Menagerie Productions v. IAC/InteractiveCorp, No. 2:2008cv04263 - Document 108 (C.D. Cal. 2009)

Court Description: ORDER GRANTING IN PART AND DENYING IN PART PLAINTIFFS' MOTION FOR CLASS CERTIFICATION 73 by Judge Christina A. Snyder: The Court GRANTS plaintiffs' motion for national class certification with regard to their breach of contract claims, an d their claim under the "fraudulent" prong of the UCL. The Court DENIES class certification to plaintiffs with regard to their claim under the "unfairness" prong of the UCL. The national class is as follows: All persons or entitie s in the United States who entered into form contracts for pay-per-click advertising through Citysearch.com, paid money for this advertising service, and experienced click fraud by reason of double clicks or Citysearch's failure to apply automatic filters to traffic from its syndication partners up through 3/23/2007. (gk)

Download PDF

Menagerie Productions v. IAC/InteractiveCorp Doc. 108 1 2 3 4 5 6 7 8 UNITED STATES DISTRICT COURT 9 CENTRAL DISTRICT OF CALIFORNIA 10 WESTERN DIVISION 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 MENAGERIE PRODUCTIONS, et al., ) ) ) Plaintiff(s), ) ) vs. ) ) CITYSEARCH, et al., ) ) ) Defendant(s). ) ) ) ) _________________________________ ) Case No. CV 08-4263 CAS (FMO) ORDER GRANTING IN PART AND DENYING IN PART PLAINTIFFS’ MOTION FOR CLASS CERTIFICATION (filed 08/11/09) INTRODUCTION On May 27, 2008, plaintiff Tom Lambotte (“Lambotte”), on behalf of himself and all others similarly situated, filed the instant class action suit in the Los Angeles County Superior Court against defendants IAC/Interactive Corp., Ticketmaster, d/b/a Citysearch.com, Citysearch.com (collectively, “Citysearch”), and Does 1-20. On June 27, 2008, Citysearch removed the action to this Court. The class action complaint asserted claims for breach of contract, violation of California’s Unfair Competition Law (“UCL”), Cal. Bus. & Prof. Code §§ 17200 et seq., and negligence. Dockets.Justia.com 1 On July 7, 2008, Citysearch filed a motion for summary judgment on Lambotte’s 2 first and second claims for relief. On July 31, 2008, the Court granted Citysearch’s 3 motion for summary judgment on Lambotte’s claims for breach of contract and 4 injunctive relief under the UCL.1 The Court granted Lambotte 30 days leave to add 5 former customers to the complaint who were not subject to the defense raised by 6 Citysearch in its motion for summary judgment. 7 On September 3, 2008, Lambotte filed a first amended complaint, in which two 8 additional named plaintiffs were included: Chad Bordeaux (“Bordeaux”) and Sarah 9 Bloch (“Bloch”). In the first amended complaint, plaintiffs asserted claims for breach of 10 contract and violation of California’s Unfair Competition Law (“UCL”), Cal. Bus. & 11 Prof. Code §§ 17200 et seq. 12 Citysearch filed a motion to dismiss plaintiffs’ breach of contract claim on 13 September 29, 2008. On November 4, 2008, the Court denied Citysearch’s motion to 14 dismiss, except insofar as it sought dismissal of Lambotte as a plaintiff on the breach of 15 contract claim. 16 On July 23, 2009, plaintiffs filed a second amended complaint (“SAC”), in which 17 Menagerie Productions (“Menagerie”) and Redwolf, LLC (“Redwolf”) replaced 18 Bordeaux, Bloch, and Lambotte as plaintiffs. In the SAC, plaintiffs again assert claims 19 for breach of contract and violation of California’s Unfair Competition Law (“UCL”), 20 Cal. Bus. & Prof. Code §§ 17200 et seq. The gravamen of the SAC was that plaintiffs 21 entered into a contract with Citysearch to place “pay-per-click” advertisements on the 22 Citysearch website, and that Citysearch failed to detect and prevent “click fraud.”2 23 24 25 26 27 28 1 The Court granted summary judgment for Citysearch on the breach of contract claim because, at the time of the motion, Citysearch had refunded all of the advertising fees that Lambotte had paid to Citysearch, and, under the express terms of the contract at issue, Lambotte was only entitled to recover fees actually paid to Citysearch. 2 The SAC alleges that the term “click fraud” is understood in the industry “to (continued...) 2 On August 11, 2009, plaintiffs filed the present motion for class certification. 1 2 Citysearch filed its opposition thereto on September 10, 2009. Plaintiffs filed their reply 3 on September 24, 2009. After carefully considering the arguments set forth by the 4 parties, the Court finds and concludes as follows. 5 II. 6 BACKGROUND Plaintiffs allege that Citysearch provides, among other things, online advertising 7 services, which generate most of Citysearch’s revenue. SAC ¶ 14. Such advertising 8 services may be purchased for a flat monthly fee or on a pay-per-click basis. SAC ¶ 15. 9 If a customer chooses the pay-per-click advertising option, Citysearch charges the 10 11 customer only when a user clicks on the customer’s advertisement. SAC ¶ 16. Plaintiffs contend that Citysearch began offering pay-per-click advertising in mid 12 to late 2004. Mot. at 3: Kabateck Decl., Exhib. 7. Plaintiffs assert that in late 2005, 13 Citysearch claimed internally that “[o]ur industry leading traffic quality systems are 14 constantly analyzing user behavior across our site and our partner network to detect 15 unusual and fraudulent click behavior. Attempts to artificially drive up an advertiser’s 16 clicks, whether manually or via robots or other deceptive tools, will be detected by our 17 systems and automatically thrown out.”3 Mot. at 4: Kabateck Decl., Exhib. 9. Plaintiffs 18 19 20 21 22 23 24 25 2 (...continued) describe clicks on a search advertisement with no intention of doing business with the advertiser and for some purpose other than that contemplated by the ad.” SAC ¶ 24. Citysearch argues that plaintiffs cannot proceed with their claims because, as an initial matter, plaintiffs failed to include “double clicks” or “a failure to apply automatic filters to traffic from syndication partners” in their definition of “click fraud” in the SAC. Opp’n at 1. The Court hereby deems the SAC to include “double clicks” and “a failure to apply automatic filters to traffic from syndication partners” in plaintiffs’ definition of “click fraud.” Thus, notwithstanding plaintiffs’ offer to amend the complaint, they need not do so. 26 3 27 28 Defendants contend that Exhibit 9 was for internal use only and would have been viewed in an ad hoc manner by Citysearch employees, but that it was never shown to (continued...) 3 1 further assert that by May 2006, Citysearch was giving its customers a “Fraud 2 Prevention Guarantee.”4 Mot. at 4-5: Kabateck Decl., Exhib. 11. Plaintiffs assert that 3 Citysearch did not update its invalid click filters until 2007. Mot. at 4: Kabatech Decl., 4 Exhib. 7. 5 6 Plaintiffs allege that in January 2008, Menagerie entered into a written form contract for the placement of pay-per-click advertising on Citysearch.com.5 SAC ¶ 33. 7 8 9 3 (...continued) advertisers. Asher Decl., ¶ 15. Accordingly, the Court does not rely on it. 4 10 11 12 It appears that the Fraud Prevention Guarantee relied upon by plaintiffs was disseminated for internal use and training. Accordingly, the Court does not rely on the Fraud Prevention Guarantee. The parties have filed numerous additional evidentiary objections. In so far as the Court does not rely on the evidence that is the subject of these objections, the Court overrules them as moot. 13 5 14 15 The Agreement describes the pay-per-click performance package as follows: 22 The Performance Package is Citysearch’s pay for performance advertising program where businesses set a monthly advertising budget (the “CAP”) and pay for Click-Throughs or Program Calls (defined below) up to the CAP. On a monthly basis, Business shall pay to Citysearch a monthly listing fee and the cost per click and/or cost per call for each Click-Through and/or for each Program Call, up to the CAP set forth on the Enrollment Form. “Clickthroughs” are defined as (i) clicks on Business’ advertising located on the Citysearch website or a Citysearch distribution partner website, which directs users to Business’ website or Business’ profile page on the Citysearch website, (ii) clicks on certain content links located on Business’ Citysearch profile page. 23 SAC, Ex. A (plaintiffs’ contract) ¶ 3a. The Agreement includes the following disclaimer: 16 17 18 19 20 21 24 25 26 27 28 Business acknowledges and agrees that Citysearch’s services are provided to business on an “as is” basis, and Citysearch disclaims any and all express or implied warranties . . . . Furthermore, to the fullest extent permitted by law, Citysearch disclaims all warranties and guarantees regarding . . . the quality or timing of click-throughs, click-through rates, (continued...) 4 1 2 3 4 5 6 7 8 5 (...continued) conversions or other performance or results for any advertising. SAC, Ex. A (plaintiffs’ contract) ¶ 8. Furthermore, the Agreement states: The terms of this Agreement and the Enrollment form may be changed by Citysearch from time to time, and any notices hereunder shall be made, by providing you with email or written notice, or by posting any such changes on the Citysearch website, and you agree to be bound by any changes . . . . ” 9 10 11 12 13 14 15 16 17 18 SAC, Ex. A (plaintiffs’ contract) ¶ 8. Plaintiffs allege that Citysearch’s “About Us” page states the following to potential customers: “We connect you to more customers. You only pay for results. Advertise on Citysearch today and only pay for clicks to your Website or business profile page.” SAC ¶ 21; SAC Exhib. B. Plaintiffs further allege that the “FAQ” page that Citysearch makes available to potential customers during the sign-up process states the following: Q: How do I know that clicks to my website are legitimate? A: Citysearch proactively researches and develops processes, policies, and technologies to identify invalid click activity with regard to our customers’ advertising. Citysearch employs advanced security filters and blocks out clicks from spiders and robots. 19 20 21 SAC ¶ 23; SAC Exhib. C. Plaintiffs allege that Citysearch devotes an entire webpage to explain its “Invalid Click Policy” to potential customers, where it states in part: 22 23 24 25 26 27 As a leader in Local Pay for Performance advertising one of Citysearch’s key concerns is detection of invalid clicks. Citysearch proactively researches and develops processes, policies, and technologies to identify invalid click activity with regard to our customers’ advertising. We are committed to protecting our customers’ investment in Local Pay for Performance Advertising with us. 28 (continued...) 5 1 Plaintiffs allege that after having received no new clients after the first billing cycle, 2 Menagerie requested a refund, but that Citysearch refused to refund the majority of the 3 fees. SAC ¶ 34. Plaintiffs allege that over a three month period, Menagerie paid $1,900 4 in fees without receiving any new customers. SAC ¶ 35. 5 Plaintiffs further allege that in March 2008, Redwolf entered into the same 6 written form contract for the placement of pay-per-click advertising on Citysearch.com. 7 SAC ¶ 37. Plaintiffs allege that shortly thereafter, Redwolf suspected click fraud and 8 requested that the charges it incurred be reversed, and that Citysearch refused. SAC ¶ 9 38. Plaintiffs allege that Redwolf cancelled its account in August 2008, at the end of its 10 11 contractual obligation, after paying over $700 in fees to Citysearch. SAC ¶ 39. The gravamen of plaintiffs’ lawsuit is two-fold. First, plaintiffs allege that they 12 entered into a written form contract (the “Agreement”) for the placement of pay-per- 13 click advertising on Citysearch.com, and that the Agreement contained an implied 14 covenant of good faith and fair dealing “that Citysearch would not do anything that 15 would have the effect of injuring the rights of Plaintiffs and the Class to receive the 16 benefits of the contract.” SAC ¶ 52. Plaintiffs allege that Citysearch breached both the 17 express terms of the Agreement and its covenant of good faith and fair dealing “by 18 19 20 21 22 23 5 (...continued) Citysearch actively monitors and analyzes clicks to our customers’ advertising to try to detect any abuse of Our Local Pay for Performance advertising program. Citysearch attempts to prevent clicks that are caused by automated robots or spiders, clicks generated from within the Citysearh organization, or clicks which show a pattern of fraudulent abuse . . . . 24 25 26 27 28 While we endeavor to charge customers solely for valid clicks, we cannot ensure that all invalid clicks will be detected. Therefore, please contact Citysearch Customer Service . . . if you have any questions or concerns about clicks charged to your account, and a customer service specialist will investigate your account. SAC ¶ 23; SAC Exhib. D. 6 1 collecting fees from plaintiffs and the Class for click fraud even though Citysearch knew, 2 or should have reasonably known, that the clicks were not ‘actual clicks’ but rather 3 purposeful clicks made for an improper purpose. Citysearch further breached its contract 4 with plaintiffs and the Class by failing to implement effective oversight, investigating 5 oversight and prevention of click fraud.” SAC ¶ 53. 6 Second, plaintiffs allege that Citysearch violated the unfair, unlawful, and 7 fraudulent prongs of the UCL. Plaintiffs allege that Citysearch’s business practices are 8 unfair because Citysearch “(a) fails to employ any method to track fraudulent clicks, 9 including clicks originating from its own employees and/or agent and clicks originating 10 from Citysearch’s “partner sites”; (b) fails to inform its customers that it does not 11 employ a method to track fraudulent clicks, including clicks originating from its own; 12 and (c) charges customers for invalid clicks.” SAC ¶ 58. Plaintiffs allege that 13 Citysearch’s business practices are unlawful “because the conduct constitutes a breach of 14 the Agreement.” SAC ¶ 59. Plaintiffs allege that Citysearch’s business practices are 15 fraudulent “because they are likely to deceive its customers into believing that they will 16 not be charged for ‘invalid’ clicks, when in fact, Citysearch routinely charges its 17 customers for clicks that it knows, or by the exercise of reasonable care, should know are 18 not clicks that originate from potential customers who actively and legitimately chose 19 the advertiser’s link.” SAC ¶ 60. By this motion, plaintiffs seek to certify the following nationwide class: “All 20 21 persons or entities in the United States who paid money for pay-per-click advertising 22 through Citysearch.com.” Mot. at 2. 23 III. 24 LEGAL STANDARD “Class actions have two primary purposes: (1) to accomplish judicial economy by 25 avoiding multiple suits, and (2) to protect rights of persons who might not be able to 26 present claims on an individual basis.” Haley v. Medtronic, Inc., 169 F.R.D. 643, 647 27 (C.D. Cal. 1996) (citing Crown, Cork & Seal Co. v. Parker, 462 U.S. 345 (1983)). 28 Federal Rule of Civil Procedure 23 governs class actions. A class action “may be 7 1 certified if the trial court is satisfied after a rigorous analysis, that the prerequisites of 2 Rule 23(a) have been satisfied.” General Tel. Co. of the Southwest v. Falcon, 457 U.S. 3 147, 161 (1982). 4 To certify a class action, plaintiffs must set forth prima facie facts that support the 5 four requirements of Rule 23(a): (1) numerosity; (2) commonality; (3) typicality; and (4) 6 adequacy of representation. Dunleavy v. Nadler (In re Mego Fir. Corp. Sec. Litig.), 213 7 F.3d 454, 462 (9th Cir. 2000) (internal quotations omitted). These requirements 8 effectively “limit the class claims to those fairly encompassed by the named plaintiff's 9 claims.” Falcon, 457 U.S. at 155 (quoting Califano v. Yamasaki, 442, U.S. 682, 701 10 11 (1979)). If the district court finds that the action meets the prerequisites of Rule 23(a), the 12 court must then consider whether the class is maintainable under one or more of the 13 three alternatives set forth in Rule 23(b). Plaintiffs seek certification under Rule 14 23(b)(3). A class is maintainable under Rule 23(b)(3) where “questions of law or fact 15 common to the members of the class predominate over any questions affecting only 16 individual members,” and where “a class action is superior to other available methods 17 for fair and efficient adjudication of the controversy.” Fed. R. Civ. P. 23(b)(3) 18 (emphasis added). “The Rule 23(b)(3) predominance inquiry tests whether the proposed 19 classes are sufficiently cohesive to warrant adjudication by representation.” Hanlon v. 20 Chrysler Corp., 150 F.3d 1011, 1022 (9th Cir. 1998) (citing Amchem Products, Inc. v. 21 Windsor, 521 U.S. 591 (1997)). The predominance inquiry measures the relative weight 22 of the common to individualized claims. Id. “Implicit in the satisfaction of the 23 predominance test is the notion that the adjudication of common issues will help achieve 24 judicial economy.” Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1189 (9th Cir. 25 2001) (citing Valentino v. Carter-Wallace, Inc., 97 F.3d 1227, 1234 (9th Cir. 1996)). In 26 determining superiority, the court must consider the four factors of Rule 23(b)(3): (1) the 27 interests members in the class have in individually controlling the prosecution or defense 28 of the separate actions; (2) the extent and nature of any litigations concerning the 8 1 controversy already commenced by or against members of the class; (3) the desirability 2 or undesirability of concentrating the litigation of the claims in the particular forum; and 3 (4) the difficulties likely encountered in the management of a class action. Id. at 1190- 4 1993. “If the main issues in a case require the separate adjudication of each class 5 member's individual claim or defense, a Rule 23(b)(3) action would be inappropriate.” 6 Id. (citing 7A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal 7 Practice and Procedure § 1778 at 535-39 (2d. ed. 1986) (hereinafter “Wright, Miller & 8 Kane”)). 9 IV. DISCUSSION 10 A. RULE 23(A) REQUIREMENTS 11 1. NUMEROSITY 12 Rule 23(a)(1) requires the members of a proposed class to be so numerous that 13 joinder of all of the class members would be impracticable. Fed. R. Civ. P. 23(a). 14 However, “[i]mpracticability does not mean ‘impossibility,’ but only the difficulty or 15 inconvenience in joining all members of the class.” Harris v. Palm Springs Alpine 16 Estates, Inc., 329 F.2d 909, 913-14 (9th Cir. 1964) (quoting Advertising Specialty Nat. 17 Ass’n v. FTC, 238 F.2d 108, 119 (1st Cir. 1956)). 18 Plaintiffs argue that the numerosity requirement is met. Plaintiffs contend that 19 while the exact size of the class is not known, “it clearly numbers in the thousands.” 20 Mot. at 8. Plaintiffs assert that of Citysearch’s 15,000 active customers, at least 10,000 21 pay for pay-per-click advertising. Id. at 8-9. Plaintiffs further argue that this number 22 does not include the thousands of class members who, like plaintiffs, were previously 23 pay-per-click customers but have since canceled their advertising accounts with 24 Citysearch. Id. at 9 (citing Stuart v. Radioshack Corp., 2009 WL 281941, at *5 (N.D. 25 Cal. Feb. 5, 2009) (“courts have found that numerosity is satisfied when class size 26 exceeds 40 members”)). 27 28 Citysearch does not dispute that plaintiffs’ proposed class is sufficiently numerous to satisfy Rule 23(a)(1). Joinder is impracticable. As such, there is no dispute that the 9 1 numerosity requirement of Rule 23(a)(1) is therefore met. 2 2. COMMONALITY 3 Commonality requires “questions of law or fact common to the class.” Fed. R. 4 Civ. P. 23(a)(2). The commonality requirement is generally construed liberally; the 5 existence of only a few common legal and factual issues may satisfy the requirement. 6 Jordan v. County of L.A., 669 F.2d 1311, 1320 (9th Cir. 1982). 7 Plaintiffs argue that the commonality requirement is met, because the adjudication 8 of the same core questions of fact and law will resolve all of the class members’ claims. 9 Mot. at 9. Plaintiffs assert that for the breach of contract claim, the issues to be resolved 10 are: “(1) whether Citysearch’s failure to implement objectively reasonable invalid click 11 prevention measures amounts to a breach of the express terms of its contract with 12 plaintiffs and the class, and (2) whether Citysearch’s discretionary decision to charge 13 plaintiffs and the class for objectively unreasonable clicks constitutes a breach of the 14 covenant of good faith and fair dealing contained in its contract with plaintiffs and the 15 class.” Id. at 9-10. Plaintiffs’ UCL claim hinges on a single question: “whether 16 Citysearch acted fraudulently or unfairly.” Id. at 10. 17 Citysearch does not dispute that there exist questions of law or fact common to the 18 class. See Jordan, 669 F.2d at 1320. The court agrees that the commonality requirement 19 of Rule 23(a)(3) is therefore met. 20 3. TYPICALITY 21 Typicality requires a determination of whether the named plaintiff’s claims are 22 typical of those of the proposed class that they seek to represent. Fed. R. Civ. P. 23 23(a)(3). “[R]epresentative claims are ‘typical’ if they are reasonably co-extensive with 24 those of absent class members; they need not be substantially identical.” Hanlon, 150 25 F.3d at 1020; Schwartz v. Harp, 108 F.R.D. 279, 282 (C.D. Cal. 1985) (“A plaintiff’s 26 claim meets this requirement if it arises from the same event or course of conduct that 27 gives rise to claims of other class members and the claims are based on the same legal 28 theory”). 10 1 Plaintiffs argue that their claims are “reasonably coextensive with those of the 2 absent class members,” because like every other proposed class member, they advertised 3 using Citysearch’s pay-per-click advertising program and they were charged for clicks 4 that were allegedly “clearly and objectively invalid.” Mot. at 11. Plaintiffs further assert 5 that their consumer protection claims are reasonably coextensive with those of absent 6 class members, because “they arise from Citysearch’s common scheme of failing to 7 apply reasonable methods for filtering out clearly invalid clicks as well as Citysearch’s 8 deceptive statements regarding the existence and quality of its click filters.” Id. 9 Plaintiffs contend that as victims of Citysearch’s common course of conduct, plaintiffs’ 10 and the class members’ injuries are identical, as are the claims that arise therefrom. Id. 11 (citing Schaefer v. Overland Express Family of Funds, 169 F.R.D. 124, 128-29 (S.D. 12 Cal. 1996) (“[C]ases challenging the same unlawful conduct which affects both the 13 named plaintiffs and the putative class usually satisfy the typicality requirement 14 irrespective of the varying fact patterns underlying the individual claims.”); In re First 15 Alliance Mortgage Co., 471 F.3d 977, 992 (9th Cir. 2006) (holding that where a 16 “centrally-orchestrated scheme to mislead” is alleged, it is the scheme and not the 17 precise details of any individual’s experience that forms the nucleus of the class 18 claims)). 19 Citysearch responds that here, because “the substantive claims depend on 20 individual permutations . . . the claims of the named plaintiffs who have the same 21 general complaint against the defendant as the class are not typical.” Opp’n at 15 22 (quoting Gartin v. S&M Nutec LLC, 245 F.R.D. 429, 434 (C.D. Cal. 2007)). Citysearch 23 argues that plaintiffs have not identified any actionable injury they have suffered, much 24 less a similar one by the rest of the class. Opp’n at 15. Citysearch asserts that plaintiffs 25 have not offered any evidence that they were charged for invalid clicks, but have 26 “merely advanced a theory that they and other class members want to explore whether 27 they have been charged for some as-yet-unidentified categories of ‘objectively invalid’ 28 11 1 clicks.”6 Opp’n at 16. Thus, Citysearch argues that “there is nothing before the Court 2 that would permit it even to speculate, much less conclude, that these plaintiffs paid for 3 (and thus were injured by) objectively invalid clicks.” Id. Citysearch further argues that 4 plaintiffs’ expert analyzed its click logs and did not find a single click that appeared to 5 be fraudulent. Id. (citing Jansen Depo. at 46: 5-8). Citysearch contends that plaintiffs’ 6 depositions show that their real grievance is not that they were charged for invalid clicks, 7 but that they received a poor return on their investment—a wholly different injury that 8 defeats typicality. Id. (citing Cen. Tel. Co. of Southwest v. Falcon, 457 U.S. 147, 157- 9 58 (1982); Akkerman v. Mecta Corp., Inc., 152 Cal. App. 4th 1094, 1101 (2007)). 10 Citysearch contends that even if plaintiffs could show actual injury, they still 11 cannot establish typicality because there is no evidence they read and relied upon a 12 misrepresentation made to all class members. Opp’n at 17. Citysearch asserts that 13 plaintiff Menagerie relied solely upon verbal representations and read no documents, 14 while plaintiff Redwolf relied on a combination of verbal and written representations, 15 but did not read the Terms, the Policy, or the FAQ page on Citysearch’s website. Id. 16 Furthermore, Citysearch asserts that neither defendant had heard about any purported 17 “fraud guarantee.” Id. Yet, Citysearch asserts that according to the SAC and plaintiffs’ 18 motion, other class members read and heard about Citysearch’s representations. Id. 19 Thus, Citysearch contends that because “the class, as defined by the plaintiffs, 20 potentially includes class members who received [representations] other than those 21 received by the named plaintiffs, the class definition is broad and overinclusive.” Id. 22 (quoting Scott v. Mascott, 1999 WL 33117204, *1 (W.D. Wash. Apr. 13, 1999); see also 23 Akkerman, 152 Cal. App. 4th at 1100-01)). 24 25 Plaintiffs reply that they have provided evidence that both named plaintiffs were harmed by paying for invalid clicks, and thus assert that they have standing. Reply at 2. 26 27 28 6 Citysearch argues that this new theory is not set forth in the SAC, which alleges that Citysearch failed to take any steps to combat click fraud. Opp’n at 16. 12 1 Plaintiffs argue that their expert, Dr. Jansen, identified clicks in plaintiffs’ click-logs that 2 appeared to be invalid. Id. However, plaintiffs argue that because Citysearch had not 3 produced information regarding the clicks for which plaintiffs have been charged prior 4 to Dr. Jansen’s completing his expert report, Dr. Jansen was unable to determine 5 whether plaintiffs have been charged for invalid clicks. Id. Plaintiffs assert that 6 subsequently, Citysearch produced this information, and thus they can now demonstrate 7 that they were improperly charged for objectively invalid clicks. Id. Specifically, 8 plaintiffs contend that they were both charged for two clicks on the same ad, made from 9 the same source, that were less than a second apart.7 Id. 10 Plaintiffs further reply that Citysearch’s assertion that plaintiffs are not typical 11 because “there is no evidence that plaintiffs read and relied upon a misrepresentation 12 made to all class members” completely ignores the fact that plaintiffs’ claims regarding 13 Citysearch’s deceptive practices are made pursuant to California’s Unfair Competition 14 Law (“UCL”), and not common law fraud. Id. at 5. Plaintiffs assert that “relief under 15 the UCL is available without individualized proof of decision, reliance and injury,” but 16 instead plaintiffs need only show that “members of the public are likely to be deceived.” 17 Id. (quoting In re Tobacco II Cases, 46 Ca. 4th 298, 312 (2009)). Plaintiffs argue that 18 they, like other class members, were “never informed that they would be charged for 19 invalid clicks because Citysearch did not take adequate steps to prevent invalid clicks.” 20 Id. at 6. 21 Moreover, plaintiffs reply that Citysearch’s argument that typicality is lacking 22 23 24 25 26 27 28 7 For example, plaintiffs argue that on June 19, 2008, plaintiff Redwolf received two clicks to its advertisement on 18:31:39, which both came from the same Citysearch syndication partner, “synd_abc.” Reply at 3 (citing Torrijos Decl., Exhib. 33). They were the only two clicks Redwolf received from that syndication partner that day. Id. (citing Torrijos Decl., Exhib. 33). On June 19, 2008, Citysearch charged plaintiff for thirteen clicks, including two from “synd_abc.” Id. (citing Torrijos Decl., Exhib. 33). Thus, Citysearch charged Redwolf for both clicks even though they both occurred within the same second. Id. (citing Torrijos Decl., Exhib. 33). 13 1 because of the differences in the manner in which plaintiffs Menagerie and Redwolf 2 enrolled in Citysearch’s advertising services fails because plaintiffs’ claims are based on 3 an alleged common course of conduct by Citysearch. Id. (citing Stern v. AT&T 4 Mobility Corp., 2008 WL 4382796, *7 (C.D. Cal. 2008)). Plaintiffs argue that here, as 5 in Stern, plaintiffs’ claims are based on Citysearch’s material omission to its potential 6 customers that it fails to detect and prevent invalid clicks and its conduct of charging its 7 advertisers for invalid clicks. Id. 8 The Court finds Citysearch’s arguments unpersuasive. First, plaintiffs have 9 presented evidence that they suffered an actionable injury, because they both presented 10 evidence that Citysearch charged them for double-clicks made within the same second 11 from the same source. See Torrijos Decl., Exhibs. 30-33. Indeed, Citysearch’s expert 12 acknowledges that such doubleclicks are patently “invalid,” and that customers should 13 not be charged for them. See Kabateck Decl., Exhib. 28. Second, like every other 14 proposed class member, plaintiffs advertised using Citysearch’s pay-per-click 15 advertising program and were allegedly charged by Citysearch for invalid clicks. 16 Notwithstanding any asserted differences between class members, plaintiffs’ claims are 17 based on an alleged common course of conduct by [Citysearch] to (1) charge its 18 advertisers for invalid clicks, and (2) make material omissions regarding the existence 19 and quality of its click filters. Accordingly, plaintiffs’ claims arise from the “same event 20 or course of conduct” as those of the various absent class members, as required under 21 Rule 23(a)(3), and are typical of the claims of the proposed class. Schwartz, 108 F.R.D. 22 at 282; Rosario v. Livaditis, 963 F.2d 1013, 1018 (7th Cir. 1992). 23 4. ADEQUACY OF REPRESENTATION 24 The adequacy of representation requirement of Rule 23(a)(4) involves a two-part 25 inquiry: “(1) do the named plaintiffs and their counsel have any conflicts of interest 26 with other class members and (2) will the named plaintiffs and their counsel prosecute 27 the action vigorously on behalf of the class?” Hanlon, 150 F.3d at 1020. 28 Plaintiffs argue that the plaintiffs’ interests here are fully aligned with those of the 14 1 class because the basis of their claims and the nature of their injury are exactly the same 2 as those of the class members they seek to represent. Mot. at 12. Plaintiffs further argue 3 that they understand and are prepared to fulfill their duties to the class. Id. (citing Decl. 4 of S. Bloch; Decl. of C. Bordeaux). Plaintiffs contend that the firms representing them 5 are experienced class action counsel, having successfully conducted numerous consumer 6 class actions. Id. Plaintiffs further contend that counsel remains committed to devoting 7 all necessary resources to prosecuting this matter and possesses the resources necessary 8 to represent the proposed class. Id. 9 Citysearch responds that plaintiffs are not adequate class representatives because 10 they lack standing to pursue all of the claims for relief set forth in their SAC. Opp’n at 11 18. Citysearch repeats its argument that plaintiffs have failed to present evidence that 12 they paid for invalid clicks, and thus argue that plaintiffs cannot serve as class 13 representatives. Id. Citysearch further argues that because plaintiffs are not currently 14 Citysearch advertisers, they cannot seek injunctive relief under the UCL. Id. 15 The Court concludes that the adequacy of representation prong is met in this case. 16 As discussed supra, plaintiffs have provided evidence that they paid for invalid clicks, 17 and thus have standing to serve as class representatives. While plaintiffs do not have 18 standing to request prospective injunctive relief under the UCL, see Heffelfinger v. Elec. 19 Data Sys. Corp., 2008 U.S. Dist. LEXIS 5296 (C.D. Cal. Jan. 7, 2008) (“because 20 plaintiffs are no longer employed by EDS, they lack standing to seek prospective 21 injunctive relief under the UCL.”), they have standing to pursue restitutionary relief 22 under the UCL because they paid to advertise with Citysearch. See e.g., So. Cal. 23 Housing Rights Center v. Los Feliz Towers, 426 F. Supp. 2d 1061, 1069 (C.D. Cal. 24 2005) (stating that the plaintiff “has standing because it presents evidence of actual 25 injury based on loss of financial resources in investigating this claim and diversion of 26 staff time from other cases to investigate the allegations here”). Therefore, the Court 27 finds that the plaintiffs have the same incentive as current advertisers with Citysearch to 28 prosecute the UCL claim vigorously for the class. See Hanlon, 150 F.3d at 1020. 15 1 Furthermore, while plaintiffs may not have advertised with Citysearch during the time 2 that Citysearch failed to run its syndication traffic through all of its automatic filters,8 the 3 Court finds that plaintiffs have the same incentive as those advertisers that did advertise 4 with Citysearch during that time to prosecute such claims for the class, because they 5 were allegedly harmed by other types of invalid clicks. 6 B. RULE 23(B)(3) REQUIREMENTS 7 Certification under Rule 23(b)(3) is proper “whenever the actual interests of the 8 parties can be served best by settling their differences in a single action.” Hanlon, 150 9 F.3d at 1022 (internal quotations omitted). As noted above, Rule 23(b)(3) calls for two 10 separate inquiries: (1) do issues common to the class “predominate” over issues unique 11 to individual class members, and (2) is the proposed class action “superior” to other 12 methods available for adjudicating the controversy. Fed. R. Civ. P. 23(b)(3). The latter 13 requirement requires consideration of the difficulties likely to be encountered in the 14 management of this litigation as a class action, including, especially, whether and how 15 the case may be tried. In making these determinations, the Court does not decide the 16 merits of any claims or defenses, or whether the plaintiffs are likely to prevail on their 17 claims. Rather, the Court must determine whether plaintiffs have shown that there are 18 plausible class-wide methods of proof available to prove their claims. 19 1. 20 “Implicit in the satisfaction of the predominance test is the notion that the 21 adjudication of common issues will help achieve judicial economy.” See Valentino, 97 22 F.3d at 1234. Thus, the Court must determine whether common issues constitute such a 23 significant aspect of the action that “there is a clear justification for handling the dispute 24 25 26 27 PREDOMINANCE AND COMMONALITY 8 Plaintiffs rely on an internal Citysearch document dated March 23, 2007 for the proposition that Citysearch did not apply all of its automatic filters to its syndication traffic until sometime in 2007 or later. Kabatech Decl., Exhib. 10. Plaintiff Menagerie advertised with Citysearch from December 27, 2007 to March 26, 2008, and plaintiff Redwolf advertised with Citysearch from March 27, 2008 to July 21, 2008. Asher Decl., ¶ 12 n.3. 28 16 1 on a representative rather than on an individual basis.” 7A Wright, Miller, & Kane § 2 1778 (3d ed. 2005). For the proponent to satisfy the predominance inquiry, it is not 3 enough to establish that common questions of law or fact exist, as it is under Rule 4 23(a)(2)’s commonality requirement—the predominance inquiry under Rule 23(b) is 5 more rigorous. Amchem Prods., Inc., 521 U.S. at 624. The predominance question 6 “tests whether proposed classes are sufficiently cohesive to warrant adjudication by 7 representation.” Id. at 623. The Court, therefore, must balance concerns regarding the 8 litigation of issues common to the class as a whole with questions affecting individual 9 class members. Abed v. A. H. Robins Co. (In re Northern District of California, Dalkon 10 Shield IUD Products Liability Litigation), 693 F.2d 847, 856 (9th Cir. 1982). a. 11 Breach of Contract Claims i. 12 Breach of Contract 13 Plaintiffs allege that Citysearch breached the express terms of the contract 14 between Citysearch and the class, and also breached the covenant of good faith and fair 15 dealing. Mot. at 14. With regard to each of these claims, plaintiffs assert that common 16 issues of law and fact predominate. Id. Plaintiffs argue that predominance is readily 17 satisfied in cases such as this one which involve form agreements with the class and the 18 Citysearch’s standardized conduct.9 Id. (citing Kleiner v. First Nat’l Bank of Atlanta, 97 19 F.R.D. 683, 691 (N.D. Ca. 1983) (“When viewed in light of Rule 23, claims arising from 20 interpretations of a form contract appear to present the classic case for treatment as a 21 class action . . . . ”); Heartland Commc’ns, Inc. v. Spring Corp., 161 F.R.D. 111 (D. Kan. 22 23 24 25 26 27 28 9 Plaintiffs further argue that the predominant questions raised by the claims in this case concern whether the invalid click filters employed by Citysearch were sufficient to eliminate clicks that were clearly and objectively invalid. Id. at 14 (citing Local Joint Executive Board of Culinary/Bartender Trust Fund v. Las Vegas Sands Inc., 244 F.3d 1152, 1162-63 (9th Cir. 2001) (“When common questions present a significant aspect of the case and they can be resolved for all members of the class in a single adjudication, there is clear justification for handling the dispute on a representative rather than on an individual basis.”)). 17 1 1995) (certifying a class where contracts signed by all class members contained virtually 2 the same provision as that challenged by the class representative)). Plaintiffs assert that 3 the entire class agreed to standard form agreements prepared by Citysearch and that 4 every class member is a party to the form contract provisions in dispute. Id. at 15. 5 Plaintiffs contend that even when class members’ claims arise out of similar, but not 6 identical contracts, claims arising out of pre-printed form contracts are particularly 7 appropriate for class action treatment. Id. (citing DeBoer v. Mellon Mortg. Co., 64 F.3d 8 1171 (8th Cir. 1995); Heartland Commc’ns, 161 F.R.D. at 114-17). 9 Plaintiffs further contend that under California law, in interpreting a contact, “the 10 question is not what [the parties] subjectively intended, but what a reasonable person 11 would believe the parties intended.” Id. (quoting Beard v. Goodrich, 110 Cal. App. 4th 12 1031, 1038 (2003)). Thus, plaintiffs assert that admissible extrinsic evidence does not 13 include evidence of a party’s unilateral or subjective intent as to the meaning of a 14 contract word or term. Id. (citing Abbott Lab. v. Alpha Therapeutic Corp., 164 F.3d 15 385, 387 (7th Cir. 1999); Peoples v. Sebring Capital Corp., 2002 WL 406979, *8 (N.D. 16 Ill. Mar. 15, 2002)10). Therefore, plaintiffs argue that the determination of whether the 17 express terms of the contract between Citysearch and the class impose certain 18 obligations on Citysearch will be made on a class-wide basis using an objective 19 standard, and the scope and enforceability of Citysearch’s form contracts present 20 common issues of law and fact. Id. at 16. 21 Citysearch responds that plaintiffs’ breach of contract claim raises individual 22 issues that defeat predominance. Opp’n at 25. Citysearch argues that to prove this claim 23 plaintiffs would need to rely on extrinsic evidence, which would vary widely among 24 advertisers. Id. at 25-6. Citysearch argues that plaintiffs previously argued in opposing 25 26 27 28 10 Citysearch argues that plaintiffs’ reliance on this case for the proposition that a court will not consider extrinsic evidence of the intent of each plaintiff in entering a form contract is misplaced, because the basis for the Court’s decision in Peoples was that neither party argued that the form contract was ambiguous. Opp’n at 25. 18 1 Citysearch’s prior motion to dismiss that the Court decided November 4, 2008, that only 2 clicks by “users” were chargeable under the contract, and offered a contorted 3 interpretation of the term “user” to mean a “potential client” of the advertiser, thereby 4 attempting to overcome the contract’s requirement that advertisers pay for “each click” 5 and Citysearch’s express disclaimer of warranty as to click quality. Id. (citing Jobson, 6 Exhib. F at 6). Citysearch contends that in response to this Court’s skepticism as to 7 plaintiffs’ definition of “user,” plaintiffs stated that they would introduce extrinsic 8 evidence to support their interpretation. Id. at 25-26 (citing Jobson, Exhib. F, at 8 9 (dismissal on pleadings inappropriate because “at oral argument, plaintiffs stated that 10 they intend to introduce additional extrinsic evidence beyond that which was already 11 submitted to support their interpretation”). Citysearch argues that plaintiffs have since 12 offered no such evidence. Id. 13 Citysearch further argues that a contract’s objectively reasonable meaning can be 14 determined only in context. Id. at 26. Citysearch asserts that if, for example, an 15 advertiser was actually told about a purported “fraud guarantee,” that fact might support 16 an objectively reasonable interpretation of the contract that is far different than in the 17 case of another advertiser who saw or heard only a disclaimer of click quality and 18 Citysearch’s policy that it could not guarantee it would detect all click fraud. Id. 19 Plaintiffs reply that Citysearch does not contest that the only terms of the contract 20 relevant to this case, that the advertiser will pay for clicks from “users,” do not differ in 21 any material respect among class members. Reply at 7. Plaintiffs argue that 22 Citysearch’s argument—that an individual analysis of each class members’ 23 understanding of the contract is required to determine whether they intended the contract 24 to require payment for all clicks regardless of the source—is contrary to established 25 principles of contract law. Id. Plaintiffs assert that “[a] standardized agreement ‘is 26 interpreted wherever reasonable as treating alike those similarly situated, without regard 27 to their knowledge or understanding of the standard terms of the writing.’” Id. (quoting 28 Sebring Capital Corp., 2002 WL 406979, at *8 (quoting Restatement (Second) of 19 1 2 Contracts § 211(2) 1981)). The Court finds that common issues of law and fact predominate with regard to 3 plaintiffs’ breach of contract claim. Here, the claim arises from a standard form contract 4 prepared by Citysearch to which all advertisers in the class agreed.11 See Kleiner, 97 5 F.R.D. at 691 (“When viewed in light of Rule 23, claims arising from interpretations of a 6 form contract appear to present the classic case for treatment as a class action, and 7 breach of contract cases are routinely certified as such.”). To the extent that the 8 language in the contract is unambiguous, the Court would apply that meaning and not 9 look to extrinsic evidence. Beard, 110 Cal. App. 4th at 1038 (“[I]f the meaning a 10 layperson would ascribe to contract language is not ambiguous, we apply that 11 meaning.”); Pac. Gas & Elec. Co. v. G.W. Thomas Drayage & Rigging Co., 69 Cal. 2d 12 33, 38 (1968). Extrinsic evidence, however, would be properly admitted to construe the 13 terms of the contract if the Court determines that its language is ambiguous. Pac. Gas & 14 Elec. Co., 69 Cal. 2d at 39-40 (“[R]ational interpretation requires at least a preliminary 15 consideration of all credible evidence offered to prove the intention of the parties. Such 16 evidence includes testimony as to the circumstances surrounding the making the 17 agreement—including the object, nature and subject matter of the writing—so that the 18 court can place itself in the same situation in which the parties found themselves at the 19 time of contracting. If the court decides, after considering this evidence, that the 20 language of a contract . . . is fairly susceptible of either one of the two interpretations 21 contended for, extrinsic evidence relevant to prove either of such meaning is 22 admissible.”); Pac. State Bank v. Greene, 110 Cal. App. 4th 375, 385-86 (2003). 23 Extrinsic evidence that the Court would consider in making this determination, such as 24 representations on Citysearch’s website, can be established on a classwide basis. 25 Therefore, it cannot be said that individual issues predominate as to plaintiff’s claim of 26 11 27 28 Citysearch does not contend that the form agreements varied by class member. However, class action treatment may still be appropriate when the claim arises out of similar, but not identical contracts. See Heartland Commc’ns. Inc., 161 F.R.D. at 114-17. 20 1 breach of contract. 2 3 4 ii. Breach of the Implied Covenant of Good Faith and Fair Dealing Plaintiffs argue that the determination of Citysearch’s liability for breach of the 5 implied covenant of good faith and fair dealing is a question common to the class, 6 because [a] party violates the covenant if it subjectively lacks belief in the validity of its 7 act or if its conduct is objectively unreasonable.” Mot. at 16 (quoting Carma Dev., Inc. 8 v. Marathon Development Cal., Inc., 2 Cal. 4th 342, 371-2 (1992)). Plaintiffs contend 9 that under either standard, proof of Citysearch’s liability will be common to the class: 10 Citysearch will be liable if it either subjectively knew that it should not charge its 11 advertisers for certain types of clicks, or if its failure to prevent such charges was 12 objectively unreasonable. Id. at 16-17. 13 Citysearch responds that plaintiffs’ claim for breach of the implied covenant of 14 good faith and fair dealing likewise raises individual issues. Opp’n at 27. First, 15 Citysearch asserts that because the reach of the covenant is dependent on the contract’s 16 express terms, the extent of Citysearch’s obligations will vary dramatically based on 17 what each class member saw and heard. Id. (citing Carma Dev., 2 Cal. 4th at 374). 18 Second, Citysearch asserts that while plaintiff’s have acknowledged that one element of 19 the legal test is whether defendant subjectively acted in bad faith—that is, whether 20 Citysearch knew that clicks it was charging for were fraudulent or invalid—plaintiffs’ 21 algorithm cannot make this determination anymore than it can determine the subjective 22 intent of an Internet user. Id. Third, Citysearch contends that even if plaintiffs’ theory 23 rested on objective fair dealing alone, an examination of the facts of which Citysearch 24 was aware would remain necessary because it is only by analyzing the many individual 25 characteristics of clicks that one could possibly determine whether Citysearch 26 objectively should have known not to charge for particular clicks. Id. 27 28 Plaintiffs reply that with respect to their claim for breach of the implied covenant, Citysearch will be in breach if the invalid click procedures it uniformly employs are 21 1 found to be objectively unreasonable. Reply at 9-10 (citing Carma Dev., 2 Cal. 4th at 2 372). Plaintiffs contend that Citysearch will also be in breach of the implied covenant if 3 it subjectively believed that its invalid click procedures were unreasonable. Id. at 10. 4 Plaintiffs argue that both of these inquiries are susceptible to common proof, and there 5 will be no need to analyze the individual characteristics of each click to make these 6 determinations because the determination of Citysearch’s liability will rest on whether 7 the procedures employed were reasonable. Id. 8 9 In this case, the Court finds that the alleged breach of contract—namely failing to filter and otherwise monitor for click fraud—if proven, is objectively unreasonable 10 conduct that can be established without resort to individualized proof. Carma Dev., 2 11 Cal. 4th at 371-2 (“A party violates the covenant if it subjectively lacks belief in the 12 validity of its act or if its conduct is objectively unreasonable.”). Accordingly, because 13 the Court determines that common issues of law and fact predominate with regard to 14 plaintiffs’ breach of contract claim, the Court also concludes that common issues 15 predominate with regard to plaintiffs’ breach of the implied covenant of good faith and 16 fair dealing claim. 17 18 19 b. Claims Under the UCL i. “Fraudulent” Prong Plaintiffs argue that common issues of law predominate as to their claims under 20 the “fraudulent” and “unfair” prongs of the UCL. Mot. at 17. Plaintiffs assert that their 21 claim under the “fraudulent” prong will not be adjudicated based on their individual 22 circumstances, but instead under the “reasonable consumer” test which only requires that 23 plaintiffs “show that members of the public are likely to be deceived.” Id. (citing 24 Williams v. Gerber Products Co., 523 F.3d 934, 938 (9th Cir. 2008)). Because the 25 “UCL’s focus [is] on the defendant’s conduct, rather than the plaintiffs’ damages,” 26 plaintiffs argue that proof of deception, reliance, and injury are not required. Id. 27 (quoting In re Tobacco II Cases, 46 Cal. 4th 298, 312 (2009)). Plaintiffs further contend 28 that the UCL “imposes strict liability,” and thus “it is not necessary to show that the 22 1 defendant intended to injure anyone.” Id. at 18 (quoting South Bay Chevrolet v. General 2 Motors Acceptance Corp., 72 Cal. App. 4th 861, 877 (1999)). 3 Citysearch responds that the UCL’s “fraudulent” prong requires individualized 4 analysis of what class members saw or heard. Opp’n at 29. Citysearch concedes that 5 plaintiffs are correct that whether a practice is “fraudulent” under the UCL focuses on 6 whether the representations in question are likely to deceive members of the public. Id. 7 But, Citysearch argues that plaintiffs are not correct in their assertion that this ends the 8 discussion. Id. at 29-30. Instead, Citysearch urges that there is a need for individualized 9 inquiry concerning such representations, because all advertisers were not exposed to the 10 same alleged misrepresentations. Id. (citing Hodes v. Van’s Int’l Foods, 2009 WL 11 2422214, *4 (C.D. Cal. July 23, 2009) (even though individualized proof of reliance on 12 allegedly deceptive statement is not required of non-class representatives, common 13 issues did not predominate where many other individualized inquiries remained, 14 including whether the products purchased by each class member contained the purported 15 misrepresentation); Stern, 2008 U.S. WL 4382796, at *10 (absent evidence that 16 defendant’s sales representatives used a “standardized script or forms when making sales 17 calls,” it was unclear how plaintiff could prove her claim on a classwide basis)). Here, 18 Citysearch argues that it is undisputed that the putative class members became 19 advertisers with Citysearch through different methods, and received different 20 information, and there is no evidence of any common script. Id. at 30. Thus, Citysearch 21 argues that predominance is not met. Id. 22 Plaintiffs reply that liability under the UCL stems not only from affirmative 23 misrepresentation but also from failure to disclose material information. Reply at 8 24 (citing Day v. AT&T Corp., 63 Cal. App. 4th 325, 332-333 (1998)). Plaintiffs assert that 25 they have offered sufficient evidence to show a common classwide omission because (1) 26 the information available on Citysearch’s website does not disclose that Citysearch will 27 charge its advertisers for invalid clicks, (2) the training materials Citysearch provides to 28 its sales representatives do not state that Citysearch will charge its advertisers for invalid 23 1 clicks, and (3) Citysearch admits that in most conversations with potential advertisers, 2 the potential for invalid clicks is never discussed. Id. (citing Kabatech Decl. Exhibs. 2, 3 9, 11; Asher Decl. at ¶ 19). Plaintiffs contend that having demonstrated that Citysearch 4 fails to disclose the same material information from all class members, the Court is left 5 with the question of whether this omission is likely to deceive a reasonable consumer. 6 Id. (citing In re Tobacco II Cases, 46 Cal. 4th at 320). 7 The Court agrees with plaintiffs that common issues predominate with regard to 8 plaintiffs’ claim of a common classwide omission under the “fraudulent” prong of the 9 UCL. See Day, 63 Cal. App. 4th at 332-334 (finding defendant liable for failing to 10 disclose material information to all class members under the “fraudulent” prong of the 11 UCL). This UCL claim will be adjudicated under the “reasonable consumer” standard 12 rather than by examining the individual circumstances of each plaintiff. The “reasonable 13 consumer” standard requires plaintiffs to “show that members of the public are likely to 14 be deceived.” Williams, 523 F.3d at 938. Here, the “focus is on the defendant’s 15 conduct, rather than the plaintiff’s damages.” In re Tobacco II Cases, 46 Cal. 4th at 312, 16 320 (“relief under the UCL is available without individualized proof of deception, 17 reliance and injury”). Because the UCL claim arises out of Citysearch’s common course 18 of conduct towards all class members, and the “reasonable consumer” standard is 19 employed to adjudicate the claim, the Court finds that predominance prong is met with 20 regards to plaintiffs’ claim under the “fraudulent” prong of the UCL. 21 22 ii. “Unfairness” Prong Plaintiffs argue that to establish a UCL claim based on the “unfairness” prong, 23 plaintiffs need only show a connection between the alleged wrongdoing and the resulting 24 harm. Mot. at 17 (citing In re Firearm Cases, 126 Cal. App. 4th 959, 981-82 (2005)). 25 Therefore, plaintiffs assert that the proof they must present, and the legal standard they 26 must satisfy to support their UCL claims are objective in nature and thus ideally suited 27 for class treatment. Id. 28 Citysearch responds that plaintiffs’ claim under the “unfairness” prong of the 24 1 UCL requires individualized analysis of advertisers’ expectations. Opp’n at 28. 2 Citysearch argues that this Court has stated it will apply a balancing test to determine 3 whether Citysearch’s alleged practice is “unfair” under the UCL. Id. Under this test, 4 “the determination of whether a particular business practice is unfair necessarily 5 involves an examination of its impact on its alleged victim, balanced against the reasons, 6 justifications and motives of the alleged wrongdoer.” Id. (quoting Motors, Inc. v. Times 7 Mirror Co., 102 Cal. App. 3d 735, 740 n.2 (1980)). Citysearch asserts that under this 8 test, a plaintiff’s individual expectations about the practice are relevant to determining 9 the extent of its harm, and thus could defeat predominance. Id. (citing Lozano, 504 F.3d 10 at 736 (“We agree that evidence about individual knowledge and expectations may help 11 the court determine the extent of the harm for the purposes of the UCL’s balancing 12 test.”)). 13 Plaintiffs reply that while plaintiff Menagerie advertised make-up artistry in 14 Portland, Oregon and plaintiff Redwold advertised payroll services in Charlotte, North 15 Carolina, clicks to their advertisements went through the same set of filters as any other 16 Citysearch advertiser and the same types of clicks were or were not filtered out. Reply 17 at 8. Plaintiffs thus assert that any determination of the propriety of Citysearch’s invalid 18 click procedures can be done on a classwide basis. Id. at 8-9. Plaintiffs argue that while 19 Citysearch claims that since 2004 it has used “sophisticated computer algorithms called 20 filters to identify clicks Citysearch treats as invalid,” Citysearch does not contend that its 21 automated invalid click procedures function differently depending on the particular 22 advertiser. Id. at 9. Furthermore, plaintiffs assert that Citysearch has collected and 23 stored the same relevant information on each click throughout the class period (citing 24 Rhoden Decl. ¶¶7-8; Exhibs. E-F), and plaintiffs’ expert was able to identify suspect 25 clicks using this data (including date, time, session cookie, ad id, and IP address of the 26 clicks). Id. (citing Kabatech Decl. Exhib 27, at 15-16). Plaintiffs argue that this uniform 27 application of Citysearch’s invalid click procedures to all of its advertisers makes the 28 adjudication of plaintiffs’ claims under the implied covenant of good faith and fair 25 1 dealing and the unfair prong of the ULC amenable to class treatment because liability for 2 both depends on whether Citysearch failed to filter out clicks that it reasonably should 3 have. Id. 4 Furthermore, plaintiffs reply that in determining Citysearch’s liability for 5 violation of the “unfairness” prong of the UCL, the Court will weigh the utility of 6 Citysearch’s failure to properly filter invalid clicks against the harm this practice caused 7 the class. Id. Plaintiffs argue that individual inquiries into each class members’ 8 expectations will not be required to determine liability under the unfair prong of the 9 UCL, because where the district court concludes that a defendant engaged in a single 10 course of wrongful conduct, a determination that common issues predominate is proper. 11 Id. (citing Lozano, 504 F.3d at 737). 12 The Court finds that common issues do not predominate with regard to plaintiffs’ 13 “unfairness” prong claim under the UCL. “The test of whether a business practice is 14 unfair involves an examination of that practice's impact on its alleged victim, balanced 15 against the reasons, justifications and motives of the alleged wrongdoer.” South Bay 16 Chevrolet, 72 Cal. App. 4th at 886. Under this test, a plaintiff’s individual expectations 17 about the business practice are relevant to determining the extent of its harm. Id. at 895 18 (finding defendant’s method of calculating interest not “unfair” under UCL, where 19 plaintiff had entered into the contract “knowing, understanding, agreeing, and 20 expecting” that defendant would calculate interest by this method)); see also Lozano, 21 504 F.3d at 736-37. Indeed, a plaintiff’s individual expectations about Citysearch’s 22 business practices are relevant in all three of plaintiffs’ claims under the unfairness 23 prong of the UCL. Accordingly, the court finds that predominance under Fed. R. Civ. 24 Pro. Rule 23(b)(3) is not met for plaintiffs’ “unfairness” prong claim under the UCL. 25 26 c. The Nationwide Application of California Law Plaintiffs argue that “[a] court may certify a nation-wide class for alleged 27 violations of state law,” but for “certification to be proper, application of the state’s laws 28 to out-of-state class members must comport with due process.” Mot. at 18 (quoting 26 1 Parkinson v. Hyundai Motor America, 2008 WL 5233200, at *6 (C.D. Cal. 2008)). 2 “Once plaintiffs make this showing, the burden shifts to the non-movant to show that the 3 laws of another state apply.” Parkinson, 2008 WL 5233200, at *6. “To satisfy their 4 burden, plaintiffs must show that California has a significant contact or aggregation of 5 contacts to the claims of the class members to ensure that California law is not being 6 applied arbitrarily.” Id. (quoting Parkinson, 2008 WL 5233200, at *6). Plaintiffs argue 7 that all versions of the contract between Citysearch and the class in force during the class 8 period contain a choice of law clause stating that California law governs the contract. 9 Id. at 19 (citing Kabateck Decl., Exhibs. 19-26). Therefore, plaintiffs argue that there is 10 11 no dispute that California law applies to plaintiffs’ breach of contract claims. Id. Plaintiffs contend that national certification of their unfair competition claims is 12 also warranted because Citysearch’s unfair and fraudulent conduct originated and 13 emanated from California. Id. Plaintiffs assert that defendant Citysearch, LLC is 14 headquartered in West Hollywood, California, every division responsible for 15 Citysearch’s invalid click procedures is based in California, and Citysearch’s CEO is 16 located in California. Id. Plaintiffs contend that these significant contacts insure that the 17 application of California’s UCL is not arbitrary and is fully consistent with the 18 requirements of due process. Id. at 19-20 (citing Mazza v. American Honda Co., 254 19 F.R.D. 610, 620 (C.D. Cal. 2008) (finding significant contacts where principal place of 20 business and corporate headquarters was in California, including its headquarters for 21 sales, marketing, research, and development)). Plaintiffs assert that the burden thus 22 shifts to Citysearch to show that the laws of another state should apply. Id. 23 Plaintiffs further argue that “in California, a three-step ‘governmental interest’ test 24 is used to determine choice of law questions,” where the Court (1) “first determines 25 whether the laws of each potentially concerned state are different from those of 26 California,” (2) then “examines each jurisdiction’s interest in the application of its own 27 law,” and (3) if “the non-forum state laws differ from those of the forum state and the 28 non-forum states have an interest in applying their laws in the action, the Court will 27 1 select the law of the state whose interests would be most impaired.” Id. at 20 (quoting 2 Parkinson, 2008 WL 5233200, at *7). Plaintiffs contend that “California’s consumer 3 protection statutes, upon which some of plaintiffs’ claims are based, are largely 4 homogeneous with those of the other fifty states and, if anything, afford greater 5 protection to consumers.” Id. (quoting Parkinson, 2008 WL 5233200, at *7). Plaintiffs 6 further contend that California “has a strong interest in policing wrongful conduct 7 allegedly occurring within its borders, including when the victims of such conduct are 8 out-of-state residents.” Id. (quoting Parkinson, 2008 WL 5233200, at *7; Mazza, 254 9 F.R.D. at 622-624 (holding that no other state has an interest in the litigation and, even if 10 others did, California’s interest would be more impaired if its laws were not applied)). 11 The Court concludes that California law can be applied to the plaintiffs’ breach of 12 contract claims, because all versions of the contract between the class and Citysearch in 13 force during the class period contain a choice of law clause stating that California law 14 governs the contract. See Kabateck Decl., Exhibs. 19-26. Furthermore, the Court agrees 15 that California’s UCL can be applied to the nationwide class, as Citysearch has not 16 shown that any “differences between California law and the law of other jurisdictions are 17 material,” nor that “other states have an interest in applying their laws in this case.” 18 Mazza, 254 F.R.D. at 621-24 (holding that California’s UCL can be applied to a 19 nationwide class). 20 2. SUPERIORITY 21 In addition to a predominance of common questions, a class proponent must also 22 demonstrate that the class action is superior to other methods of adjudicating the 23 controversy. See Valentino, 97 F.3d at 1235 (explaining that the party seeking 24 certification needs to make a “showing [as to] why the class mechanism is superior to 25 alternative methods of adjudication”). A class action may be superior where “class-wide 26 litigation of common issues will reduce litigation costs and promote greater efficiency.” 27 Id. at 1234. Rule 23(b)(3) provides the following non-exhaustive list of four factors to 28 consider in this assessment: 28 1 (A) prosecution or defense of separate actions; 2 3 (B) (C) 8 9 the desirability or undesirability of concentrating the litigation of the claims in the particular forum; 6 7 the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; 4 5 the interest of members of the class in individually controlling the (D) the difficulties likely to be encountered in the management of a class action. Fed. R. Civ. P. 23(b)(3). 10 The greater the number of individual issues to be litigated, the more difficult it 11 will be for the court to manage the class action. See, e.g., Dalkon Shield, 693 F.2d at 12 856. Thus, a class action is inappropriate where an individual class member would be 13 compelled to try “numerous and substantial issues to establish his or her right to recover 14 individually, after liability to the class is established.” Schwarzer et al., supra, at § 15 10:361. On the other hand, the fact that individual members seek separate damages is 16 not fatal to class treatment. Id. 17 Plaintiffs argue that each of the Rule 23(b)(3) factors that guide the superiority 18 inquiry is satisfied here. Mot. at 21. First, plaintiffs assert that individually prosecuting 19 plaintiffs’ claims is impractical because the cost of litigating a single case would exceed 20 the potential return. Id. (citing Mazza, 254 F.R.D. at 628). Plaintiffs assert that the 21 second and third factors also support certification of this action, because to plaintiffs’ 22 knowledge, there are no other actions against Citysearch for the claims asserted by 23 plaintiffs, and concentrating the litigation in this Court will allow it to proceed in an 24 efficient manner without risking inconsistent outcomes. Id. Finally, plaintiffs contend 25 that there will be little difficulty in managing this action as a class action, and that they 26 need only show that their methods of calculating damages are plausible at the 27 certification stage. Id. at 21-22 (citing Negrete v. Allianz Life Ins. Co. of North 28 America, 238 F.R.D. 482, 495-96 (C.D. Cal. 2006)). 29 1 Although the parties respectively in their moving papers and opposition dispute 2 whether there is a plausible, classwide method for establishing liability and proving 3 damages, plaintiff’s best articulate this methodology as follows: 4 5 Step 1: Plaintiffs will first establish liability by showing that there are 6 categories of clicks that are (i) clearly and objectively invalid for which 7 plaintiffs and the class should not have been charged, and (ii) that Citysearch 8 could have filtered these clicks out if it had chosen to do so. By definition, 9 these categories of clearly invalid clicks will be defined by objective criteria 10 available in the click-log files—otherwise Citysearch could never have filtered 11 these clicks out if it had chosen to do so. 12 13 Step 2: Using the objective criteria determined in Step 1, plaintiffs will then 14 establish (i) whether a class member was charged for a clearly invalid click; 15 (ii) how many clearly invalid clicks the class members were charged for; and 16 (iii) how much each class member paid for each of these clicks. Because, as 17 noted above, clearly invalid clicks will be defined by objective criteria (for 18 example, doubleclicks would be defined as multiple clicks on the same link 19 originating from the same user session within seconds of each other), 20 identifying invalid clicks will simply require running a database query against 21 Citysearch’s click-log data to extract any clicks that satisfy the objective 22 criteria. Those clicks can be compared against Citysearch’s billing database 23 in order to determine whether the class member was charged for the clearly 24 invalid click, and if so, how much the class member paid for the invalid click. 25 26 Step 3: Finally damages will be calculated using the information obtained 27 from Step 2. Calculating damages will be a simple matter of arithmetic, 28 consisting of multiplying the number of clearly invalid clicks that the class 30 1 member was charged for by the amount that the class member paid for each of 2 those clicks. 3 4 Reply at 14-15. Plaintiffs argue that based on the data stored by Citysearch, their expert will be 5 able to retroactively identify invalid clicks for the class.12 Id. at 22 (citing Jansen Rep.). 6 Plaintiffs assert that because Citysearch claims to record each day’s clicks and then filter 7 them retroactively, plaintiffs should be able to filter retroactively on a class wide basis. 8 Id. at 24. Plaintiffs acknowledge that a computer algorithm cannot always determine an 9 Internet user’s true intent and that no algorithm or filter can currently be developed that 10 will identify all invalid clicks. Id. at 23. Nonetheless, plaintiffs argue that there are 11 certain types of clicks where the user’s intent can be inferred with a high degree of 12 certainty and where there is zero probability of the click creating any value for the 13 12 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 In order to bolster their argument of a plausible methodology to determine classwide liability and damages, plaintiffs submit the report of Dr. Bernard Jansen. Dr. Jansen opines that it is possible to determine retroactively whether a click was invalid using historical data that is stored in Citysearch’s customer account databases. Jansen Rep. at 3. According to Dr. Jansen, this process would involve taking this historical data and running it through an algorithmic process to detect invalid clicks. Id. Dr. Jansen testifies that the classification of such clicks “is not a 100% accurate process” but there are “standard approaches” that achieve “industry-standard, acceptable results.” Id. Furthermore, Dr. Jansen opines that based on examining spreadsheets of Citysearch customer account click data, he noted click patterns that appeared to him to be invalid. Id. To counter Dr. Jansen’s report, Citysearch submits the report of Dr. Alexander Tuzhilin. Dr. Tuzhilin opines that “[i]t is not possible to retroactively determine, through running click log data through an algorithm, whether a click is valid or not, as Dr. Jansen has defined validity,” because “it is impossible for the advertising platform to design an algorithm that could measure the subjective intent of the user who clicks on an online advertisement and measure the “business value” of that click.” Tuzhilin Rep. at 5. Dr. Tuzhilin further opines that “[e]ven if such an algorithm could be designed, it would not and could not answer any of the following questions with respect to any click identified by the algorithm as invalid: (1) whether the advertiser was actually charged for it; (2) if so, how much that advertiser was charged; or (3) whether Citysearch knew or should have known that the click was invalid at the time it occurred.” Id. 31 1 advertiser, such as doubleclicks and clicks that violate an advertising platform’s invalid 2 click policy.13 Id. at 23-24 (citing Kabateck Decl., Exhib. 28 at 10; Exhib. 1 at 16-17). 3 Citysearch responds that the procedure proposed by plaintiffs would not provide a 4 classwide basis for proving their case, because a host of individual issues would be 5 required to be determined at every step. Opp’n at 21. First, Citysearch asserts that there 6 is no way to define all the categories of invalid clicks in purely objective classwide 7 terms, as plaintiffs’ expert defines an “invalid” click as one having “no probability of 8 generating value.” Id. (citing Jansen Rep. at 10). Citysearch argues that this formulation 9 mandates a case-by-case inquiry that is incompatible with classwide treatment since 10 what may have value to one advertiser may be worthless to another. Id. Even in the 11 case of doubleclicks, Citysearch asserts that judgment calls must be made, because there 12 is no industry standard for the time period between clicks. Id. (citing Jansen Depo. at 13 53:20-24; 55:3-6). 14 Second, Citysearch argues that any set of purportedly “objective” rules would 15 likely be at once over-inclusive and under-inclusive. Id. at 22. Citysearch asserts that an 16 automated algorithm could flag perfectly valid traffic, and could miss larger patterns in 17 the data that suggest fraud. Id. (citing Jansen Depo. at 218:1-220:7). Therefore, 18 Citysearch argues that the automated procedures proposed by plaintiffs would not avoid 19 the need for case-by-case judgments about each advertiser’s click data. Id. 20 Third, Citysearch argues that any set of “objective” invalid-click rules that might 21 22 23 24 25 26 27 28 13 Plaintiffs argue that they have presented evidence that Citysearch charges for the second click of a doubleclick. Reply at 12. While plaintiffs assert that determinations must be made regarding the number of clicks and the time span that would define a “doubleclick,” the Court would decide these questions on a classwide basis so that the definition of a doubleclick would apply to all clicks for all advertisers. Id. Plaintiffs further assert that they have presented evidence that Citysearch failed to exclude automated clicks from its syndication traffic for over two years. Id. (citing Kabateck Decl., Exhib. 10). Plaintiffs assert that whether Citysearch should charge for automated clicks is a question that can be objectively determined and this determination will apply to all class members. Id. 32 1 make sense today would not be appropriate for application to click data generated years 2 ago, because norms and expectations about which clicks should be filtered are constantly 3 evolving.14 Id. (citing Jansen Rep. at 13). Fourth, Citysearch argues that the data 4 necessary to form plaintiffs’ retrospective analysis does not exist in its entirety, and the 5 data that does exist is inconsistent, fragmented, and unsuitable for the proposed process. 6 Id. at 23 (citing Rhoden ¶¶ 20-24). Thus, Citysearch asserts that for many advertisers in 7 the putative class, plaintiffs’ proposal would afford no basis for relief whatsoever. Id. 8 Finally, Citysearch contends that the results of any algorithmic process must be 9 measured against what each advertiser actually paid, as plaintiffs concede that their 10 proposed algorithm could not determine whether any given invalid click that it detects 11 was actually charged to the advertiser. Id. (citing Mot. at 25; Jansen Depo. at 43; 12 Tuzhilin Rep. at 17-18). Citysearch asserts that without this information, liability cannot 13 be determined in the first place. Id. In sum, Citysearch argues that plaintiffs’ proposed 14 attempt at classwide adjudication falls flat, because their proposed methodology is not 15 only riddled with conceptual flaws and individual issues but is entirely unmanageable. 16 Id. 17 Moreover, Citysearch argues that plaintiffs’ method of calculating damages does 18 not include any means for addressing the issue of “invalid” clicks for which refunds may 19 already have been obtained, but as to which there is no one-to-one refund correlation, or 20 the issue of “invalid” clicks for which offsets exist in the form of valid clicks that should 21 have been charged to advertisers but were not because of the pre-set cap on monthly 22 charges. Id. at 24-25 (citing Bell Atl. Corp. v. AT&T Corp., 339 F.3d 294, 306-07 (5th 23 Cir. 2003) (finding class treatment not suitable where the calculation of damages is not 24 25 26 27 28 14 For example, Citysearch asserts that new spiders, robots, and bad IP addresses that should be added to exclusion lists are identified continuously, and conversely, addresses that once were bad may later become perfectly acceptable when they are newly assigned by service providers. Id. Citysearch argues that the doubleclick was not excluded by Google until 2005. Id. (citing Kabatech, Ex. 1 at 31). 33 1 susceptible to a mathematical or formulaic calculation, “or where the formula proposed 2 to calculate individual damages is clearly inadequate.”)). 3 Plaintiffs reply that they only need to show that their proposed method for 4 calculating damages is plausible, and that they have met their burden here. Reply at 13 5 (citing Negrete, 238 F.R.D. at 495-96). Plaintiffs reply that any need to account for 6 different click validity standards over time can be done on a classwide basis.15 Reply at 7 12. Plaintiffs further argue that Citysearch’s arguments that the calculation of damages 8 prevents certification because some class members may have received refunds or some 9 class members may have received “valid” clicks for which they were not charged and for 10 which Citysearch should receive an “offset” are baseless. Reply at 15. Plaintiffs assert 11 that supposed refunds or offsets can be addressed mechanically and efficiently since they 12 necessarily involve the analysis of billing data. Id. (citing Roper v. Consurve, Inc., 578 13 F.2d 1106, 1112 (5th Cir. 1978) (“While it may be necessary to make individual fact 14 determinations with respect to charges, if that question is reached, these will depend on 15 objective criteria that can be organized by a computer, perhaps with some clerical 16 assistance. It will not be necessary to hear evidence on each claim.”)). Plaintiffs further 17 argue that regardless of their veracity or applicability, these issues regarding the 18 calculation of damages are not sufficient to deny certification. Id. (citing Negrete, 238 19 F.R.D. at 494 (“The individuation of damages in consumer class actions is rarely 20 determinative under Rule 23(b)(3).”); Smilow v. SW Bell Mobile Sys., Inc., 323 F.3d 21 32, 40 (1st Cir. 2003) (“Where as here, common issues predominate regarding liability, 22 and then courts generally find the predominance requirement to be satisfied even if 23 individual damages issues remain.”)). 24 25 26 27 28 15 Plaintiffs assert that the factfinder may ultimately determine that different time spans should be used for the definition of a doubleclick depending on the year in which it was made. Id. But plaintiffs contend that since this set of definitions would apply to every click regardless of the advertiser, the factfinder would only need to make the determination once. Id. 34 The Court finds that the Rule 23(b)(3) factors that guide the superiority inquiry 1 2 are satisfied. First, it is clear that individually prosecuting plaintiffs’s claims is 3 impractical because the cost of litigating a single case would exceed the potential return. 4 See Mazza, 254 F.R.D. at 628 (where “potential damages for each class member are 5 approximately $4,000, the court finds that the individual class members do not have a 6 strong interest in controlling the litigation”). Second, it does not appear that any 7 members of the class have commenced any other litigation concerning the controversy 8 alleged herein. Third, concentrating the litigation in this Court will allow it to proceed in 9 an efficient manner without risking inconsistent outcomes, and there is no reason to 10 think that this is an undesirable forum to litigate these claims. Finally, notwithstanding Citysearch’s arguments to the contrary, the Court finds 11 12 that this class action will be manageable. Plaintiffs’ claims are based on an alleged 13 common course of conduct by Citysearch to (1) charge its advertisers for invalid clicks, 14 and (2) make material omissions regarding the existence and quality of its click filters. 15 Furthermore, plaintiffs’ proposed three-step method for establishing liability and 16 calculating damages is plausible. Negrete, 238 F.R.D. at 495-96. 17 V. CONCLUSION 18 In accordance with the foregoing, the Court GRANTS plaintiffs’ motion for 19 national class certification with regard to their breach of contract claims, and their claim 20 under the “fraudulent” prong of the UCL. The Court DENIES class certification to 21 plaintiffs with regard to their claim under the “unfairness” prong of the UCL. The 22 national class is as follows: 23 All persons or entities in the United States who entered into form 24 contracts for pay-per-click advertising through Citysearch.com, paid 25 money for this advertising service, and experienced click fraud by 26 reason of double clicks or Citysearch’s failure to apply automatic 27 filters to traffic from its syndication partners up through March 23, 28 2007. 35 1 2 IT IS SO ORDERED. 3 4 5 Dated: November 9, 2009 _________________________________ CHRISTINA A. SNYDER UNITED STATES DISTRICT JUDGE 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 36

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.