*1 This matter is before the Court on the issue of which party should bear the cost of producing applicant data that was requested in a report or .csv format. The parties previously participated in a telephonic conference with the undersigned to discuss this dispute, during which it was determined that Defendant Gypsum Express, L.T.D., (“Gypsum”) would produce the discovery materials in the format requested by Plaintiff Equal Employment Opportunity Commission (“EEOC”) and the parties would thereafter file statements on who should bear the cost of producing these documents. [R. 49]. The parties then filed their respective statements. [R. 51; R. 52]. The matter now stands ripe for review.

I. FACTUAL AND PROCEDURAL BACKGROUND

This case revolves around the EEOC's allegations that Defendant Gypsum violated Title VII of the Civil Rights Act of 1964 (“Title VII”) by engaging in a nationwide pattern-or-practice of discrimination against female applicants for flatbed-driver positions. [R. 1]. The EEOC also alleges “that Defendant subjected Sherry Curtis to retaliatory discharge, and forced Heather Isaacs to engage in discriminatory hiring practices which adversely affected her terms, conditions, or privileges of employment, culminating in her constructive discharge.” [Id.].

Since the filing of the Complaint, the parties have engaged in written discovery, including a set of Requests for Production of Documents that was served on Gypsum on April 25, 2023. [R. 46-2]. Request No. 14 of these discovery requests is of particular importance and requested the following:

[Id. at pg. 6 (emphasis in original)]. As part of the instructions with these requests, the EEOC specified that:

[Id. at pg. 2 (emphasis in original)]. In response to this request, Gypsum produced 22,375 applications in PDF format without objecting to EEOC's requested form of production nor did Gypsum provide notice it would be producing these documents in a PDF format. [R. 52 at pg. 5].

*2 Plaintiff EEOC then sent correspondence to Gypsum outlining its issues with Defendant's discovery production, including the format of this applicant data. [Id.]. Plaintiff maintained that the requested applicant data should be produced in a report or .csv format instead of in a PDF format. [See R. 47 at pg. 2]. The parties conferred on the outstanding discovery issues and were able to resolve some of these issues, but they were unable to come to a complete agreement on the production of the applicant data requested by the EEOC. [R. 46; R. 47]. Specifically, through discussions with TenStreet, the third-party applicant tracking platform Gypsum used to track its applicants for driving positions, it was determined that it would take “approximately 15-20 hours and cost between $6,000 and $8,000 to convert the PDF applications to CSV files.” [R. 47 at pg. 3]. While the parties were prepared to go forward and have TenStreet produce this data in a .csv format, the parties disagreed on who should bear the cost of producing this information.

To resolve this matter, the parties contacted the undersigned to conduct a telephonic conference on this discovery dispute, which was thereafter scheduled to occur on December 1, 2023. [R. 48]. The EEOC and Gypsum both submitted their summaries of the dispute and initial arguments on who should bear the cost of production. [R. 46; R. 47]. The telephonic conference took place on December 1st and the undersigned directed Gypsum to produce the requested applicant data in the report format specified by the Plaintiff. [R. 49]. Additionally, the undersigned directed the parties to file additional briefs on whether Gypsum or the EEOC should bear the cost of this production. [Id.]. Gypsum then produced the applicant data in the requested format, which resulted in the production of 23 spreadsheets to supplement Gypsum's responses to EEOC's Request for Production of Documents Nos. 14 and 15. [R. 52 at pgs. 5–6]. The cost to produce these spreadsheets was calculated to be $3,200. [Id. at pg. 6]. Gypsum and the EEOC then submitted their respective statements supporting their arguments on who should bear the cost of this production. [R. 51; R. 52].

II. STANDARD OF REVIEW

Pursuant to FED. R. CIV. P. 34(b)(1)(C), when a party requests Electronically Stored Information (“ESI”), the requesting party “may specify the form or forms in which electronically stored information is to be produced.” “If the responding party objects to a requested form--or if no form was specified in the request--the party must state the form or forms it intends to use.” FED. R. CIV. P. 34(b)(2)(D). Further, “[i]f a request does not specify a form for producing electronically stored information, a party must produce it in a form or forms in which it is ordinarily maintained or in a reasonably usable form or forms ...” FED. R. CIV. P. 34(b)(2)(E). This production may require the producing party to translate or convert the requested information into a reasonably usable form. See FED. R. CIV. P. 34(a)(1)(A). As other courts in the Sixth Circuit have held, however, once the requesting party “request[s] native format/metadata at the outset, the burden would shift to [the producing party] to object to this form as unduly burdensome.” In re Porsche Cars N. Am., Inc. Plastic Coolant Tubes Prod. Liab. Litig., 279 F.R.D. 447, 450 (S.D. Ohio 2012).

As the Supreme Court has explained, “the presumption is that the responding party must bear the expense of complying with discovery requests ...” Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 358, 98 S.Ct. 2380, 57 L.Ed.2d 253 (1978); see also In re Onglyza (Saxagliptin) & Kombiglyze XR (Saxagliptin & Metformin) Prod. Liab. Litig., No. 5:18-MD-2809-KKC, 2020 WL 2739176, at *4 (E.D. Ky. May 26, 2020). However, this presumption can be rebutted if the producing party can demonstrate that the requested ESI is not reasonably accessible either due to “undue burden or cost.” See FED. R. CIV. P. 26(b)(2)(B) & (c)(1)(B). However, the commentary on Rule 26 makes it clear that “[r]ecognizing the authority does not imply that cost-shifting should become a common practice. Courts and parties should continue to assume that a responding party ordinarily bears the costs of responding.” FED. R. CIV. P. 26 advisory committee's note to 2015 amendment (emphasis added).

*3 To determine if the cost of production should be shifted to the requesting party, the Sixth Circuit has applied the Zubulake[1] factors, which are considered in descending order of importance as follows:

Lawrence v. Paducah Ctr. for Health & Rehab. LLC, No. 5:21-CV-00092-BJB-LLK, 2023 WL 4552285, at *11 (W.D. Ky. July 14, 2023) (citing Union Ins. Co. v. Delta Casket Co., Inc., No. 2:06-cv-02090, 2009 WL 10664840, at *6–7 (W.D. Tenn. Aug. 18, 2009)). Other courts have also analyzed this issue under the proportionality factors outlined in FED. R. CIV. P. 26(b)(1). See Lawson v. Spirit AeroSystems, Inc., No. 18-1100-EFM-ADM, 2020 WL 3288058 (D. Kan. June 18, 2020) (“Even around the time Zubulake was decided, other courts focused on Rule 26(b) proportionality factors to determine which party should bear the costs of discovery without regard to whether ESI was reasonably accessible or not.”). While there is overlap between the Zubulake and the proportionality factors, the proportionality factors are as follows: (1) the importance of the issues at stake in the action; (2) the amount in controversy; (3) the parties’ relative access to relevant information; (4) the parties’ resources; (5) the importance of the discovery in resolving the issues; and (6) whether the burden or expense of the proposed discovery outweighs its likely benefit. FED. R. CIV. P. 26(b)(1).

III. ANALYSIS

The parties first argue whether Gypsum waived its right to object to the production of the TenStreet data in a report format. Gypsum maintains that the EEOC's Request No. 14 did not include a specific request that the applicant data be produced in a report format and that EEOC's attachments and instructions also failed to specify that this data should be produced in a report format. [R. 51 at pg. 8]. As Gypsum explains, “[u]ntil the parties’ October 5, 2023 meet and confer ... Gypsum did not understand Request No. 14, which seeks ‘driver applicant data,’ to be a request to convert each and every field from the IntelliApp for each and every applicant into a series of spreadsheets.” [Id.]. Gypsum also points to the fact that EEOC's Request No. 15 did include a specific request that the requested TenStreet data be produced in a report format and therefore the nature of EEOC's Request No. 14 “was not apparent from the face of the request and only became obvious after multiple meetings with TenStreet and meet and confers with counsel for the EEOC.” [Id. at pgs. 8–9; see also R. 51-6 at pg. 7].

Plaintiff, however, argues that it informed Gypsum of the need to produce this data in report format multiple times before sending its discovery requests and that Gypsum never objected to this specific form of production. The EEOC explains that “[a]s early as January 17, 2023 ... the EEOC notified Gypsum in writing of its need for data from ‘structured database systems (such as Tenstreet)’ in ‘report format,’ such as ‘Excel spreadsheet,’ and enclosed a copy of its Form of Production/Load Specifications which reiterated that data from structured database systems be produced through report format, such as Excel or delimited text format.” [R. 52 at pg. 3]. The EEOC then reiterates that during the January 26, 2023, Rule 26(f) planning meeting “EEOC counsel also explained that this data needed to be produced in electronic format, as was done during the investigation that preceded the litigation. Gypsum's counsel raised no questions or concerns regarding the EEOC's requested form of production and affirmed it would continue to work with TenStreet to meet its discovery obligations.” [Id. at pg. 4]. The EEOC then received an email from Gypsum on February 28, 2023, with questions about ESI production but Gypsum “raised no issues with the requested form of production of data from structured database systems.” [Id.].

*4 As previously stated, under FED. R. CIV. P. 34, a requesting party may specify the form in which ESI is to be produced, but if the discovery request does not specify this information the producing party is to tender the information in the form it is usually maintained in or a reasonably usable form. Further, the producing party's “response may state an objection to a requested form for producing electronically stored information. If the responding party objects to a requested form--or if no form was specified in the request--the party must state the form or forms it intends to use.” FED. R. CIV. P. 34(b)(2)(D). Finally, “[i]f the responding party ordinarily maintains the information it is producing in a way that makes it searchable by electronic means, the information should not be produced in a form that removes or significantly degrades this feature.” FED. R. CIV. P. 34 advisory committee's note to 2006 amendment.

Here, EEOC's representations that it repeatedly advised Gypsum of the need to produce this applicant data in report format multiple times before propounding its discovery requests, combined with EEOC's instructions for its requests, indicates that Defendant was on notice that Plaintiff wished to have this data produced in a report format. While this request could have perhaps been made clearer in the wording of Request No. 14, especially when compared to the wording of Request No. 15, the context of these requests with respect to the parties’ discussions on discovery production indicate that the EEOC did communicate its desire for this information to be produced in a report format. Because the EEOC specified that this information be produced in a report format, Gypsum's arguments that the data need only be produced in the format it was kept in the usual court of business must be disregarded as this only applies “[i]f a request does not specify a form for producing electronically stored information ...” FED. R. CIV. P. 34(b)(2)(E)(ii). Further, courts in the Sixth Circuit have also required producing parties to resubmit discovery materials in a usable form when they were originally produced in PDF format. See Siser N. Am., Inc. v. Herika G. Inc., 325 F.R.D. 200, 205 n. 1 (E.D. Mich. 2018).

Gypsum's failure to object to producing this information in a report format, despite the EEOC's multiple communications and instructions outlining this request, arguably means Defendant waived its objection to producing this data in a .csv format. However, even if Gypsum has not waived this objection, Defendant has otherwise failed to satisfy its burden that the cost of production should be shifted to the EEOC.

Gypsum next argues that EEOC should bear the cost of production because Request No. 14 is overly broad and unduly burdensome. Defendant first avers that this request is unduly burdensome because Plaintiff's request required TenStreet to convert this data into a new format from the one it is typically used during Gypsum's business. [R. 51 at pg. 10]. Gypsum also argues that it should not be forced to bear the cost of this production because TenStreet, the custodian of this data, is not a party to this litigation and EEOC chose to obtain this data through Defendant instead of issuing a subpoena under Rule 45. [Id. at pg. 13]. Defendant also asserts that this request is unduly burdensome because it asks for more information than was requested during the EEOC's initial investigation into this case. [Id. at pg. 5]. Gypsum particularly references the differences between the application data fields that were requested during the EEOC investigation and now in discovery. [Id. at pgs. 5–7]. Finally, Gypsum seeks to shift the cost of production to the EEOC because this request required TenStreet to “ ‘creat[e]’ something ‘totally new.’ ” [Id. at pg. 10].

*5 As the EEOC points out, however, Gypsum fails to rebut the presumption that it should bear the cost of production. To begin with, Gypsum primarily supports its argument that the EEOC should bear the cost of converting the requested data to a report format by citing to EEOC v. SVT, LLC, a case in which the Northern District of Indiana ordered that “if the EEOC wishes to obtain the data directly from Kronos as a customized program and report tailored specifically to the EEOC's discovery request at the estimated cost of $23,500, the EEOC will have to pay the cost for the discovery.” No. 2:13-CV-245-RLM-PRC, 2014 WL 2177796, at *7 (N.D. Ind. May 22, 2014). In that case, the EEOC sought a compilation of applicant data from Kronos, a third-party vendor who housed the data for the defendant, and Kronos determined it would cost $23,500 to prepare “a customized program and report tailored specifically to meet the EEOC's request,” and that “to provide the tailored report, it will be required to piece together information from twenty-four separate data tables, decrypt personal information from three tables containing that information, and build three separate temporary databases.” Id. at *3. The Northern District of Indiana applied the Zubulake factors and ultimately found that the factors weighed in favor of shifting the cost to the EEOC. Id. at *6–7. Notably, the court held that “the data compiled in new ways by Kronos will not be very important to the issues at stake in the litigation nor will the requested discovery be very important in resolving the issues at stake in the litigation because SVT's decisions would not have been made on compilations of data in formats that it did not have access to and because it appears that SVT can produce the same data, just organized differently.” Id. at *7.

Here, however, as EEOC points out, “[u]nlike in EEOC v. SVT, where the vendor was asked to create a report containing data from a multitude of applicant and employee sources which the court found marginally relevant, here the EEOC's narrowly tailored request sought one category of information, applicant data, which is critical to proving the central questions of liability in this case.” [R. 52 at pg. 18]. No evidence has been presented that TenStreet was forced to compile this data from as many sources as Kronos was required to do in EEOC v. SCT, LLC. This fact is best demonstrated by the fact the cost of production here is only $3,200 as opposed to the $23,500 proposal in the SVT case. Additionally, SVT only dealt with 23 class members and the production of 1,600 applications, while this case “is a nationwide pattern-or-practice case involving thousands of aggrieved applicants, statistical analyses and over 22,000 applications.” [Id. at pg. 17]. The relevance of this data and the need to properly navigate and search it is therefore heightened in this matter when compared to the situation in SVT. Finally, the EEOC only sought one category of information—applicant data—instead of the multiple categories that were sought in SVT. While it is true that Gypsum would not have analyzed this information in this specific format when making its hiring decisions, the fact remains that the information contained in these reports likely impacted these decisions and the .csv format allows the EEOC to search and use this information in a similar manner than Gypsum would have been able to as well. Based on these differences, the situation in this case is distinguishable from SVT and the Northern District of Indiana's findings are therefore not persuasive in this matter.

Next, as Defendant points out, FED. R. CIV. P. 45(d)(1) mandates that “[a] party or attorney responsible for issuing and serving a subpoena must take reasonable steps to avoid imposing undue burden or expense on a person subject to the subpoena.” As other courts in the Sixth Circuit have held, “Rule 45 reflects a concern to protect recipients of subpoenas from undue burden or expense, particularly recipients who are not parties to the underlying litigation.” In re Mod. Plastics Corp., 577 B.R. 690, 702 (W.D. Mich. 2017), aff'd sub nom. In re: Mod. Plastics Corp., 890 F.3d 244 (6th Cir. 2018). In other words, non-parties should not be required to subsidize the cost of litigation if they are subpoenaed in a case. Id.

Here, Defendant asserts that the EEOC would not have been able to force TenStreet to comply with this specified form of production if it had directly subpoenaed them and that “the EEOC has not proceeded with a subpoena because Gypsum has historically cooperated with the EEOC's numerous requests for TenStreet data. Gypsum's willingness to coordinate with non-party TenStreet should not, however, result in the imposition of these costs.” [R. 51 at pg. 15]. However, as the Sixth Circuit has held in the past, “federal courts have consistently held that documents are deemed to be within the ‘possession, custody or control’ for purposes of Rule 34 if the party has actual possession, custody or control, or has the legal right to obtain the documents on demand.” In re Bankers Tr. Co., 61 F.3d 465, 469 (6th Cir. 1995) (citations omitted); see also Polymeric Res. Corp. v. Pounds of Plastic, LLC, No. 3:20-CV-00013-GFVT-EBA, 2021 WL 7081114, at *5 (E.D. Ky. May 14, 2021). Additionally, as the EEOC points out, “[t]he cost of making, keeping, and preserving the requested data has already been imposed on Gypsum by Congress. See 42 U.S.C. § 2000e-8(c).” [R. 52 at pg. 16]. Because Gypsum has the right to obtain this data on demand and given the requirements imposed by Congress on this issue, Gypsum was in possession of this information and cannot use the argument that the EEOC could have subpoenaed this data from TenStreet directly to avoid paying this cost of production. Therefore, the EEOC's decision to obtain this data from Gypsum is not sufficient to demonstrate that this request is unduly burdensome.

*6 Next, Gypsum claims that this request is overly broad because “the scope of the EEOC's request here is substantially larger than the scope of the work involved in 2020 during the underlying investigation.” [R. 51 at pg. 5]. However, the EEOC clarifies that this is not the case as Gypsum was only required to produce 23 .csv files for the current litigation while Gypsum produced 30 .csv files for the 2020 investigation. [R. 52 at pg. 12]. As the EEOC explains, “[t]he smaller discovery production is due to the EEOC's decision in litigation to focus primarily on application and lead data, and not request applicant/recruiter emails, text messages, and other types of applicant/recruiter communications that it sought during investigation.” [Id.]. These differences can also be seen when comparing the requests for information that were sent to Gypsum during the 2020 investigation versus the discovery requests at issue here. [See R. 52-2; R. 51-6]. Additionally, certain differences can be explained by the fact that new information fields may be helpful in locating witnesses for trial. [See R. 52 at pg. 13]. Finally, the EEOC did limit its discovery request to not include application disclosure and release information, “sparing the production of a tremendous amount of data given that disclosures and releases comprise over half of most applications.” [Id.; see also R. 51-2]. While this production does contain a great amount of data, this is to be expected given the nation-wide focus and broad nature of the case—simply put, such large productions are to be expected in a case like this one. Therefore, this argument must also fail.

Finally, Gypsum argues that this request is unduly burdensome because “the request required TenStreet to ‘create[e]’ something ‘totally new’ because the format requested by the EEOC was ‘not something that already exists or something that is kept in the usual course of business.’ ” [R. 51 at pg. 10]. However, this assertion appears to be at odds with TenStreet's own statements to counsel. General Counsel for TenStreet Shannon Wheeler's email to counsel did indicate that this request would require the creation of something new and that “it's not something that already exists or something that is kept in the usual course of business.” [R. 51-3 at pg. 2]. However, as the EEOC points out, “TenStreet already exported application and lead data to csv. format during the [2020] investigation.” [R. 52 at pg. 14]. Further, EEOC explains that “TenStreet confirmed that it could produce the data in multiple spreadsheets and that the production would be ‘nearly identical’ to what was produced during the investigation, albeit with up-to-date status information and for a larger time period. TenStreet also stated that the ‘scripts’ which were written in 2020 to export the data could likely be used again for this production.” [Id.]. Therefore, while this exact compilation of information may have been new for TenStreet it appears that the process itself was known to the vendor and a similar process had been used to provide data to the EEOC during the 2020 investigation. Again, the Federal Rules of Civil Procedure require the producing party to translate or convert requested information into a reasonably usable form. See FED. R. CIV. P. 34(a)(1)(A). Nothing has been presented to effectively rebut the supposition that producing this information in a report format was needed so the EEOC could reasonably analyze and use this data. Thus, Gypsum has failed to demonstrate that producing this applicant information was unduly burdensome or that the request was overly broad. However, for the sake of thoroughness, the undersigned will also examine the Zubulake and Rule 26 proportionality factors to determine if the cost should otherwise be shifted to the EEOC.

Courts have used the Zubulake and Rule 26 factors to determine if the cost of production should be shifted to the requesting party. However, under either framework, the result is the same—the factors do not support shifting the cost to the EEOC. To reiterate, the courts can consider the Zubulake factors to determine if cost-shifting should occur, which are the following in descending order of importance:

Lawrence v. Paducah Ctr. for Health & Rehab. LLC, No. 521CV00092BJBLLK, 2023 WL 4552285, at *11 (W.D. Ky. July 14, 2023) (citing Union Ins. Co. v. Delta Casket Co., Inc., No. 2:06-cv-02090, 2009 WL 10664840, at *6–7 (W.D. Tenn. Aug. 18, 2009)).

The first Zubulake factor, which is considered the most important one, is whether the discovery request is narrowly tailored to discover relevant information. Gypsum maintains that the EEOC's request is not narrowly tailored because it asks for information from application fields that are irrelevant as they relate to the Plaintiff's specific claims. [R. 51 at pgs. 18–19]. However, as the EEOC explains, “[i]t is difficult to imag[ine] what could be more relevant to a case alleging the refusal to hire a class of qualified applicants because of the applicants’ sex, female, than applicant information, including driving experience, licenses, endorsements and employment history.” [R. 52 at pg. 19]. A comparison of these fields of information will help the factfinder determine if applicants were discriminated on the basis of sex or if other factors played a part in Gypsum's hiring practice. Further, the undersigned's previous order directing the production of this information to the EEOC also supports the finding that this request was tailored to discover relevant information. [See R. 49]. Similarly, the issues at stake in this case are of high importance as this litigation seeks to enforce federal civil rights laws and ensure that businesses are not or will not engage in discriminatory hiring practices. Finally, the EEOC will benefit from this information being produced in a report format as it will allow the Plaintiff to properly analyze Gypsum's hiring practices and efficiently search and analyze this data.[2] Therefore, the first, sixth, and seventh factors weigh in favor of not shifting the cost of production to the EEOC.

Next, the parties seem to agree that TenStreet and Gypsum, by virtue of its right to ask for this data, are the parties who have access to this information. [See R. 51 at pg. 16; R. 52 at pg. 20]. While the EEOC could have directly subpoenaed TenStreet for this data, the parties also agree that the decision was made to request this information from Gypsum due to Defendant's long-standing cooperation with TenStreet. [R. 51 at pg. 15; R. 52 at pg. 16]. No evidence has been presented that the EEOC could have obtained this data from other sources, therefore this factor also weighs in EEOC's favor.

*8 Finally, the total cost of producing this data is $3,200. As the EEOC points out, this is a case involving requests for injunctive and monetary relief that may result in hundreds of thousands of dollars being recovered on behalf of the affected individuals. [See R. 52 at pg. 21]. Additionally, the EEOC is a large federal agency and Gypsum is a business enterprise comprised of over 700 employees and operations in 10 different states. [Id.]. While $3,200 is not an inconsequential amount of money, compared to the resources and potential monetary judgments in this case the cost of production is relatively minor. Further, as the EEOC points out, it appears to have narrowed the scope of its request to try and control costs and Gypsum can potentially mitigate the cost of production with TenStreet while the EEOC cannot. [Id.]. Therefore, these factors also weigh in favor of the EEOC and the Zubulake factors indicate that the cost of production should not be shifted to the Plaintiff.

Finally, the Rule 26 proportionality factors indicate the same result. As previously mentioned, the proportionality factors are: “the importance of the issues at stake in the action, the amount in controversy, the parties’ relative access to relevant information, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery outweighs its likely benefit.” FED. R. CIV. P. 26(b)(1).

As indicated above, there is a great deal of overlap between the Zubulake and Rule 26 factors, therefore the Zubulake analysis would apply to these factors as well. Again, the issues in this case deal with allegations of a nationwide practice of sex-based discrimination so the importance of the issues are self-evident. The possible monetary judgments in this matter demonstrate that the cost of production is not exceptionally onerous and in fact is less expensive than TenStreet initially estimated. [See R. 52 at pg. 5]. The EEOC can only gain access to this information from Gypsum or TenStreet, so the parties’ relative access to this information is limited. The parties are both large-spanning organizations with great resources at their disposal to help with the discovery process. The information contained in the applicant data appears to be important to help examine Gypsum's hiring practices and help with other issues before trial, such as locating potential trial witnesses. Finally, the ability for the EEOC to search and examine this data in a report format and reasonably use it for trial indicates that the benefit of this discovery outweighs the burden and expense to produce it, especially considering TenStreet already produced similar reports during the 2020 investigation and Gypsum has already produced this data to the EEOC now. [Id. at pgs. 2–3, 14]. Therefore, the Rule 26 proportionality factors indicate that the EEOC's request is proportional, and that Gypsum should bear the cost of production.

IV. CONCLUSION

While the cost of producing this data is not inconsequential, Gypsum has failed to demonstrate that the EEOC's request was overly broad or unduly burdensome. It is also arguable that Gypsum waived its right to object to this production by not objecting to this request when it was first propounded to it by the EEOC. Nevertheless, even if Gypsum did not waive this objection, the Zubulake and Rule 26 factors demonstrate that the cost of production should not be shifted to the EEOC. Gypsum has failed to rebut the presumption that it should bear the cost of production. Accordingly,

IT IS ORDERED that Gypsum's request to shift the production costs as it relates to the production of the TenStreet applicant data is DENIED.