Brooks v. TRANS UNION LLC

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF PENNSYLVANIA

WILLIAM NORMAN BROOKS, III, CIVIL ACTION Plaintiff, v. NO. 22-48-KSM TRANS UNION LLC, Defendant.

MEMORANDUM

MARSTON, J. August 1, 2024

Plaintiff William Norman Brooks, III, individually and on behalf of all others similarly situated, claims Trans Union violated section 1681e(b) of the Fair Credit Reporting Act (“FCRA”) when it sold third party creditors consumer credit reports that erroneously showed the consumers had filed for bankruptcy. After two years of class discovery, Plaintiff seeks to certify a class of individuals about whom Trans Union allegedly sold similarly erroneous reports. Trans Union opposes the motion, arguing that Plaintiff has failed to satisfy the requirements of ascertainability, commonality, predominance, and superiority under Federal Rule of Civil Procedure 23(a) and (b)(3). For the reasons below, the Court grants Plaintiff’s motion for class certification. I. BACKGROUND Trans Union is a consumer reporting agency that gathers information on credit history and activity, among other things, to generate consumer credit reports. (See Doc. No. 13 ¶¶ 7–8, 10 (Am. Compl.).) When Trans Union sells a credit report to a prospective creditor, that credit report may indicate that a consumer has filed for bankruptcy. (Id. ¶ 21.) To determine whether to include a bankruptcy notation in a credit report, Trans Union consults two sources. (See Doc. No. 107 at 7–8 (Mem. of L. in Supp. of Mot. to Certify Class).) First, Trans Union consults creditors. (Id. at 8.) Specifically, Trans Union receives monthly communications from so-called “data furnishers”—credit card companies, banks, and retailers—about the status of their accounts with consumers. (Doc. No. 13 ¶ 15; Doc. No. 107-3 at 66:1–8 (Deposition of James Garst).)

The standard format for these communications allows creditors to note, among other things, whether and how the consumer’s account has been impacted by a bankruptcy filing. (Doc. No. 107 at 12; see also Doc. No. 107-6 at 32–35 (Trans Union User Guide) (listing various “remark codes” that can be used by data furnishers to indicate different bankruptcy postures).) These monthly communications are referred to in the industry as “tradelines.” (Doc. No. 107 at 8.) Second, Trans Union consults the public record. (Id. at 7–8.) To keep apace of the public record, Trans Union hires an outside vendor, which collects information about bankruptcies pending in the federal courts and submits that information to Trans Union every day. (Doc. No. 107-3 at 59:19–61:10.) Plaintiff contends that a credit report must be supported by both sources

of information about bankruptcies—the notations on tradelines and a public record of a bankruptcy filing—before Trans Union may lawfully sell a credit report stating that a consumer has filed for bankruptcy. (Doc. No. 107 at 8–10.) This is not the first time that a consumer has brought this basic contention before a federal court. In 2003, Trans Union reached a settlement agreement in a similar class action, Clark v. Trans Union Corp. and Trans Union, LLC, C.A. No. 8:00-cv-01219-CMC (D.S.C.). The Clark litigation focused on credit reports referencing bankruptcies that Trans Union learned about from tradelines regarding jointly held accounts. (See Doc. No. 107-2 ¶ 2.) Where two consumers jointly held one account with a creditor and Trans Union received a tradeline indicating that the jointly held account had been impacted by a bankruptcy filing, Clark argued that Trans Union violated the FCRA when it sold the credit report referencing the bankruptcy without specifying which of the two consumers had filed for bankruptcy. (See id.) In December 2003, Trans Union and Clark reached a class action settlement agreement, which provided that, after May 31, 2004:

Trans Union will not disseminate to persons, other than the consumer, Credit Reports . . . that contain on Joint Accounts any reference to bankruptcy . . . unless . . . a bankruptcy is referenced in the public-record section of the Credit Report or Trans Union’s records otherwise indicate that bankruptcy should be reflected for the consumer who is the subject of the Credit Report[.]

(Id. ¶ 18.) Pursuant to this settlement agreement, Trans Union instituted the so-called “Clark Rule.” (Doc. No. 107-3 at 27:21–28:7.) The Clark Rule requires that, when Trans Union receives a tradeline stating that a joint account has been impacted by a bankruptcy, but Trans Union’s consumer file lacks a corresponding public record for that bankruptcy, Trans Union does not report the bankruptcy remark on that individual’s credit report. (See id.) From 2004 until 2020, Trans Union only applied the Clark Rule to joint accounts. (See id. at 20:21–21:3, 28:24–29:1.) However, in 2020, in response to ongoing litigation, Trans Union’s legal department proposed expanding the Clark Rule to apply to non-joint accounts. (Id. at 28:15–29:1, 30:8–10.) By February 2023, after the implementation of this expansion was complete, Trans Union had completely ceased reporting bankruptcies on consumer credit reports when a file included a tradeline with a bankruptcy notation but no corresponding public record of the bankruptcy. (See id. at 35:15–36:14.) Today Trans Union has ceased engaging in the practice that Plaintiff claims harmed him in 2020. A. Plaintiff’s Experience

Plaintiff, William Norman Brooks, III, lives in California and has never filed for bankruptcy. (Doc. No. 107-10 ¶¶ 5, 7 (Declaration of William Norman Brooks, III).) However, on January 3, 2020, non-party William Eugene Brooks filed for bankruptcy in Mobile, Alabama. (Doc. No. 107-11.) Shortly thereafter, Plaintiff received several letters from Bank of America, where he had long been a customer in good standing, informing him that Bank of America was closing his bank account and suspending access to his line of credit because he had filed for

bankruptcy.1 (Doc. No. 107-10 ¶¶ 3–5; see also Doc. Nos. 107-12, 107-13.) Having never filed for bankruptcy, Plaintiff diligently investigated the source of this error and promptly informed Bank of America of its mistake. (See Doc. No. 107-17.) In a letter to Bank of America dated January 20, 2020, Plaintiff explained that he had contacted William Eugene Brooks’s bankruptcy attorney in Mobile, Alabama, and determined that Plaintiff and William Eugene Brooks shared the same last four digits of their social security numbers. (See id.) Bank of America agreed to remove the inaccurate bankruptcy notation. (Doc. No. 107-10 ¶ 11.) But Bank of America had already sent a monthly report to Trans Union, which included a bankruptcy remark on a tradeline for at least one of Plaintiff’s Bank of America accounts.2 (See

1 Bank of America contracts with a third party, Lundquist Consulting, Inc. (“LCI”), to monitor for new bankruptcy filings by Bank of America customers. (Doc. No. 107-15 ¶ 6 (Declaration of Neil Patak).) LCI incorrectly reported to Bank of America that Plaintiff had the same nine-digit social security number as the Alabama bankruptcy-filer, when in fact Plaintiff and William Eugene Brooks only share the last four digits of their social security numbers. (Id. ¶ 23.) 2 Bank of America incorrectly reported to Trans Union that more than one of Plaintiff’s Bank of America accounts had been impacted by the Alabama bankruptcy. Complicating matters further, Bank of America did not remediate all of its errors at the same time. In January 2020, when Plaintiff disputed the bankruptcy remarks with Bank of America, the bank promptly reopened Plaintiff’s credit card account, but a putative error by a bank employee delayed removal of the bankruptcy remark from Plaintiff’s home- equity line of credit until at least March 4, 2020 when Bank of America again furnished an inaccurate tradeline to Trans Union. (Doc. No. 107-15 ¶ 25; see also Doc. No. 107-19, Doc. No. 107-14 at 139:5– 141:3 (Deposition of Danielle Nowlin).) Thus, some record-evidence appears to probe the reasonableness of Trans Union’s policies for analyzing inconsistent tradeline data from data furnishers. (See, e.g., Doc. No. 107-14 at 104:13–22, 180:11–24.) However, this complication is irrelevant to Plaintiff’s class claim. This litigation challenges only Trans Union’s policy of failing to apply the Clark Rule’s cross-check procedure to non-joint accounts, not Trans Union’s policy for revising inconsistent tradelines. Relevant here is merely the fact that Bank of America furnished at least one inaccurate tradeline to Trans Union, Doc. No. 107-14 at 111:19–24 (Deposition of Danielle Nowlin).) In turn, as of January 15, 2020, Trans Union’s consumer file for Plaintiff included the tradeline bankruptcy remark even though Trans Union did not have a corresponding record of the bankruptcy in the public record section of the file. (See id. at 111:19–24, 81:5–13.) On January 17, 2020, Plaintiff sent a letter to Trans Union explaining, with documentation, that he had never filed for bankruptcy, and

affirmatively demonstrating that a different, William Eugene Brooks from Alabama had filed for bankruptcy. (See Doc. No. 107-16; Doc. No. 107-10 ¶ 10.) Notably, Plaintiff attached the public record of William Eugene Brooks’s bankruptcy proceeding, which facially did not match Plaintiff’s social security card, driver’s license, and proof of California address. (See Doc. No. 107-16.) Despite Plaintiff’s attempts to correct his Trans Union file, Trans Union sold Plaintiff’s credit report containing an incorrect bankruptcy notation in March 2020. (Doc. No. 111 at 11.) Subsequently, Plaintiff’s application to refinance his mortgage was denied. (Id.; Doc. No. 107 at 14–16.) B. Procedural History

In January 2022, Plaintiff brought a putative class action complaint against Trans Union, which he amended the following month. (Doc. Nos. 1, 13.) The amended complaint purported to bring claims under the FCRA and its California analogue. (Doc. No. 13 ¶¶ 49, 56–65.) The case was originally assigned to the late Honorable Gene E.K. Pratter, who issued an initial scheduling order directing the motion for class certification be filed by September 2, 2022. (Doc. No. 25.) The parties jointly agreed to extend this deadline several times so that they could complete class discovery. (See Doc. Nos. 30, 35, 38, 41.) Plaintiff’s motion to certify the class

and Trans Union sold a credit report containing the inaccurate tradeline without a corresponding public record of a bankruptcy. was filed on February 2, 2024. (See Doc. Nos. 43, 107.) Defendant’s opposition was filed on March 1, 2024. (Doc. No. 48.) The Court held oral argument on July 9, 2024.3 Plaintiff now seeks to certify a single class of individuals4 seeking relief under § 1681e of the FCRA: All natural persons with an address in the United States and its Territories about whom Defendant sold a consumer report to a third party from January 6, 2020 to January 31, 2023 which included a bankruptcy remark on a tradeline, but with no reference to a bankruptcy record in the public record section of the same report, and for whom there is no government-held public record of a bankruptcy filing within ten (10) years prior to the date of the report.

(Doc. No. 107 at 10.) Trans Union opposes class certification. C. The Proposed Process for Determining Class Membership

Plaintiff contends that membership in the class can be determined through a two-step process. (See generally Doc. No. 107 at 16–20.) The first step involves reviewing Trans Union’s internal files to determine which of the files sold to a third party between January 6, 2020 and January 31, 2023 included a tradeline with a bankruptcy remark, but failed to include a record of bankruptcy in the public record section of the file. (See id. at 17–19.) The second step involves using the individual’s nine-digit social security number to determine whether the public record shows that the individual identified in the first step has in fact filed for bankruptcy. (See id. at 19–20.) Plaintiff proposes to cross-check the nine-digit social security number associated with Trans Union’s files against the Public Access to Court Electronic Records system (“PACER”) to determine whether someone with the same nine-digit social security number filed

3 On May 15, 2024, the case was reassigned to the calendar of the Honorable Karen Spencer Marston. (Doc. No. 80.) 4 Plaintiff’s amended complaint purported to bring claims on behalf of eight discrete classes of individuals. (See Doc. No. 13 ¶ 49.) However, Plaintiff now only seeks to certify one class asserting one cause of action. (See Doc. No. 107 at 10.) for bankruptcy in the ten years prior to the date Trans Union sold the file to a third party. (Id.) 1. Step One: Trans Union’s Internal Files Trans Union maintains files on about 242 million consumers based on over 85,000 data sources, and its high-volume file keeping practices are complex. (Doc. No. 111-3.) To determine the feasibility of using Trans Union’s existing systems to isolate files exhibiting the

class characteristics, Plaintiff deposed several Trans Union employees, including Trans Union’s Vice President of Data Governance, James Garst, (Doc. No. 107-3 at 9:11–12); a Senior Consultant in Trans Union’s Online Support Team, Mary Wang-Chang, (Doc. No. 107-9 at 10:7–22); a Senior Analyst in Trans Union’s Litigation Support Department, Danielle Nowlin, (Doc. No. 107-14 at 18:21–24); and an Advisor on Trans Union’s Linking and Matching Team, Corinne Wodzinski,5 (Doc. No. 107-4 at 60:1–12). Additionally, Plaintiff offers the expert report and testimony of Jonathan Jaffe, an independent technology consultant and data scientist. (See Doc. No. 107-24 ¶ 2 (Expert Report of Jonathan Jaffe).) Trans Union produced targeted data samples from its internal files for specific months

within the class period. The first data set included 22,141 files that: (i) as of June 30, 2022, had (a) at least one tradeline with a bankruptcy remark but (b) no corresponding bankruptcy in the public record section of that Trans Union file; and (ii) satisfied the same criteria on May 31, 2022. (Doc. No. 111-4 ¶ 33 (Corinne Wodzinski Report); see also Doc. No. 107-23 at 5 (Defendant’s Fourth Supplemental Responses to Interrogatories).) Trans Union was able to determine that 1,100 of these internal files had been sold to a third party because the files had a

5 Trans Union offered Wodzinski as an expert in this matter. As discussed in the Court’s contemporaneously filed Memorandum ruling on Plaintiff’s Motion to Strike Certain Documents and Motion to Preclude the Expert Report and Testimony of Corinne Wodzinski, (Doc. Nos. 56, 109), the Court strikes Wodzinski’s putative expert opinions but finds her to be a competent fact witness. hard credit inquiry in June 2022. (Doc. No. 107-23 at 2, 5–8 (providing information about hard inquiries in response to question about whether Trans Union “provided a consumer report to a third party”).) Trans Union also produced similar data for January 2020 and March 2021. In January 2020, 185,232 Trans Union files contained tradelines with a bankruptcy remark but no

corresponding public record of the bankruptcy, and 17,098 of those files were the subject of a hard inquiry. (Id. at 6.) In March 2021, 138,077 files contained tradelines with a bankruptcy remark but no corresponding public record of a bankruptcy action, and 14,816 of those files were the subject of a hard inquiry. (Id.) Trans Union had mixed success in furnishing Plaintiff with so-called “input/output logs” for the same months. An input/output log “contains a record of the true and accurate copy of each Trans Union report on a subject transmitted to a third party.” (Doc. No. 107-25 ¶ 19 (March 2023 Declaration of Jonathan Jaffe).) Specifically, each input/output log memorializes both the third-party request for information and Trans Union’s response thereto, both of which

adhere to a fixed format that does not vary from log to log. (Doc. No. 107-26 at 9:17–10:8 (Deposition of Jonathan Jaffe).)6 In other words, input/output logs are Trans Union’s uniform record of credit report exchanges with third parties. Trans Union located input/output logs documenting most of the hard inquiries into the 1,100 files from June 2022 that contained a bankruptcy remark on a tradeline, lacked a corresponding public record of bankruptcy, and were the subject of a hard inquiry that month. (Doc. No. 107-23 at 7, 9.) However, Trans Union did not locate input/output logs for similarly situated files from either January 2020 or March 2021.

6 As Jaffe explained during his deposition, input/output logs combine the specific request made by third parties to Trans Union—also known as a fixed format inquiry or “FFI”—with the data sent by Trans Union in response—also known as a fixed format response or “FFR.” (Doc. No. 107-26 at 9:17–10:16.) (See id. at 9–10.) Trans Union’s failure to locate input/output logs for January 2020 and March 2021 arises from Trans Union’s practice of segregating data into different storage systems based on the age of the data. Relatively new data is stored in a system called “Splunk,” which Trans Union uses to “ingest and index textual data.” (Doc. No. 107-26 at 18:10–18.) Data is stored in Splunk for

about two years.7 (Doc. No. 107-25 ¶ 36.) The majority of the June 2022 logs were more easily accessible because they were contained in Splunk. (See Doc. No. 107-23 at 9 (explaining that the fraction of missing June 2022 logs were inaccessible “[f]or a variety of reasons, including because certain logs were apparently not contained in Splunk”).8 This accords with Jaffe’s opinion that class members’ credit reports “for the most recent two years can be identified using” functionalities of Splunk itself. (Doc. No. 107-25 ¶ 52.) Thus, there appears to be little difficulty with accessing input/output logs for data that is contained within Splunk. Older logs are more difficult to access. Trans Union stores logs that are over two years old, such as those from January 2020 and March 2021, in long term file storage (“LTFS”).9

(Doc. No. 107-25 ¶ 25.) Trans Union’s counsel concedes it is not impossible to locate input/output logs stored in LTFS, but only that it is highly time-consuming to do so. (See Doc.

7 But see Doc. No. 107-9 at 58:22–59:7 (Deposition of Mary Wang-Chang) (stating that Trans Union only keeps “up to three to six months of data” in Splunk). Jaffe suspected that Wang-Chang was only able to access 3 to 6 months of data in Splunk “because of permissions limiting her access.” (Doc. No. 107-25 at 14 n.25 (March 2023 Declaration of Jonathan Jaffe).) In any case, because the proposed class runs from January 2020 to January 2023, much of the data that will need to be analyzed will not be in Splunk under either timeline. 8 Trans Union has not explained why some June 2022 logs were not contained within Splunk. 9 Trans Union in fact stores its older data in two distinct formats, namely LTFS and physical tape archives of data formerly contained in Splunk. (See Doc. No. 107-25 ¶¶ 25, 37; Doc. No. 111 at 16.) However, counsel for Trans Union conceded during oral argument that LTFS and tape archives are “similar,” and only LTFS is relevant here. (See July 9, 2024 Hrg. Tr. at 109:23–110:2.) Thus, the accessibility of tape archives appears to be moot, notwithstanding its discussion in the briefing. No. 107-23 at 9–10 (“[T]he process Trans Union used required more than one day to identify, extract, and process the single day’s worth of Input/Output Logs from March 31, 2021.”) But, Jaffe offers a much shorter estimate of the amount of time that would be required to process this data, opining that seven full years of older data could be analyzed to determine class membership in “a matter of hours to days to weeks, depending on how you optimize it.” (Doc. No. 107-26 at

87:12–88:4.) Specifically, Jaffe opined that “UNIX shell script commands” could be used to isolate the applicable input/output logs held in LTFS. (Doc. No. 107-25 ¶¶ 27–35.) And Trans Union’s own Senior Consultant appears to agree with this opinion. (Doc. No. 107-9 at 59:15–18 (“[Plaintiff’s Counsel]: And it’s your understanding that the raw body log data stored in that LTFS location can be accessed using a Linux command, you said? [Wang-Chang]: Yes.”).10 Because the data contained within Trans Union’s input/output logs adheres to a fixed format,11 Jaffe explained that a UNIX command could be written that would isolate and pull “the subject name, address, social security number, date of birth, transaction id, along with trade records containing bankruptcy indicators and the public records with codes indicating bankruptcy” for

potential class members. (Doc. No. 107-25 ¶ 32.) He added that Trans Union’s enterprise-scale system is capable of streamlining this process because it can simultaneously run parallel searches of multiple files via a UNIX/Linux functionality called “parallel.”12 (Id. ¶ 34.)

10 Unix and Linux commands do not appear to materially differ in this context. (See Doc. No. 107-25 at 11 n.17.) 11 Counsel for Trans Union continues to argue that LTFS is not searchable in a fixed format, but his argument on this point is unconvincing. He stated that, because files stored in LTFS have a variable number of fields containing fixed-format data, the data is not searchable. (July 9, 2024 Hrg. Tr. at 114:10–23.) However, files stored in the searchable Splunk system also have a variable number of datapoints associated with each file. (See, e.g., Doc. Nos. 108-2, 69-1 (documenting files with 0, 1, 2, and 3 additional names on credit reports).) There is no evidence before the Court explaining why a variable number of fields containing data in a fixed format would render that fixed format data unsearchable. 12 Trans Union has pointed to no record evidence that rebuts Jaffe’s opinion about the streamlined 2. Step Two: The Public Record The second step of the class identification process involves determining whether potential class members have filed for bankruptcy. Plaintiff proposes determining whether a public record of bankruptcy exists by searching for the public record via PACER.13 (Doc. No. 107 at 19–20.) This determination is significant because one element of an FCRA claim under § 1681e(b) is that

“inaccurate information was included in a consumer’s credit report.” Cortez v. Trans Union, LLC, 617 F.3d 688, 708 (3d Cir. 2010) (quoting Philbin v. Trans Union Corp., 101 F.3d 957, 963 (3d Cir. 1996)). If Trans Union sold an individual’s file to a third party during the class period, and that file contained a tradeline notating bankruptcy without a corresponding public record of a bankruptcy action, but that individual in fact did file for bankruptcy, then the report would be accurate, and no cause of action under § 1681e(b) would accrue. Thus, after the set of potential class members has been derived from Splunk and LTFS, potential class members who have filed for bankruptcy remain to be excluded. This exclusion is the focus of step two. Plaintiff proposes to use potential class members’ nine-digit social security numbers from

the files located at step one to search PACER for a bankruptcy filing. (See Doc. No. 107 at 19.) Nine-digit social security numbers for potential class members can be determined from the fixed- format data contained in the input/output logs. (See Doc. No. 107-26 at 136:22–137:7.) Through a simple, objective search, nine-digit digit social security numbers can be used to

searchability of Trans Union’s older files. 13 Trans Union grumbles at the fact that Plaintiff’s amended complaint originally defined the class at issue exclusively with reference to Trans Union’s internal files, (see Doc. No. 13 ¶ 49(a) (Am. Compl.)), but Plaintiff now seeks to certify a class with reference to government-held files. Trans Union also speculates that Plaintiff “changed his theory of liability and class definition based on Wodzinski’s opinions.” (Doc. No. 113 at 15.) The Court, however, discerns no impropriety in amending a class definition in light of facts learned through class discovery. determine whether PACER contains a corresponding public record of a bankruptcy action. (Id. at 126:10–127:6, 148:2–149:14.) If PACER contains a public record of a bankruptcy from the previous ten years under that social security number, then Plaintiff excludes the associated individual from the class.14 Jaffe tested this method for excluding bankruptcy filers using the June 2022 dataset.

First, he isolated the nine-digit social security numbers of potential class members from June 2022. (Doc. No. 107-29 ¶ 6 (Declaration of Lauren KW Brennan, Esq.); see also id. at 5–42 (displaying nine-digit social security numbers for potential class members from June 2022 dataset).) A paralegal in the office of Plaintiff’s counsel then determined that, out of the 1,100 potential class members from June 2022 that were isolated at step one, at least 270 lacked a public record of a bankruptcy action on PACER. (Id. ¶ 9.) Additionally, 311 individuals had a public record of a bankruptcy that was more than 10 years old in June 2022.15 (Id. ¶ 10). Finally, Plaintiff notes that, because Trans Union began expanding the Clark Rule to non-joint accounts in April 2021, the volume of class members for the months preceding the Clark Rule expansion

will be much higher than the volume of members from June 2022. (Doc. No. 107 at 19–20.) Thus, the class is likely to number in the tens of thousands.16 (Id.)

14 Counsel for Trans Union takes issue with the fact that step two of Plaintiff’s procedure is designed to exclude non-class members who have filed for bankruptcy rather than affirmatively include class members who have not filed for bankruptcy. (See July 9, 2024 Hrg. Tr. at 96:7–17.) However, this ponderous ontological distinction makes no legal difference. Class counsel should always endeavor to exclude potential class members with a legally dispositive characteristic from an overly inclusive class. Eliminating such individuals—a desirable end—will always be a negative procedure. 15 It is against Trans Union policy to report bankruptcies that are more than 10 years old. (Doc. No. 107- 14 at 67:19–68:14.) 16 Trans Union contends that step two will not exclude everyone who has filed for bankruptcy, but much of Trans Union’s argument to this effect appears to be based on irrelevant information provided by Trans Union’s longtime employee, Corinne Wodzinski. Wodzinski explained that Trans Union routinely creates secondary files when it receives new information that may belong to an individual, but which Trans Union is not sufficiently confident actually belongs to that individual to include it in the Using the same June 2022 dataset, Trans Union’s counsel double-checked 50 of the 270 individuals for whom Plaintiff’s PACER search confirmed no public record of bankruptcies existed. (Doc. No. 111-5 ¶ 13 (Declaration of Albert E. Hartmann, Esq.).) Instead of relying on nine-digit social security numbers to search PACER, as Plaintiff did, Trans Union searched for matching names and addresses contained within the putative class member’s credit report. (See

id. ¶¶ 14–17.)17 Of the fifty individuals comprising Trans Union’s dataset, Trans Union reports that it found seventeen public records of bankruptcies on PACER where, although the nine-digit social security number did not match, “the name, street address, city, state, and zip code of a bankruptcy debtor exactly matched the corresponding information on the Trans Union credit report.” (Id. ¶ 18.) Trans Union also reportedly found thirteen public records of bankruptcies on PACER where, again, although the nine-digit social security number did not match, “the name of a debtor, as well as the state in the address from PACER, matched the corresponding name and state from the Trans Union” credit report. (Id. ¶ 19.) II. CLASS CERTIFICATION

“The class action is an exception to the usual rule that litigation is conducted by and on

individual’s primary file. (Doc. No. 111-4 ¶ 24.) Wodzinski originally reported that when Trans Union sells a file to a third party, the company typically only sells one file. (See id. ¶ 25.) Confusingly, during the Rule 702 hearing, Wodzinski testified that Trans Union aggregates primary, secondary, and even tertiary files in real-time whenever a third party requests a credit report. (July 9, 2024 Hrg. Tr. at 38:10– 41:9.) In any case, Trans Union argues that, even if it sold a primary file to a third party containing a bankruptcy remark on a tradeline but lacking a public record of a bankruptcy, Trans Union might still have the corresponding public record of a bankruptcy for the primary file in a non-aggregated, secondary file. (See Doc. No. 111-4 ¶ 3b.) The Court understands Trans Union’s logic, but this feature of Trans Union’s internal file keeping system does not help the Court evaluate the methodology Plaintiff actually proposes, namely to search the public record via PACER, not Trans Union’s internal files. Thus, Trans Union’s argument is entirely irrelevant to the class Plaintiff actually proposes. 17 To be clear, Trans Union only searched PACER using name and address data, ignoring whether the 9- digit social security on the file it sold matched the public record of a bankruptcy to which they “matched” it. (See Doc. No. 111-5 ¶¶ 14–16.) behalf of the individual named parties only.” Wal-Mart Stores v. Dukes, 564 U.S. 338, 348 (2011) (internal quotation marks and citation omitted). To invoke this exception, every putative class action must satisfy the four requirements of Rule 23(a) and the requirements of either Rule 23(b)(2) or (b)(3). See Fed. R. Civ. P. 23(a)–(b). Rule 23(a) requires that (1) the class be so numerous that joinder is impracticable; (2) there are questions of law or fact common to the

class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a). Plaintiff bears the burden of proving the prerequisites to class certification are met and that the class fits within one of the 23(b) categories. Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 591 (3d Cir. 2012). Plaintiff relies on Rule 23(b)(3) as the basis for certification. (Doc. No. 107 at 28–33.) Rule 23(b)(3) requires that common class questions predominate over questions affecting only individual members and that the class action mechanism be superior to other available methods for fairly and efficiently adjudicating the controversy. Fed. R. Civ. P.

23(b)(3). “Although the plaintiff need not establish the merits of his case at this stage, the Third Circuit has held that ‘[a]n overlap between a class certification requirement and the merits of a claim is no reason to decline to resolve relevant disputes when necessary to determine whether a class certification requirement is met.’” In re Suboxone (Buprenorphine Hydrochloride & Nalaxone) Antitrust Litig., 421 F. Supp. 3d 12, 45–46 (E.D. Pa. 2019) (quoting In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 316 (3d Cir. 2008)). Ultimately, class certification is “proper only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23 are met.” Hydrogen Peroxide, 552 F.3d at 309 (cleaned up). Before considering Rule 23(a) and Rule 23(b)(3), the Court must determine two matters. First, that the class has clearly defined parameters and claims to be given class treatment as required by Rule 23(c)(1)(B). Marcus, 687 F.3d at 591–92. Second, that the class is currently and readily ascertainable based on objective criteria. Id. at 592–93. Here, the first issue is satisfied as the class in this matter has clearly defined parameters and shares one common claim under § 1681e of the FCRA. The Court will thus begin with the ascertainability requirements

before turning to the Rule 23(a) and Rule 23(b)(3) requirements. A. Ascertainability

Plaintiff moves to certify a Rule 23(b)(3) class under the following definition: All natural persons with an address in the United States and its Territories about whom Defendant sold a consumer report to a third party from January 6, 2020 to January 31, 2023 which included a bankruptcy remark on a tradeline, but with no reference to a bankruptcy record in the public record section of the same report, and for whom there is no government-held public record of a bankruptcy filing within ten (10) years prior to the date of the report.

(Doc. No. 107 at 10.) To meet the ascertainability requirement, the plaintiff has the burden of showing, by a preponderance of the evidence, that “(1) the class is ‘defined with reference to objective criteria’; and (2) there is ‘a reliable and administratively feasible mechanism for determining whether putative class members fall within the class definition.’” Byrd v. Aaron’s Inc., 784 F.3d 154, 163 (3d Cir. 2015) (quoting Carrera v. Bayer Corp., 727 F.3d 300, 306 (3d Cir. 2013)). Courts have discussed three reasons for this standard. “First, ascertainability and a clear class definition allow potential class members to identify themselves for purposes of opting out of a class. Second, it ensures that a defendant’s rights are protected by the class action mechanism, and that those persons who will be bound by the final judgment are clearly identifiable. Finally, it ensures that the parties can identify class members in a manner consistent with the efficiencies of a class action.” City Select Auto Sales Inc. v. BMW Bank of N. Am. Inc., 867 F.3d 434, 439 (3d Cir. 2017) (cleaned up). The movant “must propose a classification method with evidentiary support.” In re Niaspan Antitrust Litig., 67 F.4th 118, 130 (3d Cir. 2023). The Court “must engage in a rigorous analysis” to determine whether the ascertainability requirement has been demonstrated by a preponderance of the evidence. Id. However, this requirement “does not mean that a plaintiff

must be able to identify all class members at class certification—instead, a plaintiff need only show that class members can be identified.” Id. (quoting Byrd, 784 F.3d at 163). Here, the proposed class is defined with reference to four objective criteria: (1) Trans Union sold a class member’s credit report to a third party between January 6, 2020 and January 31, 2023; (2) the credit report contained a bankruptcy remark on a tradeline; (3) the credit report lacked a corresponding public record of a bankruptcy; and (4) there is no government-held public record of a bankruptcy filing by the class member within ten years of the month in which Trans Union sold the credit report. The Court finds that there are reliable and feasible mechanisms for determining whether

class members meet these four criteria. As discussed above, Plaintiff proposes that he will first determine which Trans Union files meet criteria (1), (2), and (3) by employing various mechanisms for searching Trans Union’s internal data, which differ based on how that data is stored. (See Doc. No. 107 at 17–19.) Then for step two, Plaintiff proposes searching the PACER database for public records of bankruptcies to determine whether the subjects of files meeting the first three criteria also meet criteria (4). (Id. at 19–20.) 1. Step One: Trans Union’s Internal Files

As an initial matter, the Court notes that criteria (2) and (3)—requiring that class members’ files contain a bankruptcy remark on a tradeline and lack a public record of a bankruptcy, respectively—are readily determinable from Trans Union’s internal files regardless of how the company stores them. Trans Union readily ascertained the files from January 2020, March 2021, and June 2022 that exemplified these characteristics. (Doc. No. 107-23 at 6–7.) And Trans Union was also able to readily determine whether files exemplifying these characteristics for the sample months were the subject of a hard inquiry during those months,

which would trigger the sale of the report. (See id.) The difficulty reportedly arises when Trans Union attempts to isolate input/output logs memorializing those hard inquiries. Because Trans Union segregates its files based on their age, and the class period covers both the newer files Trans Union stores in Splunk and the older files it stores in LTFS, two discrete mechanisms must be used to isolate input/output logs. Plaintiff has demonstrated by a preponderance of the evidence that the logs of newer data contained within Splunk can be searched using the functionalities of the Splunk system itself. (See Doc. No. 107-25 ¶ 52.) In particular, the Court notes that Trans Union successfully located input/output logs for June 2022 and shared this data with Plaintiff.18 (See Doc. No. 107-23 at 8–9 (“Trans Union was able to

design and execute queries that identified and extracted 1,527 of the input/output logs associated with the June 2022 Inquiries.”).) Plaintiff has also demonstrated by a preponderance of the evidence that a reliable method exists to isolate input/output logs that are stored in LTFS. At the class certification stage,

18 Trans Union flagged that it was unable to find 170 logs from June 2022, but the only reason it supplies for this failure is that “certain logs were apparently not contained within Splunk.” (Doc. No. 107-23 at 9.) Trans Union has not explained this condition, and the Court finds that this condition does not rebut evidence that the proposed method for isolating input/output logs in Splunk is reliable and feasible. See Kelly v. RealPage Inc., 47 F.4th 202, 223 (3d Cir. 2022) (“[W]here a defendant’s lack of records makes it more difficult to ascertain members of an otherwise objectively verifiable class, the individuals who make up that class should not bear the cost of the defendant’s faulty record keeping.” (cleaned up) (quoting Hargrove v. Sleepy’s LLC, 974 F.3d 467, 470 (3d Cir. 2020))). Plaintiff “need not ‘be able to identify all class members[,] . . . instead, [he] need only show that class members can be identified.’” City Select Auto Sales Inc., 867 F.3d at 439 (some internal quotation marks omitted) (quoting Byrd, 784 F.3d at 163). Both Plaintiff’s expert, Jaffe, and Trans Union’s Senior Consultant of Online Support, Wang-Chang, presented evidence that UNIX shell script commands could be used to isolate the applicable input/output logs held in

LTFS. (See Doc. No. 107-25 ¶¶ 27–35; see also Doc. No. 107-9 at 59:15–18.) Trans Union’s lawyers disagree with Jaffe and Wang-Chang, but the evidence trumps their arguments. Trans Union argues that Jaffe’s opinions are speculative, and Plaintiff fails to sufficiently explain how Trans Union can search LTFS efficiently. (See Doc. No. 111 at 17.) Trans Union contends that, because Jaffe’s opinions about Trans Union’s ability to process logs from LTFS are partly based on data that he reviewed from the Splunk system, (see Doc. No. 107- 26 at 88:5–13), his opinions do not actually address how to query the older data. (See Doc. No. 111 at 16–17). During oral argument, defense counsel pointed out that Jaffe has not provided the exact query that could be used to search LTFS for the relevant logs, arguing that this omission means Plaintiff has failed to meet his burden.19 (July 9, 2024 Hrg. Tr. 94:10–12, 97:22–98:5.)

But this argument is akin to faulting Plaintiff for failing to solve a math problem that has never been presented to him. Trans Union only provided input/output logs from Splunk for June 2022. Trans Union did not provide the January 2020 or March 2021 input/output logs that were requested.20 (See Doc. No. 107-23 at 9–10.) And Trans Union stopped its own efforts to collect

19 Plaintiff’s counsel disputed this contention during the hearing, directing the Court to paragraph 48 of Jaffe’s March 28, 2023 declaration. (July 9, 2024 Hrg. Tr. at 118:8–16.) Paragraph 48, however, contains “pseudocode” that could be modified to identify class member reports for the most recent two years in Splunk. (Doc. No. 107-25 ¶¶ 46–52.) Trans Union complains that Jaffe failed to provide a query that would work for LTFS, not Splunk. Thus, the Court concludes that Jaffe did not provide a specific query that could be used to search LTFS data for the relevant logs. 20 Defense counsel suggests that Plaintiff should have requested all three years of the relevant LTFS data, this data because it was taking “more than one day” to process the logs from March 31, 2021, and it appeared to Trans Union that it would take too long to process the entire month of March 2021 given the then-pending deadline for turning over materials. (Id.) Given Jaffe and Wang-Chang’s testimony, the Court is not persuaded by Trans Union’s self-serving argument that it will take Trans Union years to isolate this data when Trans Union

admits it only attempted to do so for a day. In assessing a motion for class certification, the Court’s rigorous analysis often “involves judging credibility, weighing evidence, or deciding issues that overlap with the merits of a plaintiff’s claims.” Harnish v. Widener Univ. Sch. of L., 833 F.3d 298, 304 (3d Cir. 2016) (citing Hydrogen Peroxide, 552 F.3d at 316–25). After indulging the parties in two years of class discovery, the Court is skeptical of Trans Union’s failure to complete a process that would have taken about a month to complete.21 The Court declines to draw an adverse inference against Jaffe’s opinions because he was never given the opportunity to analyze data that Trans Union failed to produce, even though Trans Union agrees it developed a procedure for obtaining it. (See Doc. No. 107-23 at 9–10.) And, even assuming

that Trans Union’s LTFS is as onerous to search as Trans Union contends, defendants may not “defeat ascertainability with a strategic decision to house records across multiple sources or

which would involve about three (3) billion Trans Union credit reports. (See July 9, 2024 Hrg. Tr. at 93:24–95:4, 108:6–20.) According to defense counsel, Jaffe then should have designed a query to isolate the relevant input/output logs from the entire class period from those billions of reports. (See id.) However, Plaintiff need not actually determine class members at this stage; he merely needs to “show that class members can be identified.” City Select, 867 F.3d at 439 (cleaned up). It was perfectly reasonable for Plaintiff to request just two months of the relevant logs—those from January 2020 and March 2021— and attempt to devise a solution based on that targeted sample set, just as the parties had proceeded with respect to the June 2022 data contained in Splunk. 21 Though Trans Union’s response to Plaintiff’s interrogatory suggested that the process of isolating one day’s worth of data took more than one day, during oral argument defense counsel stated that it took one day to do so. (July 9, 2024 Hrg. Tr. at 109:11–12.) There is also evidence before the Court that this process could have been optimized if Trans Union had not simply abandoned it. (See Doc. No. 107-26 at 87:21–88:4.) databases.” Kelly v. RealPage Inc., 47 F.4th 202, 223 (3d Cir. 2022). The preponderance of the evidence demonstrates that the data in LTFS adheres to fixed format and is searchable by means of a shell script command. This is a reliable and feasible method for isolating input/output logs that memorialize Trans Union’s sale of a credit report to a third party when that data is stored in LTFS.

2. Step Two: PACER

All that remains to be determined is whether there is a government-held public record of a bankruptcy filing by the class member within ten years of the month in which Trans Union sold the class member’s credit report. Plaintiff proposes to search PACER using nine-digit social security numbers, which are determinable from Trans Union’s input/output logs. (See Doc. No. 107 at 19–20; Doc. No. 107-26 at 136:22–137:7.) The Third Circuit Court of Appeals has consistently found that record-matching “is precisely the sort of exercise” that is “sufficiently administrable to satisfy ascertainability.” See Kelly, 47 F.4th at 223 (collecting cases). And Trans Union’s counsel even conceded during oral argument that searching PACER with nine- digit social security numbers is an objective procedure.22 (July 9, 2024 Hrg. Tr. at 106:23–25.) Trans Union attempts to argue that using an individual’s nine-digit social security number is not sufficiently accurate, though, and proposes a variety of other pieces of information that could be matched across records. (See Doc. No. 111 at 22–24.) However, this muddying of the waters is not an ascertainability argument; it is an argument on the merits of the class FCRA claims. For the reasons explained in the Court’s predominance analysis below, while the Court is not precluded from considering the merits of Plaintiff’s claims at the certification stage, see Harnish,

22 During the Rule 702 hearing, Wodzinski, too, exclaimed that it would be “great” to have access to nine- digit social security numbers when endeavoring to match public records of bankruptcies, as opposed to the four-digit numbers to which Trans Union has access. (July 9, 2024 Hrg. Tr. at 27:1–18.) 833 F.3d at 304, the Court finds that these proposed alternative matching standards are not fatal to the merits of the class’s § 1681e claims. Matching records on the basis of nine-digit social security numbers is a feasible and reliable process, and the lack of a corresponding record on PACER evidences that the individual to whom that nine-digit social security number belongs has not filed for bankruptcy. Wodzinski

presented evidence that Trans Union’s internal files sometimes contain data entry errors. (Doc. No. 111-4 ¶ 18.) However, she has no experience whatsoever with PACER. (Doc. No. 107-4 at 66:5–67:2.) The Court declines to find, in what would amount to a non-sequitur, that the nine- digit social security numbers included on public records of bankruptcies on PACER are unreliable because Trans Union’s internal files contain typographical errors. Cf. Van v. LLR, Inc., 61 F.4th 1053, 1068 (9th Cir. 2023) (“We do not permit a defendant to support its invocation of individualized issues with mere speculation.”). There is no evidence before the Court that nine-digit social security numbers on PACER unreliably reflect the nine-digit social security numbers of the individuals who filed for bankruptcy. Nor is there direct evidence that

any of the social security numbers relied on to perform PACER searches on the June 2022 dataset were inaccurate. 23 Importantly, there is evidence that these nine-digit social security

23 There is evidence before the Court that individuals whose nine-digit social security number returns no public record of a bankruptcy action on PACER do have characteristics in common with individuals for whom there is a public record of a bankruptcy action on PACER. For example, Trans Union has submitted evidence that, for at least 17 of the 270 putative class members from June 2022 who had never filed for bankruptcy, a public record of a bankruptcy on PACER existed where, although the nine-digit social security number did not match, “the name, street address, city, state, and zip code of a bankruptcy debtor exactly matched the corresponding information on the Trans Union credit report[.]” (Doc. No. 111-5 ¶ 18.) This level of matching might appear to be strong evidence that the PACER record found by Trans Union belongs to the corresponding class member, but it is not. The Consumer Financial Protection Bureau has explained that, especially for Hispanic, Asian, and Black communities with low name-diversity, name-only matching is likely to lead to inaccuracies on credit reports. Consumer Financial Protection Bureau, Advisory Opinion on Name-Only Matching Procedures and Fair Credit Reporting (Nov. 10, 2021), 86 Fed. Reg. 62468, at 62470. And Plaintiff’s counsel convincingly argued during the July 9 hearing that name and full address are insufficient to match records. (See July 9, 2024 Hrg. Tr. at 89:5–15.) In particular, she argued that, because parents often share names with their children numbers are a more reliable method for determining whether individuals have filed for bankruptcy than any of the shifting alternative standards proposed by Trans Union. (See Doc. Nos. 108-2, 69-1 (documenting numerous mismatches between the nine-digit social security numbers from Trans Union’s credit report and the final four digits of social security numbers from PACER public bankruptcy records for individuals who Trans Union argues filed for

bankruptcy based on comparisons of name, address, and four-digit social security numbers).) Here, the Court finds that the evidence supports the proposition that searching PACER for public records of bankruptcies based on nine-digit social security numbers is a reliable and administratively feasible method for determining whether there is a government-held public record of a bankruptcy filing by the class member within ten years of the month in which Trans Union sold a credit report. The Third Circuit Court of Appeals’ decision in Kelly supports this conclusion. In Kelly, the Third Circuit Court of Appeals overturned the district court’s denial of class certification on the basis of ascertainability24 in a case brought under § 1681g of the FCRA. Kelly, 47 F.4th at

205. RealPages, a credit reporting agency, allegedly unlawfully failed to disclose to the plaintiffs the sources of public information it used to produce the plaintiffs’ erroneous credit

and live with them, matching by name and address can be equivocal. (Id., accord Doc. No. 107-4 at 275:18–277:5 (Deposition of Corinne Wodzinski) (explaining instance where Trans Union’s algorithm segregated two files that shared name and address but contained different social security numbers and dates of birth).) A second illustration of the problem with relying on names and addresses would be that large apartment buildings, especially in certain communities, could contain multiple residents who share the same name and address. (See July 9, 2024 Hrg. Tr. at 89:6–10.) Thus, the Court finds that Trans Union’s name and address matching does not convincingly undermine the reliability of searching by nine- digit social security number. (Accord Doc. No. 107-4 at 271:24–272:2 (invoking the “unique cardinality” of the social security number).)

24 The district court also found that the predominance requirement had not been met, but the predominance analysis in Kelly depended on a statutory interpretation issue regarding a distinct section of the FCRA than is at issue here. See Kelly, 47 F.4th at 216–22. The primary relevance of Kelly to the present action is thus the ascertainability analysis. reports. Id. at 207–08. The plaintiffs sought to certify a class of individuals who received reports from RealPage that did not include, as required by the FCRA, the public record information upon which RealPage had relied in producing the credit report. Id. at 208–09. The district court concluded that this class was not ascertainable because figuring out which credit reports “contained public record information . . . would require ‘[a] review of each individual

file,’” which the district court found to be “not administratively feasible.” Id. at 210. The Third Circuit Court of Appeals explained, however, that there is no per se rule that file-by-file review is not administratively feasible. See id. at 223–25. Kelly bolsters Plaintiff’s case for ascertainability because it supports the conclusion that individually searching PACER for public records of bankruptcies using nine-digit social security numbers is administratively feasible. The Kelly Court explained that it was administratively feasible to verify “whether there is public record information in [RealPage’s] file” because it merely required “an examination of the face of Rental Reports that are indisputably in RealPage’s possession.” Id. at 224. Similarly, determining the nine-digit social security number

merely requires an examination of the face of the input/output logs that are in Trans Union’s possession. Searching that nine-digit number in PACER is administratively feasible.25 Finally, the Court finds Trans Union’s reliance on Niaspan does not militate against finding that the ascertainability requirement has been met here. In Niaspan, the Third Circuit Court of Appeals upheld the district court’s denial of class certification based on its finding that the class was not ascertainable. See Niaspan, 67 F.4th at 134–35. Niaspan was an antitrust suit brought against the manufacturer of a drug, alleging that the drug was too expensive because of

25 Jaffe opined that these PACER searches could be partially automated. (Doc. No. 107-26 at 137:19– 138:7.) the defendants’ anticompetitive behavior. Id. at 124–25. The plaintiffs moved to certify a class of insurance entities that overpaid for the drug, but because prescription drug purchases are complex transactions involving multiple end-payors, determining membership in the class required determining what role class members played in each transaction. Id. at 128–29. The contractual arrangements of different drug-purchasers involved potential class members in a

multitude of ways, and the district court concluded that “it would have to examine the underlying contractual relationships of each transaction to distinguish between class members” and other transaction participants. Id. at 138. The Third Circuit Court of Appeals upheld the district court’s finding that such contractual analysis was not administratively feasible. Id. Although the Niaspan Court’s general discussion of the Rule 23(b)(3) framework is instructive, the case is clearly distinguishable on its facts.26 There are no individualized contractual arrangements that need to be consulted to ascertain the identity of class members here. Instead, there are only Trans Union’s input/output logs, which have a uniform format that includes nine-digit social security number. PACER either does or does not contain public

records of bankruptcies associated with the nine-digit social security numbers from the input/output logs. This binary is a far-cry from the complex contractual relationships that complicated the analysis in Niaspan. Thus, Niaspan helps to illustrate why Plaintiff’s class determination method is administratively feasible. Plaintiff’s streamlined, objective process for ascertaining class membership will “save[] the resources of both the courts and the parties by permitting” Plaintiff’s class-wide FCRA claims “to be litigated in an economical fashion under Rule 23.” Id. at 132 (quoting Califano v.

26 Counsel for Trans Union agreed that Niaspan is not factually similar to the present case. (July 9, 2024 Hrg. Tr. at 107:9–19.) Yamasaki, 442 U.S. 682, 701 (1979)). The Court discerns no “administrative burdens that are incongruous with the efficiencies expected in a class action.” Id. (quoting Marcus, 687 F.3d at 593). The Court notes in closing that it would be a peculiar result if the present class were not ascertainable today, even though essentially the same class was certified twenty years ago in Clark, especially given the advances in modern technology for searching such data.27 Notable

too is the absence of any direct evidence from a Trans Union employee explaining exactly why Trans Union is incapable of searching LTFS data.28 Plaintiff has demonstrated by a preponderance of the evidence that the proposed class is ascertainable. Accordingly, the Court considers whether the proposed class meets the requirements of Rule 23(a) and (b)(3). B. Rule 23(a) Now that the Court has found that Plaintiff met the preliminary requirement of

27 The Court raises a judicial eyebrow at the fact that counsel for neither side was able to explain at the July 9 hearing how the class was ascertained in Clark. (See July 9, 2024 Hrg. Tr. at 73:11–74:3, 75:11– 76:6, 95:11–17, 113:7–114:9, 116:24–118:1.) Counsel for Trans Union represented that Plaintiff had “never asked” for information about ascertaining the Clark class, but Plaintiff’s counsel protested that they had submitted an interrogatory about ascertaining the Clark class to Trans Union during discovery. (Id. at 95:11–17, 116:23–117:3.) After the hearing concluded, Trans Union filed a Motion for Leave to Supplement the Record, requesting leave to add both Trans Union’s supplemental response to that interrogatory and several items from the public docket in Clark. (Doc. No. 92.) Plaintiff does not oppose this motion. (See Doc. No. 93 at 1.) However, the Court notes that Trans Union’s submissions appear to militate in favor of granting class certification here. Specifically, Trans Union has submitted the December 2003 declaration of William Stockdale, Trans Union’s Senior Director of Customer Information Services, in which Stockdale states that Trans Union successfully removed from the final class list “those consumers who had filed for bankruptcy during the previous 10 years based on public record data.” (Doc. No. 99 ¶¶ 1, 6.) While Trans Union’s late-arriving supplement to the record underscores the Court’s conclusion that determining whether potential class members have filed for bankruptcy using the public record is reliable and feasible (because it has been done before), this evidence is not necessary to the Court’s conclusion. 28 The Court notes that defense counsel was already “shocked” once to find that his client could isolate input/output logs for June 2022, even though he thought Trans Union could not achieve it. (See July 9, 2024 Hrg. Tr. at 122:15–25.) ascertainability, we turn next to the requirements set forth in Federal Rule of Civil Procedure 23. Rule 23(a) requires that (1) the class is so numerous that joinder is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class. Fed. R. Civ. P. 23(a). Trans Union only

challenges commonality. (Doc. No. 111 at 26.) As such, the Court briefly addresses numerosity, typicality, and adequacy. Then the Court turns to commonality as part of the Court’s discussion of predominance under Rule 23(b)(3).29 1. Numerosity The class that Plaintiff seeks to certify is so numerous that it would be impracticable to join all members in a single action. The Third Circuit Court of Appeals has explained that, while there is “[n]o minimum number of plaintiffs . . . required to maintain a suit as a class action, . . . generally if the named plaintiff demonstrates that the potential number of plaintiffs exceeds 40,” the numerosity requirement is satisfied. Stewart v. Abraham, 275 F.3d 220, 226–27 (3d Cir.

2001). In the month of June 2022 alone, Plaintiff’s proposed method for determining class membership located 581 class members. (See Doc. No. 107-29 ¶¶ 9–10.) Even though Trans Union failed to provide input/output logs for January 2020 and March 2021, Trans Union did identify 17,098 files from January 2020 and 14,816 files from March 2021 that were the subject of a hard inquiry and contained tradelines with a bankruptcy remark but no corresponding public

29 When the proponent of class certification moves under Rule 23(b)(3), as Plaintiff has in the present action, the Rule 23(a)(2) commonality requirement is subsumed by the Rule 23(b)(3) predominance requirement. As the Third Circuit Court of Appeals has explained, “we consider the Rule 23(a) commonality requirement to be incorporated into the more stringent Rule 23(b)(3) predominance requirement, and therefore deem it appropriate to ‘analyze the two factors together, with particular focus on the predominance requirement.’” Sullivan v. DB Investments, Inc., 667 F.3d 273, 297 (3d Cir. 2011) (quoting In re Ins. Brokerage Antitr. Litig., 579 F.3d 241, 266 (3d Cir. 2009)). record of a bankruptcy action. (Doc. No. 107-23 at 6.) Plaintiff’s analysis of the June 2022 input/output logs determined that 270 of 97330 such files for which 9-digit social security numbers were available, or 27.7%, returned no bankruptcy on PACER, and 311 of those files, or 32.0%, returned only a stale bankruptcy that was more than ten years old. (Doc. No. 107 at 19.) If these proportions hold steady, January 2020 would yield an additional 4,736 files returning no

bankruptcy on PACER and 5,471 files returning only a stale bankruptcy. March 2021 would yield 4,741 files that returned no bankruptcy on PACER and 4,623 files that returned only a stale bankruptcy.31 These numbers represent class membership over just three months, and the class period is 37 months long. The Court thus finds that the numerosity requirement is met. 2. Typicality Plaintiff’s claims and defenses are typical of the class. The typicality requirement “ensures the interests of the class and the class representatives are aligned ‘so that the latter will work to benefit the entire class through the pursuit of their own goals.” In re National Football League Players Concussion Injury Litig., 821 F.3d 410, 427–28 (3d Cir. 2016) (quoting Newton

v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 182–83 (3d Cir. 2001)). The Third Circuit Court of Appeals has “set a ‘low threshold’ for typicality.” Id. at 428 (quoting Newton, 259 F.3d at 183). “‘Even relatively pronounced factual differences will generally not preclude a finding of typicality where there is a strong similarity of legal theories’ or where the claim arises

30 Plaintiff’s counsel mentioned in passing for the first time at the July 9, 2024 hearing that only 973 of the 1,100 input/output logs from June 2022 contained a nine-digit social security number. (July 9, 2024 Hrg. Tr. at 85:1–12.) There has been no explanation nor argument about the meaning of this fact. However, the Court notes its confusion about how Trans Union could sell a credit report about an individual to a third party if Trans Union does not have the nine-digit social security number of that individual. 31 Multiple methods exist to de-duplicate any repetitious files. (Doc. No. 107-26 at 110:2–15.) from the same practice or course of conduct.” In re Prudential Ins. Co. Am. Sales Practice Litig. Agent Actions, 148 F.3d 283, 311 (3d Cir. 1998) (quoting Baby Neal for and by Kanter v. Casey, 43 F.3d 48, 58 (3d Cir. 1994)). Plaintiff’s claim is typical of the class. Trans Union sold a credit report about Plaintiff during the class period that contained an inaccurate bankruptcy remark on a tradeline and lacked

a public record of a bankruptcy action. (Doc. No. 107 at 14–16.) Because Plaintiff has never filed for bankruptcy, a search of Plaintiff’s nine-digit social security number on PACER confirms there is no public record of bankruptcy. (See Doc. No. 107-28 (screenshot of PACER search).) Plaintiff seeks the same statutory damages pursuant to 15 U.S.C. § 1681n that he seeks on behalf of the class. (See Doc. No. 13 at 13 (Prayer for Relief); Doc. No. 107 at 26.) The sole distinction between Plaintiff and some class members is that, while Plaintiff never filed for bankruptcy, some class members filed for bankruptcy more than ten years prior to Trans Union’s sale of the report. Nonetheless, Plaintiff’s theory of liability under 15 U.S.C. § 1681e is the same for class members who never filed for bankruptcy as for those who had only a stale bankruptcy

wrongfully reported: Trans Union failed to “follow reasonable procedures to assure maximum possible accuracy” of bankruptcy notations by declining to apply the Clark Rule to non-joint accounts when selling class members’ credit reports. 15 U.S.C. § 1681e(b). Plaintiff has met the low threshold of Rule 23(a)(3) and demonstrated by a preponderance of the evidence that his claims and defenses are typical of the class. 3. Adequacy Plaintiff has also demonstrated by a preponderance of the evidence that he “will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). Satisfaction of the adequacy requirement “depends on two factors: (a) the plaintiff’s attorney must be qualified, experienced, and generally able to conduct the proposed litigation, and (b) the plaintiff must not have interests antagonistic to those of the class.” New Directions Treatment Servs. v. City of Reading, 490 F.3d 293, 313 (3d Cir. 2007) (quoting Wetzel v. Liberty Mut. Ins. Co., 508 F.2d 239, 247 (3d Cir. 1975)). The second factor aims “to uncover conflicts of interest between named parties and the class they seek to represent.” Id. (quoting In re Warfarin Sodium Antitrust

Litig., 391 F.3d 516, 532 (3d Cir. 2004)). Plaintiff satisfies both factors of the adequacy analysis. His attorneys are highly qualified, experienced, and capable. Plaintiff’s law firm, Francis Mailman Soumilas (“FMS”), has served as class counsel in over 70 class actions. (Doc. No. 107 at 26.) And FMS has been recognized for specialized expertise in litigating FCRA cases such as this one. See, e.g., White v. Experian Info. Sols., 993 F. Supp. 2d 1154, 1169 (C.D. Cal. 2014) (describing FMS as “FCRA specialists”); McIntyre v. RealPage, Inc., No. 18-cv-03934, 2023 WL 2643201, at *3 n.5 (E.D. Pa. Mar. 24, 2023) (noting FMS’s “significant experience litigating [FCRA] class actions” and “particular skill and efficiency” in representing class); Ramirez v. Trans Union, LLC, No. 12-cv-

00632, 2022 WL 17722395, at *6 (N.D. Cal. Dec. 15, 2022) (“Courts have consistently recognized [FMS] ‘for its expertise in FCRA litigation and the high caliber of its work for the classes it represents.” (quoting briefing)). Plaintiff’s attorneys certainly satisfy the standards of Rule 23(a)(4). The Court is aware of no conflict of interest between Plaintiff and the class he seeks to represent. Upon learning of the incorrect bankruptcy notation on his credit report, Plaintiff diligently investigated, gathered documentation, and contested the inaccurate report. (See Doc. No. 107 at 14–16.) He is knowledgeable about the case and the Court is confident that Plaintiff, along with his highly accomplished counsel, will adequately represent the interests of class members. Plaintiff has satisfied the adequacy requirement by a preponderance of the evidence. * * * In sum, the numerosity, typicality, and adequacy requirements of Rule 23(a) are met. That leaves the predominance and superiority requirements of Rule 23(b)(3). C. Rule 23(b)(3)

Rule 23(b)(3) requires both: “[i] that the questions of law or fact common to class members predominate over any questions affecting only individual members, and [ii] that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). As noted above, when the proponent of class certification moves under Rule 23(b)(3), as Plaintiff has in the present action, the Rule 23(a)(2) commonality requirement is subsumed by the Rule 23(b)(3) predominance requirement. The Court will first analyze the commonality and predominance requirements together before turning to superiority. 1. Commonality and Predominance

Rule 23(a)(2) requires the movant to demonstrate that their claims “depend upon a common contention,” the resolution of which “will resolve an issue that is central to the validity of each one of the claims in one stroke.” Wal-Mart, 564 U.S. at 350. “This does not mean merely that they have all suffered a violation of the same provision of law.” Id. The commonality test is meant to ensure that “claims can productively be litigated at once.” Allen v. Ollie’s Bargain Outlet, Inc., 37 F.4th 890, 900 (3d Cir. 2022) (quoting Wal-Mart, 564 U.S. at 350). It is not sufficient for the movant to merely pose a hypothetical question; “[t]here must be evidence the class proceeding will likely ‘produce a common answer’” to the question posed. Id. at 901 (quoting Wal-Mart, 564 U.S. at 352). The predominance requirement of Rule 23(b)(3) “imposes a more rigorous obligation upon a reviewing court” than the commonality requirement of Rule 23(a)(2). Sullivan v. DB Investments, Inc., 667 F.3d 273, 297 (3d Cir. 2011) (citing In re Ins. Brokerage Antitrust Litig., 579 F.3d 241, 266 (3d Cir. 2009)). The predominance inquiry “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” In re Ins. Brokerage, 579 F.3d at 266. The focus of this inquiry “is on whether

the defendant’s conduct was common as to all of the class members, and whether all of the class members were harmed by the defendant’s conduct.” Sullivan, 667 F.3d at 298. Trans Union argues that Plaintiff has failed to establish commonality or predominance in this case. First, although Trans Union agrees that Plaintiff has identified two common questions—the reasonableness of Trans Union’s procedures and whether those procedures are so unreasonable that they rise to the level of willful—it disagrees that Plaintiff has offered answers to these questions that can be proven with common evidence. (See Doc. No. 111 at 26.) Second, Trans Union argues individualized issues predominate over common questions and the need for individualized proof of the inaccuracy of each class member’s credit report defeats a finding of

predominance. (See id. at 27–32.) The Court addresses each argument in turn. a. Commonality Plaintiff asserts that Trans Union’s decision not to apply the Clark Rule to non-joint accounts for almost twenty years was both (a) unreasonable under § 1681e(b) of the FCRA and (b) willful. (See Doc. No. 108 at 26.) These questions are relevant to each class member’s claims against Trans Union. Yet, the Supreme Court has explained that the Court cannot merely identify questions common to the class—in this instance whether Trans Union’s policy was reasonable and willful—the Court “must answer the very question . . . asked.” Allen, 37 F.4th at 901; see also Wal-Mart, 564 U.S. at 349 (explaining that all “competently crafted” classes raise banal common questions, but Rule 23(a)(2) requires “the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation” (quoting Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L. Rev. 97, 131–32 (2009))). Here, a class proceeding is well-adapted to resolving the reasonableness and willfulness of Trans Union’s decision not to apply the Clark Rule to non-joint accounts. The evidence

necessary to substantiate this contention is uniform across all class members. The only evidence relevant to determining whether Trans Union’s policy was unreasonable and willful is related to Trans Union’s formulation and implementation of its own company-wide policies. The non- application of the Clark Rule to joint accounts was indisputably Trans Union’s policy following settlement of the Clark litigation, and this policy equally impacted all proposed class members— each of whom was the subject of a Trans Union credit report sold to a third party that indicated bankruptcy on a tradeline but lacked a corresponding public record of a bankruptcy within 10 years of the date the report was sold. These reports would not have been sold if the Clark Rule had applied to these class members.32 This is partially demonstrated by the fact that Trans Union

decided to expand the Clark Rule to apply to class members and by early 2023 completed this expansion. (Doc. No. 107-3 at 28:15–29:1, 30:8–10, 35:15–36:14.) Thus, this class of individuals ceased to exist altogether. (See id. at 29:1–41:9.) Relevant in this regard, the Court also notes that the proposal to expand the Clark rule came from Trans Union’s legal department

32 This is as true of class members who have never filed for bankruptcy as it is of class members who filed for bankruptcy over ten years ago. If an individual has an eleven-year-old bankruptcy, for example, there might be a stale reference to a bankruptcy record in the public record section of Trans Union’s internal consumer file, but the reference would not be reportable in a credit report according to Trans Union’s own reasonable policy and should be suppressed. If Trans Union had cross-checked tradelines with public records of bankruptcies pursuant to the Clark Rule, then it presumably would have caught that stale public record bankruptcies could not appear on a credit report. Since it can’t appear on a credit report next to the tradeline, the bankruptcy could not be reported at all pursuant to the Clark Rule. and “was driven by ongoing litigation . . . [in] multiple cases.” (Id. at 29:2–11.) Contrasting this case with Wal-Mart underlines why Rule 23(a)(2) is satisfied here. In Wal-Mart, the plaintiffs sought to certify a class of approximately 1.5 million female Wal-Mart employees, alleging that Wal-Mart violated Title VII by denying class members equal pay and promotions on the basis of sex. Wal-Mart, 564 U.S. at 342–43. In reversing the lower courts’

grant of class certification, the Supreme Court highlighted that Wal-Mart’s express policy forbade sex discrimination and gave managers discretion over employment matters, explaining that “demonstrating the invalidity of one manager’s use of discretion will do nothing to demonstrate the invalidity of another’s.” Id. at 353–56. By contrast, Trans Union’s express policy declined to apply the Clark Rule to non-joint accounts as a categorical matter. (See Doc. No. 107-3 at 20:21–21:3, 27:21–29:1.) Whereas Wal-Mart managers’ discretion refracted Wal-Mart’s policy against sex discrimination into 1.5 million fact patterns, Trans Union’s policy applied with equal, laser-like precision to every proposed class member. A jury could conclude that Trans Union’s policy was either reasonable

or unreasonable based on common evidence, and the jury would not need to do a case-by-case review to determine how Trans Union’s uniform policy applied to each class member. Instead, the reasonableness of Trans Union’s policy and Trans Union’s intent in enacting that policy will be evidenced by the realities of the credit reporting industry and Trans Union’s deliberations about how to conduct its business. Although individual cases could be used to illustrate how Trans Union’s policy operates, these examples would only serve to exemplify the overall (ir)rationality of the policy as it was uniformly applied—not to demonstrate how Trans Union’s policy was mis-applied in individual cases. Cf. Wal-Mart, 564 U.S. at 355 (“The only corporate policy that the plaintiffs’ evidence convincingly establishes is Wal-Mart’s ‘policy’ of allowing discretion by local supervisors over employment decisions. On its face, of course, that is just the opposite of a uniform employment practice that would provide the commonality needed for a class action; it is a policy against having uniform employment practices.” (original emphasis)). Plaintiff has thus satisfied the commonality requirement by a preponderance of the evidence. b. Predominance

Next, Trans Union argues that Plaintiff fails to demonstrate that common questions about the reasonableness and willfulness of Trans Union’s policy predominate over individualized questions regarding the accuracy of each report sold by Trans Union. “Considering whether questions of law or fact common to class members predominate begins, of course, with the elements of the underlying cause of action.” Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809 (2011) (internal quotation marks omitted). The elements of a cause of action under § 1681e(b) are: “(1) inaccurate information was included in a consumer’s credit report; (2) the inaccuracy was due to defendant’s failure to follow reasonable procedures to assure maximum possible accuracy; (3) the consumer suffered injury; and (4) the consumer’s injury

was caused by the inclusion of the inaccurate entry.” Cortez, 617 F.3d at 708 (quoting Philbin, 101 F.3d at 963). Trans Union argues that individualized issues regarding the first element—whether a report was inaccurate—predominate over common issues regarding Trans Union’s policy of limiting the Clark Rule to joint accounts. (Doc. No. 111 at 27–32.) If Plaintiff’s PACER-search method for determining the existence of public records of bankruptcies based on comparing nine-digit social security numbers sweeps in too many individuals who have filed for bankruptcy, then, Trans Union reasons, individual issues regarding whether each class member has filed for bankruptcy will predominate over common issues regarding Trans Union’s policy. (See id. at 30–33.) Basically, the parties disagree about what constitutes competent evidence for demonstrating whether an individual has filed for bankruptcy. Plaintiff proposes a unitary standard for determining whether a potential class member filed for bankruptcy: [W]here a search of PACER using the nine-digit social security number of the individual who was the subject of the report turns up no results because there is no bankruptcy record with a matching [social security number], that is sufficient evidence to conclude that the individual did not file for bankruptcy.

(Doc. No. 108 at 17.) Because the class includes stale bankruptcy-filers, Plaintiff also submits that “where the only publicly available bankruptcy records that match an individual’s [social security number] are more than ten years old, that is evidence that any reference to a bankruptcy is too outdated to accurately reflect the individual’s credit history.” (Id. at 20–21.) And, Plaintiff asserts that, because nine-digit social security numbers are “the quintessential unique identifier for U.S. consumers,” (id. at 17), matching nine-digit social security numbers to public records of bankruptcy filings yields competent evidence to demonstrate whether class members have filed for bankruptcy. Cf. Thompson v. San Antonio Retail Merchants Ass’n, 682 F.2d 509, 511 (5th Cir. 1982) (per curiam) (“The social security number is the single most important identifying factor for credit-reference purposes.”). By contrast, Trans Union asserts that three different categories of potential class members may have, in fact, filed for bankruptcy, arguing that inquiring into these categorizations will require individualized mini-trials. (See Doc. No. 111 at 22–24.) The first category consists of class members with matching names and full addresses on both the credit report sold by Trans Union and a public record of a bankruptcy on PACER. (See Doc. No. 111-5 ¶ 18.) This category is based on evidence that, out of 50 of the 270 individuals from June 2022 for which Plaintiff’s method of searching for nine-digit social security numbers returned no public record of a bankruptcy on PACER, Trans Union found that 17 public records of bankruptcies on PACER exactly matched “the name, street address, city, state, and zip code” of the allegedly inaccurate report Trans Union sold. (Id.) The second category of class members who, according to Trans Union, may have, in fact,

filed for bankruptcy consists of class members with matching names and states on both the credit report sold by Trans Union and a public record of a bankruptcy on PACER. (See id. ¶ 19.) This category is based on evidence that, out of 50 of the 270 individuals from June 2022 for which Plaintiff’s method returned no public record of a bankruptcy on PACER, Trans Union found 13 public records of bankruptcies on PACER “where the name of a debtor, as well as the state in the address from PACER, matched the corresponding name and state from the Trans Union” credit report. (Id.) The third category consists of class members with a matching name, address, and last four social security number digits on both the credit report sold by Trans Union and the public

record of a bankruptcy on PACER. (See id. ¶ 5.) Trans Union’s supplemental exhibit illustrates this category.33 Specifically, out of 24 primary files lacking a public record of a bankruptcy for which Wodzinski failed to find a secondary file containing a public record of a bankruptcy, Trans Union’s counsel found 19 of those primary files matched the name, address, and last four digits of social security number of a public record of a bankruptcy on PACER. (Id. ¶ 5.) Trans Union contends this supports its argument that individualized “mini-trials” will need to take

33 As discussed in the Court’s contemporaneously filed Memorandum ruling on Plaintiff’s Motion to Strike Certain Documents and Motion to Preclude the Expert Report and Testimony of Corinne Wodzinski, (Doc. Nos. 56, 109), the Court strikes this exhibit as a supplement to Wodzinski’s expert report but allows it as factual evidence. place in order to determine whether class members who fall into any of these three categories have in fact filed for bankruptcy. (Doc. No. 111 at 29–30.) After careful consideration, the Court concludes that Trans Union’s argument is a red herring that obfuscates more than illuminates the predominance inquiry. Trans Union’s argument boils down to the proposition that there is no single conclusive way (nine-digit social

security number or otherwise) to determine whether individuals have filed for bankruptcy because bankruptcy data can be matched with individuals in many different ways.34 But if Trans Union’s theory was true, Trans Union likewise would be unable to sell credit reports with any reasonable degree of certainty. Cf. Soutter v. Equifax Information Services, LLC, 307 F.R.D. 183, 197–98 (E.D. Va. 2015) (“In general, courts do not look favorably upon the argument that records a defendant treats as accurate for business purposes are not accurate enough to define a class.”). Trans Union’s shifting “standards” for suggesting that a public record of a bankruptcy may belong to a class member—name plus full address (category one); name plus state (category

two); name plus address plus four-digit social security number (category three)—simply confuse the issues. There is no evidence before the Court to suggest that any of these three standards is a sufficient, objective basis for determining whether the public record of a bankruptcy from PACER actually belongs to the consumer who is the subject of the report Trans Union sold.

34 The Court agrees with Plaintiff that Trans Union’s argument “boils down to an assertion that inaccuracy can never be determined without consulting the consumer.” (Doc. No. 108 at 25.) However, there is no evidence before the Court that either party ever actually tried interviewing potential class members. Thus, it appears to the Court that the parties are simply arguing about what method, short of interviewing class members, is best for matching public records of bankruptcy actions to individuals. The Court notes that matching records with individual consumers is precisely what Trans Union does to create the credit reports it sells, (see July 9, 2024 Hrg. Tr. at 39:10–40:3), and there is no evidence before the Court that individual interviews constitute any part of Trans Union’s process. Such a contention is merely conjectural. Cf. In re U.S. Foodservice Inc. Pricing Litig., 729 F.3d 108, 121–22 (2d Cir. 2013) (explaining that “conjectural ‘individualized questions . . . do not undermine class cohesion and thus cannot be said to predominate for purposes of Rule 23(b)(3)’” (quoting Amgen Inc. v. Conn. Retirement Plans & Tr. Funds, 568 U.S. 455, 469–70 (2013))). And, importantly, Trans Union has an objective algorithm that links files, but that algorithm did

not link the public records that form the evidentiary basis for categories one through three with the files that Trans Union sold.35 Because Trans Union cannot offer an objective basis for determining whether these public records of bankruptcies belong to the subject of their credit report, Trans Union offers three somewhat arbitrary categories that could suggest they belong together. However, these categories and the data that undergirds them merely constitute evidence that the class may not be able to prove that Trans Union’s reports were inaccurate, which is a necessary element of their FCRA claims. A jury will decide this issue. The key insight here is that the dispute between Plaintiff and Trans Union about how to determine whether class members have filed for bankruptcy can be resolved “in one stroke” on

the merits. Wal-Mart, 564 U.S. at 350. Of course, “[b]ecause the decision whether to certify a class ‘requires a thorough examination of the factual and legal allegations,’ the court’s rigorous analysis may include a ‘preliminary inquiry into the merits[.]’” In re Hydrogen Peroxide, 552 F.3d at 317 (quoting Newton, 259 F.3d at 166, 168). The Court may also “consider the substantive elements of the plaintiffs’ case in order to envision the form that a trial on those issues would take.” Id. (quoting Newton, 259 F.3d at 166). In a trial on the merits, Plaintiff appears poised to argue that the lack of a public record of a bankruptcy on PACER matching the

35 Wodzinski confirmed that whenever a public record of a bankruptcy was segregated from a primary file containing a tradeline with a bankruptcy notation, this data-segregation occurred because of Trans Union’s matching logic. (See July 9, 2024 Hrg. Tr. at 17:25–18:7.) nine-digit social security number on Trans Union’s credit report establishes the inaccuracy of that report. Trans Union appears poised to argue that it does not. Resolving this issue will not require tens of thousands of mini-trials. Instead, it will require class-wide evidence about how to objectively determine, by using both Trans Union’s files and the public record, whether class members have filed for bankruptcy—class-wide evidence that saturates the record presently

before the Court. At this juncture, Trans Union has mis-framed its evidence on the merits of Plaintiff’s class claim as demonstrating that the predominance requirement has not been met.36 The Court finds that Plaintiff has satisfied the commonality and predominance requirements of Rule 23(a) and (b)(3). 2. Superiority Trans Union further argues that Plaintiff fails to meet the superiority requirement under Rule 23(b)(3). A class action certified under Rule 23(b)(3) must be “superior to other available methods” for adjudicating class members’ claims. Several factors are relevant here: (A) the class members’ interests in individually controlling the prosecution or defense of separate actions;

(B) the extent and nature of any litigation concerning the controversy already begun by or against class members;

(C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and

(D) the likely difficulties in managing a class action.

Fed. R. Civ. P. 23(b)(3). This requirement “mandates that the district court determine that the class action is the best method of fairly and efficiently resolving the controversy.” Johnston v.

36 Trans Union helps frame one of the issues for trial: whether the lack of a public record of a bankruptcy with a matching nine-digit social security number on PACER in the ten years preceding Trans Union’s credit report sufficiently demonstrates that Trans Union’s report was inaccurate under 15 U.S.C. § 1681e. HBO Film Mgmt., Inc., 265 F.3d 178, 185 (3d Cir. 2001). Individual class members do not have a strong interest in individually controlling this litigation because the two issues animating this action are resolvable with class-wide evidence. First, whether Trans Union’s decision not to apply the Clark Rule to non-joint accounts was reasonable and willful requires no individualized evidence because the policy was uniformly

applied. Second, whether the lack of a public record of a bankruptcy on PACER with a matching nine-digit social security number in the ten years preceding Trans Union’s report sufficiently demonstrates that Trans Union’s report was inaccurate is a question the jury can resolve in one stroke. Trans Union’s argument against superiority simply recycles its argument against predominance. Trans Union asserts that individualized inquiries into whether class members have filed for bankruptcy will render this class action unmanageable. (Doc. No. 111 at 34.) But, as discussed above, the Court disagrees. The inaccuracy element of the class FCRA claim is resolvable on the merits. Moreover, here the Plaintiff seeks only statutory, not individualized

damages which also alleviates the concern over individualized inquiries. And the Court is not aware of any class members who would prefer to advance individualized evidence that they did not file for bankruptcy.37 Even assuming that such members exist, that interest is strongly outweighed by the many claims that will be litigated via the class action mechanism but not through individual litigation. Trans Union is alleged to have violated the consumer rights of tens of thousands of class members spread across the entire nation. It is far more desirable and efficient to resolve these

37 Trans Union’s counsel confirmed that he was unaware of litigation against Trans Union brought by potential class members in an individual capacity. (July 9, 2024 Hrg. Tr. at 116:3–12.) tens of thousands of cases in one class action lawsuit than to have scattered individual lawsuits across the country, which will inevitably fail to address the entire claims of the entire class. See In re Prudential, 148 F.3d at 316 (affirming district court’s finding of superiority where it “examined the relatively modest size of individual claims and the sheer volume of those claims in the aggregate, and concluded a class action presented the ‘only rational avenue of redress for

many class members’” (quoting district court)). Thus, a class action is “the best method of fairly and efficiently resolving the controversy.” Johnston, 265 F.3d at 185. The Court finds Plaintiff has satisfied the superiority requirement under Rule 23(b)(3). * * *

In sum, Plaintiff’s proposed class is ascertainable and meets the requirements of Rule 23(a) and (b)(3) by a preponderance of the evidence. III. Appointing Class Representatives Appointment of class representatives is governed by Rule 23(a) of the Federal Rules of Civil Procedure, which requires that their claims be “typical of the claims . . . of the class” and that they “will fairly and adequately protect the interests of the class.” For the reasons previously discussed in Part II.B.2, those requirements are met here, and the Court will appoint Plaintiff William Norman Brooks, III as representative of the Class. IV. Appointing Class Counsel Rule 23(g) of the Federal Rules of Civil Procedure provides the standard for the appointment of class counsel. Class counsel are responsible not only for representing the interests of the class representative, but also for “fairly and adequately represent[ing] the interests of the class.” Fed. R. Civ. P. 23(g)(4). As discussed above in Part II.B.3, Rule 23(g) provides a non-exclusive list of factors courts should consider when appointing class counsel. For the reasons previously discussed, the Court finds that Francis Mailman Soumilas has the ability and resources to “fairly and adequately represent the interests of the class.” Accordingly, Francis Mailman Soumilas will be appointed as Class Counsel. V. Trial Plan Because the Court is granting Plaintiff's motion for class certification, we must also

address Trans Union's arguments about the scope of the issues that will define this litigation as it moves forward on a class-wide basis. Rule 23(c)(1)(B) provides that “[a]n order certifying a class action must define the class and class claims, issues or defenses.” Fed. R. Civ. P. 23(c)(1)(B). The Third Circuit Court of Appeals has explained that this provision requires that the Court’s certification order, or “an incorporated memorandum opinion,” contain “the precise parameters defining the class and a complete list of the claims, issues, or defenses to be treated on a class basis,” all of which must be “readily discernible from the text.” Wachtel ex rel. Jesse v. Guardian Life Ins. Co. of America, 453 F.3d 179, 185 (3d Cir. 2006). To this end, the Third Circuit has encouraged district courts to use trial plans submitted by the proponent of class

certification. See id. at 188 n.7 (describing the submission of trial plans as “an advisable practice within the class action arena”). Here, Trans Union complains that Plaintiff’s trial plan does not contemplate an inquiry into the accuracy element of his FCRA claim at trial. (Doc. No. 111 at 35.) But just because the Third Circuit considers the submission of trial plans an "advisable practice," that does not render a plaintiff’s trial plan the last word on how a case will proceed. Instead, as the Wachtel opinion makes clear, it is the Court--through the certification order and memorandum opinion--that sets the “precise parameters defining the class” and the “claims, issues, or defenses to be treated on a class basis.” Wachtel, 453 F.3d at 185. This Memorandum and its accompanying Order articulate that one of the issues in this case is the sufficiency of Plaintiff’s proposed means of determining whether class members have filed for bankruptcy for purposes of establishing inaccuracy under 15 U.S.C. § 1681e. Plaintiff's proposed trial plan does not supersede that determination. VI. CONCLUSION

For the reasons discussed above, the Court also grants Plaintiff’s motion for class certification. An appropriate Order follows.