the failure to produce relevant information that seems likely to have existed but is no longer available on more easily accessed sources;
(4) the likelihood of finding relevant, responsive information that cannot be obtained from other, more easily accessed sources;
(5) predictions as to the importance and usefulness of the further information;
(6) the importance of the issues at stake in the litigation; and
(7) the parties’ resources.
Advisory Committee Notes to Rule 26(b)(2). The similarity between these considerations and the factors in Zubulake I is readily apparent. The court finds all of these factors to be instructive. The Court agrees, however, that these factors should not be treated simply as a check-list which is mechanically applied to every case and where the decision to shift costs is made solely on the basis of how many factors are present in a particular case. See e.g., Zubulake I, 217 F.R.D. at 322-23; Zubulake III, 216 F.R.D. at 289.
D. Inaccessible Medium and Undue Burden
Here, as in Zubulake, the data sought is on an medium which can be classified as “inaccessible,” i.e., back-up tapes, which must be restored before they can be searched for relevant data. This suggests that the process of producing such data could constitute an undue burden on Defendant. See Zubulake, 217 F.R.D. at 318 (“whether production of documents is unduly burdensome or expensive turns primarily on whether it is kept in an accessible or inaccessible format.”). However, the Amendment to Rule 26(b)(2)(B) makes clear that any inaccessibility must be “because of undue burden or cost.” This brings into play Plaintiffs’ argument that the costs in this case are not so significant as to cause an undue burden on the City.
The City states, however, that it has already spent over $20,000 for work prior to May 17, 2005, in connection with electronic discovery requests from Plaintiffs, including the purchase of back-up tapes and two tape drives. (Doc. 121, Norman Aff. II3; Doc. 63, Norman Aff. 11113-4). The City argues that the total costs—the City’s past expenditures plus any future costs to restore and search the e-mail—are substantial and burdensome. (Doc. 121 at 1-2). For purposes of cost-shifting in this case, the court believes that it should consider not only the future costs to either (1) acquire the search hardware (Discovery Attender), or (2) to conduct a search on the .pst files using Microsoft Outlook, but it should also consider the costs which the City has already incurred that were directly related to restoring or searching e-mails.12
The court has reviewed the itemization of prior work done by the City’s IT/IS Department which was attached to Mr. Norman’s earlier affidavit. (Doc. 63, Norman Aff. H 3). It appears that much of the early work in October and November, 2004, was directed to issues concerning the back-up of the City’s TSM system which was the subject of an earlier motion by the City. See Defendants’ Motion for Protective Order, October 8, 2004, Doc. 24. In that motion, the City was concerned with Plaintiffs’ demand for preservation of electronic data and the effect any such order would have on the City’s automated back-up of the TSM system. At that time, the court was advised that the TSM system was a disaster retrieval system that was separate from the City’s e-mail system and that the City had already created back-up tapes of the City’s e-mail system. (Doc. 27 at 4, n. 1). Therefore it appears to the court *638that the acquisition of the two new tape drives and additional back-up tapes, see Doc. 63, Norman Aff. II4, had nothing to do with preservation of the City’s e-mail system. The court does note that the City staff did spend time creating .pst files for certain listed employees and supervisors and researching issues concerning the Exchange 5.5 server restoration, all of which appear to be directly related to the e-mail search issue. The court can identify such work done on December 2, 3, 6 and 8, 2004, and on March 7, 9 and 11, 2005, which appears to fall in this category. See Doc. 63, Norman Aff. f 3 and attached Consulting Time Report. The total cost reported for those specific activities is approximately $1,150.
Therefore, for purposes of considering the cost to the City of restoring and searching the single e-mail back-up tape, it appears that the total costs would not exceed $3,374.95 (prior staff time of $1,150 and acquisition of Discovery Attender and future staff time to restore and search e-mails of $2,624.95).13 While this pales in comparison to the estimated costs in Zubulake III (approximately $166,000 to restore and search 94 back-up tapes)14 and Quinby ($226,266.60 to restore and search six former employees’ e-mails on an expedited basis),15 it is not an insignificant cost.
While it is somewhat questionable whether these total costs make the e-mails stored on back-up tape “not reasonably accessible because of undue burden or cost,” as required by the proposed amendment to Rule 26(b)(2)(B), the court will proceed to analyze the cost-shifting issue based upon the factors identified in Zubulake I and the amended Rule 26(b)(2)(C).
E. Application of Cost Shifting Factors in this Case
The ultimate question is whether the burden of complying with the discovery request outweighs the likely benefit of the proposed discovery. Zubulake I, 217 F.R.D. at 322-23 (“Put another way, ‘how important is the sought-after evidence in comparison to the cost of production?’ ”). The factors to be considered should be directed toward, and subservient to, this ultimate question.
Thus, the first inquiry should be to ask what are the benefits to be derived from the discovery, particularly: (1) what is the likelihood that the discovery will uncover relevant information, and (2) what is the potential value of that information in resolving the issues in the case. Then, these benefits should be compared to the cost burden resulting from the discovery, again asking particularly: (3) the total cost of the production compared to the amount in controversy, and (4) the total cost of production compared to the resources available to each party. The court believes that these factors should all be given such weight as may be justified by the facts of the individual case as they are all critical to the ultimate question of whether the costs outweigh the likely benefits of production.
The cost for the anticipated restoration and search of the e-mail back-up tape is well established in the record and is not disputed. Neither party has speculated, however, on the possible recovery that these four named plaintiffs might receive were they to prevail in their claims. The court is unable on the record to make any such determination itself. However, since the case now deals only with these four named Plaintiffs and not with a class, any projected recovery would appear to be far less than Plaintiffs might have initially anticipated when the case was filed. Even so, the costs of restoring and searching the e-mail back-up tape does not seem excessive when compared to the possible amount in controversy.
On the other hand, any cost-shifting to Plaintiffs will necessarily be borne now by *639only the four Plaintiffs, and not spread among a larger number of class members. The parties also failed to present any evidence as to the relative financial position of the parties, but the court can properly assume that the four Plaintiffs have significantly less financial capability to pay these costs than does the City. However, it should also be considered that as a governmental entity, the City’s ability to shoulder significant discovery costs is not comparable to the investment banking organizations in both Zubulake and Quinby,16 and the source of any such monies comes from public rather than private sources. Considering all of the cost factors, the court believes that they weigh slightly against any cost-shifting.
The more problematic matter involves the two “benefit” factors.17 Discovery in the case is now concluded except for the unresolved issue concerning the e-mail back-up tape. Plaintiffs note in their Supplemental Memorandum that some e-mails have been produced in discovery and have been discussed in depositions. (Doc. 117 at 3-5). However, the deposition excerpts cited by Plaintiffs do not appear to provide any direct evidence of gender discrimination or retaliation against Plaintiffs. While the court realizes that a party cannot predict with certainty that a proposed discovery method will lead to the discovery of important evidence, the more likely it is that the back-up tape contains important, relevant evidence to Plaintiffs’ claim, the fairer it is to require the City to search at its own expense. Cf. Zubulake
I, 217 F.R.D. at 323 (citing McPeek v. Ashcroft, 202 F.R.D. 31, 34 (D.D.C.2001)).
In Zubulake III, even though several emails had been produced as a result of a sample restoration and search of only a few back-up tapes, the court concluded that the plaintiff had not been able to show that there was indispensable evidence on the back-up tapes, and the success of the full search was somewhat speculative. 216 F.R.D. at 289. The same situation appears to be present in this case. While Plaintiff has identified testimony that there have been some e-mails seen by a Plaintiff that she indicated “would probably upset a female officer,” Doc. 117 at 4, this is not a strong indication that a search of the e-mails of 117 supervisors (many of whom were not the direct supervisors of any of the four Plaintiffs)18 would likely produce direct evidence of gender discrimination or retaliation. The portions of the Plaintiffs’ depositions provided by Defendants reinforces this conclusion. See Doc. 121, Attached Deposition Excerpts. Therefore, the benefit factors weigh in favor of cost-shifting.
Most of the other factors outlined in either Zubulake or the amendments to Rule 26(b)(2), are not particularly relevant to the analysis in this ease. For example, the Court does agree with Zubulake that while other factors such as the importance of the issues at stake in the litigation and the relative benefits to the parties of obtaining the information rarely may be relevant, they may be vital to the analysis under the certain circumstances. Thus, highly important litigation 19 may require greater latitude in cost-*640shifting than more isolated cases, and eases where there is a definite benefit to the responding party may weigh against cost-shifting.20
While this obviously is an important case to the four Plaintiffs, the court is inclined to agree with Zubulake that a workplace discrimination claim probably is not the type of case that warrants special consideration as to cost-shifting. Interestingly, both Zubulake and Quinby, like the present case, involve claims of gender discrimination in the workplace. Although both courts acknowledge that discrimination is an important problem, neither considered such a case to be unique or novel in nature, and therefore gave little or no consideration to this factor. Zubulake III, 216 F.R.D. at 289; Quinby, 2006 WL 2597900 at * 15. As the court stated in Zubulake III,
[i]f I were to consider the issues in this discrimination case sufficiently important to weigh in the cost-shifting analysis, then this factor would be virtually meaningless. Accordingly, this factor is neutral.
216 F.R.D. at 289. Even though suits alleging discrimination in the workplace of a public employer may seem more like “social reform litigation” than suits involving only private employers, the court still believes that it would be unwise to treat this ease as one requiring some special consideration for purposes of cost-shifting.
CONCLUSION
After consideration of the briefs of the parties, and considering the specific circumstances of this case, the court cannot conclude that cost to the City of restoring and searching the single e-mail back-up tape of July 23, 2004, is such as to render that back-up tape “not reasonably accessible because of undue burden or cost.” See Amendment to Rule 26(b)(2)(B). The court is not unsympathetic to the City’s argument that it has already incurred significant costs and expenses in order to satisfy other electronic discovery demands of Plaintiffs in this case; however, the majority of those expenses do not directly relate to the restoration and search of the back-up tape which is the issue now before the court and therefore should not be considered. Cf. Zubulake III, 216 F.R.D. at 290 (as a general rule, where cost-shifting is appropriate, only the costs of restoration and searching should be shifted). Because the e-mails are reasonably accessible without undue burden or cost, the cost-shifting factors in the amendment to Rule 26(b)(2)(C) do not come into play. Cf, Zubu-lake I, 217 F.R.D. at 318 (cost-shifting should only be considered when electronic discovery imposes an “undue burden or expense” on the responding party).
Had the direct cost and expense of restoring and searching the back-up tape been larger in this ease, as may often be the ease with use of back-up tapes, the application of the cost-shifting factors outlined in the amendments to Rule 26(b)(2)(C) and Zubu-lake I would have easily supported a shifting of some of the costs to the Plaintiffs. This is particularly true since it appears to the court that the likelihood that the search of the July 23, 2004, back-up tape would result in the discovery of significant important evidence to support Plaintiffs’ discrimination claims in this case is somewhat questionable and speculative. In that regard, the court does note that Defendants apparently have not attempted to view any of the e-mails on the back-up tape even though e-mails for 117 mailboxes have already been transferred to .pst format and could have been searched with Microsoft Outlook search features. Thus, Defendants, like Plaintiffs, are merely *641speculating about the results of any e-mail search.21
While the court has concluded that cost-shifting is not warranted in this case, the court still retains the authority under Fed. R.Civ.P. 26(b)(2) to limit the extent of proposed discovery where the expense outweighs its likely benefit. After considering all the fact of this case, the court concludes that the scope of the search of the e-mails should be limited in two respects.
First, the court notes in this case the relative cost to restore and search the requested e-mails is much less because Plaintiffs have agreed that only one back-up tape for July 23, 2004, should be searched. On the other hand, Plaintiffs’ designation of the seventeen words to be used for searches22 as to each of the 117 supervisors’ e-mail boxes appears to be too broad in scope. The court agrees with Defendants that several of the words are common enough that they might well result in identification of a significant volume of irrelevant e-mail. Others are words that do not appear to be reasonably calculated to lead to the discovery of relevant evidence in this case. Accordingly, the Court finds that the following words should not be used in any e-mail search: Sara, Heather, Female, Ho, Kim and Dyke.
Second, while Plaintiffs may be entitled to seek discovery of an alleged department-wide gender bias from supervisors other than those who directly supervise the four Plaintiffs,23 the court believes that the cost of searching 117 mailboxes exceeds the likely benefit of such a search. Accordingly, any email search shall be limited to on 50 of the 117 identified mailboxes. Plaintiffs may select the .50 mailboxes to be searched.
Because the court has pared down the both the number of words to be used in any search as well as the number of mailboxes to be searched, this will reduce the expenses forecast by Defendants, particularly if they choose to search the .pst files which have already been copied for all 117 supervisors’ e-mail boxes. This will not, however, significantly reduce the cost if the City elects to purchase the Discovery Attender software since it searches the entire e-mail back-up file in one operation. Because it has not ordered any shifting of costs to Plaintiffs, the court believes that Defendants should retain the right to decide whether to purchase and install the Discovery Attender software or to conduct searches of the .pst files using Microsoft Outlook.
IT IS THEREFORE ORDERED that Defendants’ motion for Plaintiffs to pay all or part of the costs relating to discovery of Defendants’ electronic mail (Doc. 58), is hereby DENIED.
IT IS FURTHER ORDERED that the search of Defendants’ electronic mail shall be limited in scope as set out in this Memorandum and Order.