In Tomeo v. CitiGroup, Inc., 2018 WL 4627386, at *9–12 (N.D.Ill., 2018), Judge Ellis declined to certify a wrong-number TCPA class.
Here, Citi has put forth specific evidence establishing that a significant percentage of the putative class consented to receiving calls. Hansen intentionally avoided stating an opinion on the issue of consent in the Hansen Report, Doc. 121 Ex. 6 ¶ 63, and Citi’s experts noted that the groups that Hansen identified as individuals contacted after Citi coded the account as a Wrong Number, Do Not Call, Cease and Desist, or Case 998 (the “Flags”) did not necessarily constitute individuals who did not consent to contact. In the Taylor Rebuttal, Taylor reviewed a sample of over 1,000 Individual Note Screens and concluded that, because the Flags were not always accurate and accountholders frequently went back and forth between giving consent and revoking it, the only way to determine consent for each accountholder is to individually review Citi’s files on each accountholder. Doc. 121 Ex. 15 ¶¶ 16–19. Taylor supports his opinion with specific evidence: regarding the Cease and Desist Class, he looked at accounts with Case 998 flags, reviewing about 15% of the accounts in the Case 998 data that Hansen reviewed, and 17% of the time, the accountholder re-consented to contact. Id. ¶¶ 32–33. Because some accounts in Hansen’s lists received significantly more calls than others, the number of calls disqualified by individual analysis could be much higher. Similarly, with the Wrong Number Class, Taylor examined a large portion of the accounts within the Individual Note Screens production that contained the Wrong Number flag and determined that 15% of those accounts appeared to be instances where the number did in fact belong to a borrower—instances where either the so-called wrong number later placed an incoming call to Citi, or where multiple phone numbers for an account were all flagged as wrong numbers. Id. ¶ 49. As with the consent issue, Taylor noted that the process for determining actual wrong numbers was “time-intensive” and “manual” and required review of each Individual Note Screen to know with certainty that the Citi placed the call to a Wrong Number. Id. ¶¶ 45, 48. In addition, Sponslor conducted a more exhaustive review of the four original named plaintiffs in his rebuttal. In his review, he considered the entire Citi file for each potential named plaintiff—Individual Note Screens, FileNet, DRI system information, and recordings of calls Citi maintains. He determined that, whenever Citi called any of the three individuals other than Tomeo, it had consent to do so. Doc. 121 Ex. 13 ¶¶ 48, 54, 57. With regard to Tomeo, he found that Tomeo’s mother had represented to Citi that it could reach her through Tomeo’s phone number, which she provided to Citi on multiple occasions. Id. ¶ 66. On this basis, Sponslor concluded that it would be necessary to consider information in the files beyond the Individual Note Screens to determine whether there was consent. Id. ¶ 40. The level of potential error within both classes is significant. . . . And even taking Hansen at his word that he could mass search the Individual Notes Screens for instances where each class member cycled between consent and revocation, the Court would still need to conduct individual inquiries into whether the interactions described in those notes constituted a change in the status of consent. The named plaintiff’s situation is instructive here: to determine whether Citi had consent to call Tomeo, the parties reviewed Individual Note Screens, FileNet, Citi recordings of calls with Tomeo, and three depositions. Doc. 146-1 at 17. Moreover, the parties still dispute whether Citi had consent to call Tomeo, so a fact finder would need to sort through this evidence to determine whether Citi had consent to place calls to Tomeo. Id. Hansen’s mass-search of the Individual Note Screen for Tomeo would not be conclusive here; the fact finder would still require further individualized inquiry to determine the issue of consent. Simply put, neither Tomeo nor his experts adequately identify a common way to address the individual variations of consent and revocation that occurred in this case.* * * * *
The Wrong Number Class is equally as problematic. The Taylor Rebuttal identifies two ways that he determined accounts flagged WRNG were in fact accountholder numbers. But the analyses that Hansen proposes in response do not convince the Court that he has developed a common way to isolate those who have already consented to calls. In Davis v. AT&T Corp., the court considered similar circumstances: in considering certification of a class of calls to wrong numbers, the court noted that “many customers tell callers they have reached the wrong number, though the customer’s number was dialed, as a ‘procrastination tool’ to avoid speaking on the phone” and “if Defendant’s customer provided a number belonging to another person, such as a spouse or other family member, an inquiry into that customer’s authority to provide consent to call that number would be required.” No. 15cv2342-DMS (DHB), 2017 WL 1155350, at *6 (S.D. Cal. 2017). These same problems are present here—although Hansen proposes that he could identify some of the situations where accountholders tell Citi it has the wrong number, his proposed method would only catch situations where the accountholder called Citi afterward, or told Citi it had the wrong number for multiple phone numbers. Situations where the accountholder merely gave consent through a form or gave a family member’s phone number would slip through the cracks. Ultimately, the Davis court found that “a complete analysis of the customer status issue would require an inquiry into each call recipient’s individual circumstances.” Id. The Court finds that the same is true here. Tomeo attempts to distinguish Davis by noting that the class definitions are different: in Davis, the class only included non-customers, while here, the class includes accountholders who previously told Citi that it called a wrong number. This distinction does not save Tomeo’s Wrong Number Class—even assuming that accountholders telling Citi representatives they had the wrong number would be enough to withdraw consent, the Court would still need to conduct an individualized consent inquiry for all of the reasons discussed above for accountholders. If anything, this adds to the individualized analysis. Moreover, Hansen has not actually executed any of the methods that he proposes for determining consent or actual wrong numbers in the Hansen Reply. Citi argues that his intention to do so later is insufficient for establishing predominance for class certification. . . . Tomeo also contends that, even if consent requires some individual attention, it is not significant enough to prevent other common issues from predominating. This goes against all of the cases cited above where courts have found that, where the defendant provides specific evidence that a significant number of putative class members consented to contact, individual issues of consent predominate. See, e.g. Legg, 321 F.R.D. at 578. . . . . Tomeo’s last argument is that the percentage of potential consent issues that Taylor identified is not material, and the parties could alternatively agree to reduce damages by 15 or 17 percent to account for the calls Taylor contends were made with consent. This argument does not hold water. The purpose of Taylor’s exercise was to identify some flaws in the potential classes, not all of the flaws. He merely considered Individual Note Screens, rather than all of the information Citi has regarding each account. As Sponslor’s more exhaustive review of the potential named plaintiffs’ files revealed, it is often necessary to go beyond the Individual Note Screens to fully determine the issue of consent. Doc. 121 Ex. 13 ¶ 40. In addition, Citi’s counsel reviewed a sample of Individual Note Screens for 1,000 accounts and determined that they would press for individualized inquiry on the issue of consent regarding more than 70 percent of those accounts. Based on this analysis, it is clear that the problems Citi has identified regarding consent in the potential Classes is material, and likely goes beyond the specific percentages that Taylor identified. Tomeo’s reliance on Mullins v. Direct Digital, LLC is entirely off-base here: Mullins addresses an entirely different aspect of class certification (ascertainability). Moreover, the quote that Tomeo cites (“the addition or subtraction of additional class members affects neither the defendant’s liability nor the total amount of damages”) contemplates a situation where money is stolen from a pension fund, and so no matter the number of employees subscribed to the pension fund who make up the class, the liability and amount of damages remain the same. Mullins v. Direct Digital, LLC, 795 F.3d 654, 670 (7th Cir. 2015). This does not apply in this situation, where the number of TCPA violations (and thus liability and damages) changes based on the number of calls, which changes based on the number of phone numbers included in the potential classes. In the face of Citi’s evidence that it had consent for a significant percentage of potential class members, Tomeo has failed to establish a way to determine consent on a classwide basis. Thus, individualized issues of consent predominate, and Tomeo has not carried his burden under Rule 23(b)(3).