Sapan v. Yelp, Inc.

• Court: District Court, N.D. California
• Decided: November 15, 2021
• Docket: 3:17-cv-03240
• Status: Unknown

Opinion

1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 JONATHAN SAPAN, Case No. 3:17-cv-03240-JD

8 Plaintiff, ORDER RE MOTION TO CERTIFY 9 v. CLASS AND MOTION TO STRIKE

10 YELP, INC., Re: Dkt. No. 66, 129, 136 Defendant. 11

12 Named plaintiff Jonathan Sapan asks to certify a national class for claims alleged under the 13 Telephone Consumer Protection Act (TCPA) against Yelp, Inc. (Yelp). Dkt. No. 66. Yelp asks to 14 strike a declaration of a putative expert filed by Sapan. Dkt. No. 129. The parties’ familiarity 15 with the record is assumed, and certification is denied. The case will go to trial on Sapan’s 16 individual claim. The motion to strike is denied. 17 DISCUSSION 18 I. CLASS CERTIFICATION 19 A. PROCEDURAL BACKGROUND 20 Before getting to the merits of certification, it is useful to recount the rather tortured 21 history of Sapan’s class certification efforts. Sapan’s theory of the case posed a challenging 22 problem for certification. Yelp operates to publish reviews of businesses and serve the advertising 23 needs of businesses and commercial enterprises. Dkt. No. 66-1 at 1. Unlike other social media 24 platforms, its operations are not directed to personal information in the first instance. As a result, 25 the exemptions in the TCPA for calls to numbers that are linked to an existing business 26 relationship, or which otherwise have provided consent to be called, are critical factors in 27 1 determining whether Yelp may liable as Sapan alleges, and whether the question of liability can be 2 answered on a classwide basis.1 3 For certification purposes, Sapan had to come up with a method that would reasonably 4 account for the TCPA exemptions without entailing individualized inquiries for each putative 5 class member. Despite a number of opportunities to do that, Sapan has not succeeded. The 6 original certification motion said almost nothing about how class members would be identified 7 within the parameters of the proposed class definition, or how the TCPA exemptions would be 8 reasonably accounted for on a classwide basis. The motion was not accompanied by any expert 9 opinions with respect to these and other Rule 23 questions. Overall, the original motion presented 10 no good reason for certifying a class in this case. 11 Rather than denying certification with finality on that paltry effort, which would have been 12 perfectly appropriate, the Court directed Sapan to file an amended motion to explain how he 13 intended to deal with the TCPA exemptions in the context of defining a class, and related issues. 14 See Dkt. No. 118; Dkt. No. 121 at 15:14-17:3. The Court granted a second bite at the certification 15 apple to ensure that the interests of the putative class were adequately addressed. See Gulf Oil Co. 16 v. Bernard, 452 U.S. 89, 100 (1981) (“[A] district court has both the duty and the broad authority 17 to exercise control over a class action and to enter appropriate orders governing the conduct of 18 counsel and parties.”). 19 At that point in the litigation, all fact and expert discovery had long been closed. Dkt. No. 20 121 at 32:1-11; 33:22-34:2. Even so, Sapan filed with the amended motion a declaration by Anya 21 Verkhovskaya, who he retained to opine on a methodology for certifying a class. Dkt. No. 122-1. 22 Yelp has objected to the declaration because Sapan proffered it after discovery and expert 23 proceedings had ended. See Dkt. No. 129. It is true that in the normal course, the Court would 24 have no hesitation to strike the untimely declaration, as Yelp urges. But the Court left the door 25 open when it asked Sapan to try again, and will not hold that against him now. The details leading 26 to this conclusion are discussed in Section II. 27 1 The Court held a hearing on the revised certification motion. See Dkt. No. 151. Another 2 major problem came to light, namely that Sapan had dropped the ball during discovery and had 3 not obtained anything close to a usable record of Yelp’s calls to putative class members, which 4 was vital evidence for certification purposes. Dkt. No. 153 at 4:14-18. This was a situation 5 entirely of his own making. Sapan requested Yelp call records during discovery, but in such an 6 overly broad and unduly burdensome fashion that the Court could not allow them to go forward 7 without modifications. See Dkt. No. 42; Dkt. No. 66-1 at 5. Sapan’s revised requests were 8 equally deficient. Dkt. No. 43; Dkt. No. 44. After these tries, Sapan inexplicably gave up, and did 9 not pursue any further requests for the records. The result is that Sapan, as he has acknowledged, 10 see Dkt. No. 153 at 4:14-18, has only a few months of call records, well short of the several-year 11 liability period alleged for the proposed class. 12 On this patchy record, Sapan asks to certify under Federal Rule of Civil Procedure 23(b)(3) 13 a class of “[a]ll persons located within the United States of America who provide a phone bill or 14 statement showing they had had a phone number subscribed to a residential tariff and provide a 15 confirmation e-mail from the federal National Do Not Call Registry showing their number was on 16 the Registry to whom Yelp transmitted more than one solicitation call within any 12-month period 17 anytime from June 5, 2013 to the present.” Dkt. No. 66-1 at 6. 18 B. DISCUSSION 19 The standards governing class certification are well established, and the Court has written 20 extensively about them. See, e.g., Meek v. SkyWest, Inc., No. 17-cv-1012-JD, 2021 WL 4461180 21 (N.D. Cal. Sep. 29, 2021). The overall goal is “to select the metho[d] best suited to adjudication 22 of the controversy fairly and efficiently.” Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 568 23 U.S. 455, 460 (2013) (internal quotations omitted) (modification in original). Plaintiffs must show 24 that their proposed classes satisfy all four requirements of Rule 23(a), and at least one of the 25 subsections of Rule 23(b). Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013); Zinser v. Accufix 26 Research Inst., Inc., 253 F.3d 1180, 1186 (9th Cir. 2001), amended by 273 F.3d 1266 (9th Cir. 27 2001). Plaintiff has elected to proceed under Rule 23(b)(3) only. Dkt. No. 66-1 at 1. As the party 1 seeking certification, plaintiff bears the burden of showing that the requirements of Rule 23 are 2 met for each proposed class. Mazza v. Am. Honda Motor Co., 666 F.3d 581, 588 (9th Cir. 2012). 3 The Court’s class certification analysis “must be rigorous and may entail some overlap 4 with the merits of the plaintiff’s underlying claim,” though the merits questions may be considered 5 to the extent, and only to the extent, that they are “relevant to determining whether the Rule 23 6 prerequisites for class certification are satisfied.” Amgen, 568 U.S. at 465-66 (internal quotations 7 and citations omitted). The class certification procedure is decidedly not an alternative form of 8 summary judgment or an occasion to hold a mini-trial on the merits. Alcantar v. Hobart Service, 9 800 F.3d 1047, 1053 (9th Cir. 2015). The decision of whether to certify a class is entrusted to the 10 sound discretion of the district court. Zinser, 253 F.3d at 1186. 11 1. Numerosity (23(a)(1)) 12 Sapan’s renewed certification motion hits serious trouble on this initial inquiry. Rule 13 23(a)(1) requires that a proposed class be “so numerous that joinder of all members is 14 impracticable.” Fed. R. Civ. P. 23(a)(1). Sapan says that he meets this requirement because Yelp 15 acknowledged in 2013 that approximately 81% of its listings had accurate phone numbers, which 16 he takes to mean that approximately 19% of the numbers were not properly on Yelp’s call lists. 17 Dkt. No. 66-1 at 8. Sapan adds that roughly 71% of telephone lines in the United States are on the 18 Do Not Call Registry, and then blends all of this together to deduce that Yelp must have made 19 millions of calls every year to numbers on the Do Not Call Registry. Id. at 8-9. 20 This is not evidence that might warrant a finding of numerosity. Sapan has not 21 demonstrated how general percentages about allegedly “wrong” numbers on Yelp’s call lists 22 translate into a reasonable estimate of numerosity for the class he proposes. See Siles v. ILGWU 23 Nat’l Retirement Fund, 783 F.2d 923, 930 (9th Cir. 1986) (plaintiff failed to establish numerosity 24 by demonstrating how many employees lost their jobs without establishing other characteristics of 25 the class); Derval v. Xaler, No. 19-cv-1881, 2020 WL 430781, at *3 (C.D. Cal. Jan. 28, 2020) 26 (establishing the number of customers without establishing how many received unwanted 27 messages failed to demonstrate numerosity). It is not enough that Yelp possibly made calls to 1 Sapan’s class definition, a person listed on the Registry must have been contacted twice within a 2 twelve-month period. Dkt. No. 66-1 at 6. Sapan has not adduced any evidence showing that 3 Yelp’s calls met his own class definition criteria. This is so even though, as discussed, the Court 4 specifically gave him an opportunity to amend his motion for class certification to address this 5 issue. Dkt. No. 118. 6 In lieu of facts, Sapan relies on the declaration of Verkhovskaya, who holds “a Bachelor of 7 Science degree from Molloy College in Long Island, New York,” and represents that she has 8 appeared in a number of TCPA cases on the topic of call identification. See Dkt. No. 122-1 at 9 ECF pp. 30-31. Verkhovskaya opines that it is possible to take a list of numbers that Yelp called 10 and analyze them to determine which numbers belong to potential class members. Id. at ¶¶ 3-7. 11 Her proposed method would entail culling Yelp’s call records to identify phone numbers that were 12 on the Do Not Call Registry for at least 31 days before Yelp called the number. Id. at ¶¶ 36-37. 13 These numbers would be analyzed in a rather unclear fashion to see if any had been called twice 14 within a twelve-month period. Id. at ¶ 38. She would cross check those numbers against a Lexis- 15 Nexis database to filter out business numbers. Id. at ¶¶ 43-45. The resulting numbers would 16 presumably constitute the class, and she would use commercial databases to connect a specific 17 person to each number. Id. at ¶¶ 46-52. She calls this process the “reverse-append” method of 18 analyzing phone call data. Id. at ¶ 51. 19 Sapan’s reliance on Verkhovskaya’s declaration is highly problematic. To start, the 20 declaration is little more than a generic summary of her reverse-append method. More than half of 21 the declaration’s 28 pages is given over to Verkhovskaya’s credentials and a listing of the cases in 22 which she has appeared. See id. at ¶¶ 9-28. She devotes a scant 6 pages to a description of the 23 reverse-append approach, and even that is presented in highly general terms. See id. at ¶¶ 29-52. 24 Nothing in the declaration is tethered to the evidence in this case. Verkhovskaya did not 25 do any work, or offer any opinions, based on the record developed during the litigation. This is 26 undoubtedly due in large measure to the fact Sapan has virtually no data for Verkhovskaya to 27 analyze as a result of failing to pursue discovery. Verkhovskaya promises that she will analyze 1 that evidence in hand, and the time for him to obtain it is long past. For reasons that Sapan did not 2 explain, Verkhovskaya did not run her analysis even on the data Sapan managed to acquire in 3 discovery. She offers no opinions about numerosity, or any other Rule 23 element. 4 Consequently, Verkhovskaya’s declaration is of no utility to the issue of certification. Her 5 proposed reverse-append method is also of questionable reliability and validity. Yelp did not 6 bring a Federal Rule of Evidence 702 or Daubert challenge on these grounds, ostensibly because 7 Verkhovskaya was not properly proffered as an expert, and so the Court need not engage in a full 8 admissibility analysis.2 It is instructive to note that other courts have found Verkhovskaya’s 9 reverse-append approach to be doubtful. See, e.g., Sandoe v. Boston Scientific Corp., 333 F.R.D. 10 4 (D. Mass. 2019). Sandoe observed that Verkhovskaya’s method had a known flaw of failing to 11 identify even named plaintiffs as class members. Id. at 9. It also found cause for concern about its 12 capacity to exclude calls within a TCPA exemption. Id. at 9-10. Verkhovskaya’s declaration did 13 not provide any reason to conclude that similar problems would not arise here.3 14 The salient point is that Sapan, after two opportunities, has not reasonably demonstrated 15 numerosity for the proposed class, and so certification flounders on the first element of Rule 23(a). 16 This not a common outcome in the Court’s experience, but because Sapan was required to prove 17 that he satisfies each of the Rule 23(a) requirements before class certification can be granted, the 18 failure to demonstrate numerosity is fatal to his motion. See Comcast, 569 U.S. at 33. 19 2. Typicality and Adequacy (23(a)(3)-(4)) 20 Mainly for the sake of completeness, the Court will review the other Rule 23(a) elements, 21 even though a full stop is already warranted by the lack of evidence of numerosity. Rule 23(a) 22

23 2 While Yelp did not file a separate Daubert motion challenging Verkhovskaya’s declaration, it mentioned the Daubert standards in a supplemental opposition brief. See Dkt. No. 124 at 12. The 24 Court declines to construe this as a proper Daubert challenge.

25 3 Even the courts that have allowed Verkhovskaya’s method and opinions acknowledged their shortcomings. See, e.g., Chintz v. Intero Real Estate Servs., No. 18-cv-5623-BLF, 2020 WL 26 731299, at *6-7 (N.D. Cal. Jul. 22, 2020) (Verkhovskaya’s reliance on inaccurate data did not render her report inadmissible, but impacted the weight of her report); Perez v. Rash Curtis & 27 Assocs., No. 16-cv-3396-YGR, 2019 WL 1491694, at *4-5 (N.D. Cal. Apr. 4, 2019) 1 also requires Sapan to demonstrate that his claims are typical of the putative class, and that they 2 are capable of fairly and adequately protecting the interests of the class. Fed. R. Civ. P. 23(a)(3)- 3 (4). 4 Yelp says that Sapan does not meet the typicality requirement because he lacks Article III 5 standing and is a professional litigant. Dkt. No. 75 at 16. These criticisms are not uncommon 6 against named plaintiffs, and are rarely well taken. Yelp has not shown that Sapan may be a 7 “professional plaintiff,” whatever that might mean, and also has not shown that a frequent litigant 8 should be disqualified on that basis from filing new cases. Nothing in the record suggests that 9 Sapan has engaged in collusive, abusive, or otherwise questionable conduct in this case. 10 For Article III standing, Yelp says that Sapan is not a typical class member because he has 11 an established business relationship with Yelp. Id. at 17. This argument is duplicative of issues 12 raised for predominance under Rule 23(b)(3), which the Court discusses below. The summary 13 point here is that Sapan has standing to sue, and typicality of claims. 14 The adequacy issue, which goes in part to the ability of plaintiff’s counsel to fairly and 15 adequately represent the class, raises more concerns. The Court has presided over this litigation 16 for some time, and concludes that many of the reasons for the denial of certification are due to the 17 questionable manner in which plaintiff’s counsel handled discovery and the presentation of 18 ostensible expert opinions and Rule 23 arguments. The Court declines to conclude that plaintiff’s 19 counsel are inadequate, but it has not been a bravura performance by any means. Yelp’s 20 objections to Sapan’s adequacy duplicate its typicality arguments, and are denied for the same 21 reasons. 22 3. Commonality (23(a)(2)) and Predominance (23(b)(3)) 23 The commonality requirement under Rule 23(a)(2) is satisfied when “there are questions of 24 law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). Because “any competently crafted 25 class complaint literally raises common questions,” the Court’s task is to look for a common 26 contention “capable of classwide resolution -- which means that determination of its truth or 27 falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” 1 “capacity of a class-wide proceeding to generate common answers apt to drive the resolution of 2 the litigation.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011) (internal quotations 3 omitted) (emphasis in original). This does not require total uniformity across a class. “The 4 existence of shared legal issues with divergent factual predicates is sufficient, as is a common core 5 of salient facts coupled with disparate legal remedies within the class.” Hanlon v. Chrysler Corp., 6 150 F.3d 1011, 1019 (9th Cir. 1998). The commonality standard imposed by Rule 23(a)(2) is, 7 however, “rigorous.” Leyva v. Medline Indus. Inc., 716 F.3d 510, 512 (9th Cir. 2013). 8 Rule 23(b)(3) sets out the related but nonetheless distinct requirement that the common 9 questions of law or fact predominate over the individual ones. This inquiry focuses on whether 10 the “common questions present a significant aspect of the case and [if] they can be resolved for all 11 members of the class in a single adjudication.” Hanlon, 150 F.3d at 1022 (internal quotations and 12 citation omitted); see also Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 422, 453 (2016). Each 13 element of a claim need not be susceptible to classwide proof, Amgen, 568 U.S. at 468-69, and the 14 “important questions apt to drive the resolution of the litigation are given more weight in the 15 predominance analysis over individualized questions which are of considerably less significance 16 to the claims of the class.” Torres v. Mercer Canyons Inc., 835 F.3d 1125, 1134 (9th Cir. 2016). 17 Rule 23(b)(3) permits certification when “one or more of the central issues in the action are 18 common to the class and can be said to predominate, . . . even though other important matters will 19 have to be tried separately, such as damages or some affirmative defenses peculiar to some 20 individual class members.” Tyson, 577 U.S. at 453 (internal quotations omitted). 21 “Rule 23(b)(3)’s predominance criterion is even more demanding than Rule 23(a),” 22 Comcast, 569 U.S. at 34, and the main concern under subsection (b)(3) is “the balance between 23 individual and common issues.” In re Hyundai and Kia Fuel Economy Litigation, 926 F.3d 539, 24 560 (9th Cir. 2019) (en banc) (internal quotations omitted). The Court finds it appropriate to 25 assess commonality and predominance in tandem, with a careful eye toward ensuring that the 26 specific requirements of each are fully satisfied. See, e.g., Just Film, Inc. v. Buono, 847 F.3d 27 1108, 1120-21 (9th Cir. 2017). 1 Yelp raises a predominance issue principally with respect to determining whether an 2 individual call may come within an exception to the TCPA. Dkt. No. 75 at 11-14. Under the 3 TCPA, “certain unsolicited advertisements sent via phone” are unlawful. True Health 4 Chiropractic, Inc. v. McKesson Corp., 896 F.3d 923, 926 (9th Cir. 2018); 47 U.S.C. § 227(b)(1). 5 Such unsolicited advertisements may be sent though, if “(1) the sender and recipient have ‘an 6 established business relationship,’ (2) the recipient voluntarily provided his or her contact 7 information to the sender either directly or indirectly through ‘a directory, advertisement, or site 8 on the Internet,’ and (3) the ‘unsolicited advertisement contains’ an opt-out notice meeting certain 9 statutory requirements.” Id. at 926 (quoting 47 U.S.C. 227(b)(1)(C)(i)-(iii)). An established 10 business relationship and prior consent are affirmative defenses, which the defendant typically has 11 the obligation of proving, but at the class certification stage, Sapan must show that individual 12 issues will not predominate in this case. Id. at 931-32. Evidence of a consent defense in a TCPA 13 class action will weigh against predominance. Id. 14 That is the situation here. Yelp has identified evidence in the record showing an 15 established business relationship with, or other prior consent provided by, putative class members. 16 For example, the record establishes that some putative class members provided their residential 17 phone numbers to Yelp sales employees via e-mail, and some small businesses and sole 18 proprietors gave their cell phone numbers or residential phone numbers to Yelp sales employees, 19 so they could be reached directly at a convenient time. Dkt. No. 78 ¶ 5; Dkt. No. 78-1. Business 20 owners also listed their residential phone numbers when they registered their business pages. Dkt. 21 No. 76 ¶ 8. 22 As a result, determining whether an individual is within the proposed class would 23 necessarily entail an individualized analysis of his or her communications with Yelp. To illustrate 24 what this would require, Yelp explained the steps that were necessary to determine whether it had 25 an established business relationship with Sapan. Dkt. No. 75 at 12. Yelp reviewed Sapan’s “log- 26 in history, email and notification subscription preferences, registration and agreement to Yelp’s 27 Terms of Service, Yelp profile page, history of authoring reviews on Yelp,” and several other 1 comments to determine whether any created an established business relationship. Id. Yelp states 2 that it would need to do all of these steps for each putative class member. See, e.g., Dkt. No. 78-1. 3 Sapan did not challenge any of this. To the contrary, he forthrightly acknowledged that 4 individuals who consented to calls via an established business relationship with Yelp could not be 5 class members. See Dkt. No. 122 at 6. The crucial shortfall for Sapan’s certification request is 6 that he has not proposed a method of identifying class members without individualized inquiries 7 along the lines Yelp described. Even assuming that Verkhovskaya could identify phone numbers 8 on the Do Not Call Registry that were called twice by Yelp in a twelve month period -- a 9 possibility that is not at all clear in her declaration -- her method does not account for people who 10 consented to being called in the ways that the record documents. 11 Consequently, Sapan has not demonstrated a level of predominance and commonality that 12 would make this case is suitable for adjudication on a class wide basis. Certification of the 13 proposed class is denied. 14 II. MOTION TO STRIKE AND OTHER OBJECTIONS 15 In response to the filing of the Verkhovskaya declaration with Sapan’s amended motion for 16 class certification, Yelp moved to strike the declaration primarily on the grounds that the 17 declaration was untimely, unjustified, and prejudicial. Dkt. No. 129 at 2-3. But when the Court 18 directed Sapan to file an amended motion for certification, the possibility of a new declaration was 19 raised, and the Court in effect allowed it. Dkt. No. 121 at 31:14-16. Yelp was afforded an 20 opportunity to depose Verkhovskaya, but chose not to do so. Dkt. No. 144; see also Dkt. No. 137 21 at 4. Consequently, the Court declines to strike the declaration, although as discussed, it has 22 carried no force in Sapan’s favor. 23 Sapan has objected to Yelp’s supplemental briefing in opposition of the motion for class 24 certification. See Dkt. No. 136. He specifically objects to new declarations and exhibits and new 25 documents that were not disclosed during discovery. Id. at 2-3. The Court has not relied on any 26 of this additional evidence, and the objections are denied as moot. 27 1 CONCLUSION 2 The stay is lifted and the case is administratively re-opened. The case will proceed solely 3 on the basis of Sapan’s individual claim. The parties are directed to file by December 6, 2021, a 4 || joint proposal for a trial date in the second quarter of 2022. 5 IT IS SO ORDERED. 6 Dated: November 15, 2021 7 8 JAMES MONATO 9 United Ptates District Judge 10 11 12

15 16

= 17

Z 18 19 20 21 22 23 24 25 26 27 28