November 26, 2013
RICK HARLOW, et al., Plaintiffs,
SPRINT NEXTEL CORPORATION, et al., Defendants.
MEMORANDUM AND ORDER
David J. Waxse U.S. Magistrate Judge
Plaintiffs, current and former employees of Defendants’ Business Direct Channel, bring this action under the Kansas Wage Payment Act and Kansas contract law, alleging Defendants failed to pay them earned commissions due to systematic problems with Defendants’ computer systems. The matter is currently before the Court on Plaintiffs’ Motion to Compel (ECF No. 310). Plaintiffs request an order compelling Defendant Sprint to produce data relevant to the calculation of commissions contained in two specific data warehouses, as well as data dictionaries and entity-relationship diagrams for these data warehouses responsive to Plaintiff’s sixth set of request for production of documents Nos. 48–49 and 52–53. As set forth below, the motion is granted in part and denied in part.
I. Relevant Facts
During discovery in this case, Plaintiffs sought production of Sprint’s billing data used for the calculation of sales commissions of its Business Direct Channel employees. For the relevant class period, Sprint maintained Ensemble, which was one of two billing systems used to calculate commissions for Sprint’s Business Direct Channel salespersons. The other billing system, P2K, was decommissioned in 2008. Defendants produced the Ensemble data in December 2011.
In addition to the Ensemble database, Sprint also has two data warehouses called the Nextel Data Warehouse (“NDW”) and Nextel Online Repository (“NOR”). These data warehouses are partial replications of data in the Ensemble database and store billing information, as well as confidential customer information unrelated to the calculation of commissions. Sprint explained their function as follows:
[NDW and NOR] were databases that were maintained and operated by Sprint. Each database contained, among other things, customer-, purchase-, or service-related data. The databases also contained partial replications of the Ensemble database. . . . On a daily basis, referential data from approximately seven tables in the NDW database was copied into [the commissions system] for use in the Business Direct commissions process.
Throughout this case, the parties have conducted data discovery in three stages. During Stage 1, Sprint produced documents and data for the named plaintiffs. The Stage 1 documents and data production served as the basis for Stage 2 discovery, during which required the parties negotiated what documents and data were needed for the entire class. During Stage 2 of the data discovery period, the parties negotiated, reviewed and approved all of the documents and data to be produced, both in form and substance, which would be analyzed by the parties’ respective experts. The agreed-upon data set was referred to as the “Stage 3 data.”
Based upon Sprint’s identification of NDW and NOR as potentially relevant datasets during Stage 1, Plaintiffs sought production of NDW and NOR data during Stage 2. Sprint objected to producing NDW and NOR data, arguing that the data was duplicative of other data it had already produced. In response to Sprint’s objections, Plaintiffs thereafter agreed that a separate production of NDW and NOR data was unnecessary “[b]ased on [Sprint’s] representation that [NDW and NOR] only contain archived data that originated in the order entry system and basically are duplicative of what [Plaintiffs] will be getting from the order entry systems.” Plaintiffs then attempted to get assurances from Sprint that its representation that the data from the order entry systems had not been subject to changes that would make the archives stored in NDW and NOR a more accurate and cleaner version of the transaction at the time it occurred. When Sprint did not respond to this letter or clarify that NDW or NOR data were not duplicates of the order-entry data, Plaintiffs served Sprint with requests for admissions. These requests sought sworn confirmation of Sprint’s previous representations regarding the duplicate nature of data in NDW and NOR and the data already produced by Sprint from the Stage 3 order entry systems. Specifically, Plaintiffs’ Request for Admission No. 8 asked Defendants to admit that “there are no differences between the data stored in [NDW] and the data being produced [by Sprint] from the Stage 3 order entry systems.” Request for Admission No. 9 sought the same admission but with respect to the NOR data warehouse. Defendants objected to both requests for admission as “vague and overbroad with respect the meaning of ‘data stored in [NDW or NOR]’ and ‘the data produced by [Sprint] from the Stage 3 order entry systems.’” They also objected to the requests insofar as they implied that the NDW and NOR and data produced by Defendants are mirror images of one another. Defendants then admitted that the NDW and NOR are “not the original source of data for the Stage 3 order entry systems data, but rather the order entry systems Stage 3 data productions are the most original source of the requested data, which is what Plaintiffs requested and agreed to.”
In December 2012, a year after it produced the Ensemble billing data, Sprint made a partial Stage 3 production. This production included data from the NDW for customer-purchased accessories that are credited to the appropriate salesperson. Sprint called this production the “NDW Accessory Pull, ” which it claims contains all of the potentially relevant data from the NDW that was possibly used for calculation of commissions for the Business Direct Channel. Plaintiffs claim that Sprint’s NDW Accessory Pull was not part of earlier Stage 1 and Stage 2 inquiries and therefore was produced without their review or approval. Sprint later produced the related Fulfillment Desktop Tool (“FDT”) order tables for the Ensemble database on June 26, 2013.
On May 3, 2013, Plaintiffs served their sixth set of requests for production of documents. Request Nos. 48 and 49 sought all data dictionaries and entity-relationship diagrams (“ERDs”) for the NDW. Request Nos. 52 and 53 sought data dictionaries and ERDs for the NOR. Defendants objected to the requests.
After making further efforts to confer about the NDW and NOR data, Plaintiffs filed the instant Motion to Compel Defendants to produce the NDW and NOR data and related information sought by Plaintiffs’ sixth requests for production.
II. Duplicative and Relevance Objections to Producing the NDW and NOR Data
Sprint objects to producing the NDW and NOR data. It argues that it has already produced all the relevant data from the NDW that played a limited role in commissions calculations for the Business Direct Channel by its production of the NDW Accessory Pull data and the original source Ensemble billing data, which includes both the billing data and the FDT order tables. It contends that neither the NDW nor the NOR contains billing information that is different from the billing data included in the Ensemble database. Also, there is no additional data in the NDW that was used or relied upon to create the data in the NDW Accessory Pull. Because it has already produced all the relevant data, Sprint argues that it makes no sense to compel production of partial replicated data unrelated to Business Direct Channel commissions, especially at this late stage in the litigation. Sprint also agues that, other than the NDW Accessory Pull, the NDW does not contain data relevant to the calculation of commissions, and Plaintiffs have failed to demonstrate how any unproduced NDW and NOR data is relevant to their claims or to the calculation of commissions for Business Direct Channel employees.
Plaintiffs assert that Sprint’s own documents and testimony confirm that the NDW and NOR data warehouses were used or referenced in some fashion in the commissions system relating to more than just accessory transactions. They point to the deposition testimony of Timothy Rassette, a former IT business consultant, who testified in the companion case, Sibley v. Sprint Nextel Corp., that “additional information about commission transactions can be obtained in NDW. . . . [T]here’s information in NDW that commissions may need to use to make the transactions complete for commission payments. . . . An example would be a phone model number would be found in NDW.”Plaintiffs contend that this deposition testimony suggests that the NDW and NOR data warehouses played a role in commissions and are therefore relevant to their claims in this case.
Plaintiffs also claim that their document review unearthed evidence suggesting that NDW and NOR actually played a more extensive role in commissions processing. This evidence also suggests that the NDW data is potentially different than order-entry data in that it is stored at a different point in the commissions processing and therefore may have been more/less processed data than what Sprint extracted and produced from the order-entry or Ensemble systems. Plaintiffs allege that Sprint’s data was routinely altered and discarded, sometimes erroneously, at various points in the commissions process. Therefore, the information from the NDW and NOR data warehouses could be compared to Sprint’s billing and FDT order-entry data to identify differences in support of Plaintiffs’ allegations.
Plaintiffs also argue that the NDW and NOR data relevancy extends beyond their potential role in commissions processing because these systems were also referenced as part of Sprint’s internal appeals process. According to Plaintiffs, when business representatives discovered discrepancies in their pay and submitted an appeal to Sprint, an analyst would investigate the issue, and often would use data in the NDW and NOR warehouses to determine whether a claim was valid and a commission was due. Given that the heart of their case is the accuracy of Sprint’s commissions payments, Plaintiffs argue that databases used by Sprint to investigate alleged commissions discrepancies are relevant.
In this case, Plaintiffs have sufficiently convinced the Court that the NDW and NOR data warehouses contain data that is only partially duplicative of the Ensemble and NDW Accessory Pull data already produced by Sprint. The initial production by Sprint of only the Ensemble billing data followed by its later unanticipated production of the NDW Accessory Pull data a year later suggests that Sprint later realized that the Ensemble data did not include all the data relevant to calculations of Business Direct Channel commissions, particularly those relating to accessory sales. This is further supported by testimony from the Sibley case where a former IT business consultant for Sprint testified that the NDW provided additional information such as phone model numbers to make transactions complete for commission payments.
The Court has reviewed the deposition testimony cited by Sprint in support of its assertion that no data from the NDW (other than NDW Accessory Pull data) was used or played any role in commission calculations for the Business Direct Channel. The Court notes that Craig Davis, an Sprint employee who worked on the NDW data warehouse, testified that the NDW data warehouse “should reflect data housed in the commission system.” When asked the question “Is data fed from NDW or was data fed from NDW to the commission system for use in processing, ” Mr. Davis answered, “It should not have been, no.” Sprint claims that this testimony supports its claim that NDW data was not used in commissions processing. Plaintiffs argue that Mr. Davis’s use of “should” in his response does not equate to a “no” answer. The Court does not read Mr. Davis’s response as an unequivocal negative response, but rather that NDW data should not have been fed to the commissions system for use in processing. If accepting Mr. Davis’s testimony is that no data from NDW was fed into the commission system, then this is contradicted by Sprint supplementing its Ensemble data production with the NDW Accessory Pull data. This also does not foreclose the possibility that commissions were calculated erroneously because NDW data was not fed into the commissions system.
The Court has also reviewed the deposition transcripts for Sprint’s referenced deponents Venu Muluka and James McCarthy. Mr. Muluka testified that he did not recollect whether the Business Channel commission process used any data from the NDW data warehouse. Similarly, Mr. McCarthy’s Rule 30(b)(6) testimony was that he had no knowledge of the NDW playing any role in commission calculations for the Business Direct Channel. Because the deponents did not recollect or did not have any knowledge of the NDW’s role in commissions, these deponents’ testimony does not add much in terms of direct support to Sprint’s claim that no data from the NDW (other than NDW Accessory Pull data) was used or played any role in the calculation of Business Channel commissions.
Sprint is in the best position to establish that its own NDW and NOR data warehouses do not contain data used or referenced in the calculation of commissions for Business Direct Channel salespersons, other than the data it has already produced. Sprint has not convinced the Court that all data relevant to Business Direct Channel commissions calculations has been produced given the need for the NDW Accessory Pull a year after the initial Ensemble data production, as well as conflicting testimony about the exact role of the NDW data in Sprint’s process for calculating commissions. This, combined with testimony that Sprint employees used NDW data to investigate discrepancies Business Direct Channel commissions and Plaintiffs’ allegation that Sprint’s data was routinely altered and discarded in the commissions process, bolsters the reasonableness of Plaintiffs’ request for the production of the NDW and NOR data.
Sprint’s responses to Plaintiffs’ requests for admission on the issue also suggest that Sprint cannot provide Plaintiffs with assurances that the NDW and NOR data warehouses may contain relevant data not previously produced as part of the Ensemble or NDW Accessory Pull data productions. When asked to admit the statement “there are no differences between the data stored in [NDW or NOR] and the data being produced [by Sprint] from the Stage 3 order entry systems, ” Sprint objected and then admitted the statement that NDW and NOR are “not the original source of data for the Stage 3 order entry systems data, but rather the order entry systems Stage 3 data productions are the most original source of the requested data, which is what Plaintiffs requested and agreed to.” Although the absolute and unqualified wording of the requests—asking Sprint to admit there are “no differences” between the data stored in the NDW and NOR and the data it produced—makes it virtually impossible to admit without some type of qualification, it appears that Sprint is admitting something different from what the requests are asking it to admit. Plaintiffs have thus sufficiently convinced the Court that the NDW contains data relevant to the calculation of Business Direct Channel commissions, and the Ensemble and NDW Accessory Pull data already produced by Sprint is not entirely duplicative of data stored in the NDW and NOR. Sprint should therefore be compelled to produce the NDW and NOR data.
III. Burdensome Objection
Sprint also argues that producing the NDW and NOR data is unduly burdensome because the requested data sets contain approximately 108 terabytes of data. Due to the volume of data in the repositories, Sprint claims that it would take it several months to extract the data.
Plaintiffs state that they are willing to work with Sprint to narrow the size of the data production. As part of their motion, they request an order requiring Sprint to work with them in good faith to reduce the data set size of the NDW and NOR data production.
Federal Rule of Civil Procedure 26(b)(2)(B) sets specific limits on discovery of electronically stored information (“ESI”). It provides that a party need not provide discovery of ESI from sources that the party identifies as “not reasonably accessible because of undue burden or cost.” Upon the filing of a motion to compel discovery, “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). Pertinent to this case, the Court is required to limit the extent of discovery otherwise allowed if “the discovery sought is unreasonably cumulative or duplicative” under Rule 26(b)(2)(C)(i).
In this case, the Court has determined that the NDW and NOR data sought is not unreasonably cumulative or duplicative of discovery already produced by Sprint. Thus, the Court is not required under Rule 26(b)(2)(C)(i) to limit the extent of discovery sought by Plaintiffs. Thus under Rule 26(b)(2)(B), Sprint must make a showing that the ESI sought—the NDW and NOR data—“is not reasonably accessible because of undue burden or cost.”
Sprint offers the affidavit of its Manager of Database Strategy, Analysis and Design in Database Services, John Mueller, in support of its claim that producing the NDW and NOR data would be unduly burdensome. Mr. Mueller states that the NDW contains around 800 data tables, contains 100 terabytes of data, and is a 90-95% replication of Ensemble database. The NOR contains about 200 data tables, contains 8 terabytes of data, and is about a 40% replication of Ensemble database. Mr. Mueller also states that because the relevant wireless customer billing databases maintain data at the customer level for business purposes, it is a complex task to extract and aggregate data in the manner requested. A data query would need to be developed to pull the requested data and the query tested to verify accuracy of the data extracted and optimization for efficiency of access. The data pull would need to be done during non-peak hours so as not to disrupt normal business activity. Mr. Mueller estimates that given the length of time the data request covers, it will take several months to develop a process to extract and pull the customer data.
While Sprint has shown that the NDW and NOR data warehouses contain many terabytes of data and its production would require the development of a special data query to pull the requested data, it has not shown that the requested production would be unduly burdensome. Given the parties’ history of working together on their staged-discovery process with minimal involvement by the Court to date and given Plaintiffs’ offer to work with Sprint to narrow the size of the production, the Court determines that the parties can find an agreed solution that provides Plaintiffs with the NDW and NOR data relevant to the Business Direct Channel commissions calculations while attempting to minimize the burden upon Sprint. The Court will therefore order that all parties work together in good faith and in a cooperative manner toward the goal of narrowing the size of the production of the NDW and NOR data to reduce the amount of data to be produced, as well as the cost and time needed to produce it.
IV. Sixth Requests for Production Nos. 48–49 and 52–53
Plaintiffs also request an order compelling Sprint to produce documents responsive to its sixth requests for production Nos. 48–49 and 52–53,  which seek all data dictionaries and entity-relationship diagrams for the NDW and NOR data warehouses. Plaintiffs ask the Court to order Sprint to produce this information so that they can work together to narrow the size of the NDW and NOR data production. They claim that they would be better equipped to make suggestions about how to narrow their request if Sprint would produce entity-relationship diagrams and data dictionaries related to these warehouses. This information, along with dictionaries defining data, will act as a map and allow Plaintiffs to understand the data stored in the warehouses and the data’s relationship to one another.
Plaintiffs’ Sixth Request Nos. 48 and 49 seek all data dictionaries and ERDs for the NDW, while Request Nos. 52 and 53 ask for data dictionaries and ERDs for the NOR. In the definitions section accompanying these requests, Plaintiffs define the term “data dictionary” as referring to “a data structure or centralized repository of information about data, such as its meaning, relationships to other data, origin, usage, and format.” Plaintiffs define “ERD” or “Entity-Relationship Diagram” as “a data modeling technique that graphically or textually illustrates an information system’s entities, the relationships between those entities, and/or the attributes within the system.”
Defendants initially objected, inter alia, to the requests as exceeding the production agreed upon in Stage 2 discovery negotiations with respect to Stage 3 discovery, both of which have closed. They further objected that Plaintiffs had not demonstrated why such information is relevant or necessary to the completion of the Stage 3 expert analysis and/or to evaluate commission earnings for class members. Defendants also objected that responding to the requests would require them to create documents and to manipulate data for Plaintiffs’ benefit, which goes beyond Defendants’ obligations under the Federal Rules of Civil Procedure. Defendants state that they have produced, and will only continue to produce, data and corresponding documents to Plaintiffs in the format in which they were maintained in the normal course of business as previously agreed between the parties as part of Stage 3 discovery.
With respect to the NDW, Defendants more specifically objected to the requests as seeking documents and/or data that do not exist because there is not a “data dictionary” or “ERD” for the NDW or the NDW Accessory Pull. For the NOR, Defendants state there is no NOR data relevant to the calculation of class member commissions in the Business Direct Channel, and thus they are under no obligation to produce “data dictionaries for NOR, ” or “ERDs for NOR” even if they were to exist.
Despite their objections to Request Nos. 48 and 49, Defendants have responded that there is not a data dictionary or ERD for the NDW or the NDW Accessory Pull. Based upon this representation that the NDW does not have a data dictionary or ERD to be produced, the Court denies Plaintiffs’ motion to compel Request Nos. 48 and 49.
In contrast to their statement that the NDW does not have a data dictionary or ERD, Defendants state in response to Request Nos. 52 and 53 that they do not have an obligation to produce data dictionaries or ERDs for the NOR data warehouse, even if they exist. This statement suggests that NOR data dictionaries and ERDs may exist. If they do exist, then Sprint should produce them to Plaintiffs. Defendants’ objections to Request Nos. 52 and 53 are overruled for the same reasons discussed above for compelling Sprint to produce the NOR data.
V. Summary of Rulings
In summary, the Court finds that Plaintiffs have sufficiently convinced the Court that the NDW and NOR data warehouses contain data that is only partially duplicative of the Ensemble and NDW Accessory Pull data already produced by Sprint. Plaintiffs have also shown that NDW and NOR data played a role in the calculation of commissions, refuting Sprint’s claim that the data warehouses were only used in the commissions system for accessory transactions. Finally, Plaintiffs have established the data warehouses contain relevant data due to Sprint’s own use of them as a resource in its internal appeals process for investigating reported discrepancies in commissions paid. The motion is therefore granted in part and denied in part. Plaintiffs’ request for an order compelling Sprint to produce NDW and NOR data relevant to the calculation of commissions for the Business Direct Channel employees is granted. Plaintiffs’ request for an order directing Sprint to work cooperatively with Plaintiffs in good faith to reduce the data set size and compelling Sprint to produce information responsive to Plaintiffs' Sixth Request Nos. 52–53 is also granted. Plaintiffs’ motion is denied with respect to Request Nos. 48–49.
Although Plaintiffs do not request their expenses incurred in making their Motion to Compel, under Fed.R.Civ.P. 37(a)(5)(C) if a motion to compel is granted in part and denied in part, the court “may, after giving an opportunity to be heard, apportion the reasonable expenses for the motion.” Here, Plaintiffs persuaded the Court that the data contained in the NDW and NOR data warehouses is not entirely duplicative of the Ensemble and NDW Accessory Pull data already produced by Sprint and is relevant to the calculation of commissions, and thus justifies an order compelling Sprint to produce the NDW and NOR data. Sprint, however, was substantially justified in asserting its objections to the requested production. Considering these circumstances, the Court will apportion the expenses so that each party shall bear their own expenses related to the motion.
IT IS THEREFORE ORDERED that Plaintiffs’ Motion to Compel (ECF No. 310) is granted in part and denied in part, as set forth herein. Sprint is ordered to (1) produce documents responsive to Plaintiffs’ Sixth Request Nos. 52–53 within 30 days of this Memorandum and Order (or the time frame agreed by the parties); (2) forthwith start the process of working cooperatively with Plaintiffs in good faith to reduce the data set size in the NDW and NOR data warehouses; and (3) produce the NDW and NOR data relevant to commissions for Sprint’s Business Direct Channel within 90 days of the date of this Memorandum and Order (or the time frame agreed by the parties).
IT IS FURTHER ORDERED that the parties shall bear their own expenses related to this motion.