Rabin, CPA v. Google LLC

Case Information

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA STEVE RABIN, et al., Case No. 22-cv-04547-PCP Plaintiffs, ORDER CERTIFYING CLASS AND v. APPOINTING CLASS REPRESENTATIVE AND CLASS GOOGLE LLC, COUNSEL Defendant.

In this case a purported class of Google customers sues the company for its cancellation of services it allegedly promised would remain free in perpetuity. Plaintiffs move to certify this class. For the following reasons, the Court grants the motion in part. The Court certifies a Rule 23(b)(3) class of commercial users who were not eligible to “opt out” of Google’s transition to a paid version of the service. The Court appoints plaintiff Steve Rabin as class representative and Lieff Cabraser Heimann & Bernstein, LLP and Webb, Klase & Lemond, LLC as class counsel. The Court denies certification of a Rule 23(b)(2) class and denies the motion to appoint plaintiff Ian Graves as class representative.

BACKGROUND

In August 2006, Google launched a free “beta” service called “Google Apps for Your Domain” or “Google Apps Standard Edition.” Google offered this service to commercial and non- commercial users alike. The Standard Edition initially offered features that included website hosting, Gmail, custom domain email addresses (such as firstname@lastname.com or @businessname.com), two gigabytes of email storage, a calendar, and various customization and administration tools. The Standard Edition also allowed for multiple users per account, a key offering for Google’s commercial customers, many of whom added multiple employees as users. Google gave users access to these tools free of charge and promised to provide continued access for as long as Google made the service available to its customers. In turn, Google collected valuable user data and feedback that it employed to guide product development and investment. Over time, Google added additional features to the Standard Edition like Google Docs and Google Sheets, many of which remain core to Google’s product offerings today.

In February 2007, Google began offering a paid version of its service called the “Premier Edition.” In exchange for payment, users received access to additional tools that users of the Standard Edition could not access. The Premier Edition initially offered access to Gmail, Google Talk, Calendar, Start Page, Docs, and Spreadsheets. It also offered ten gigabytes of storage per user. In 2016, Google discontinued the Premium Edition and introduced G-Suite, which offered access to many of the same products as well as to new tools like Forms, Slides, Sites, Admin, and Vault. In 2020, Google discontinued G-Suite and introduced Workspace, which similarly kept many of the tools from G-Suite but also introduced new features and enhanced integration between various Google tools. From 2007 onward, Google continued to improve the paid version of its services. In December 2012, Google stopped allowing new users to sign up for the Standard Edition. Instead, Google required any new commercial user to sign up for the paid version of the Google service available at that time. Google allowed those who had signed up for the Standard Edition before 2012 to continue using its services for free. Google refers to these users as “legacy users” or “Legacy Free customers.” Over time, Google stopped actively maintaining the Standard Edition, which eventually fell out of synch with Google’s paid, commercial offerings.

When they signed up for the Standard Edition between 2006 and 2012, users agreed to a set of contract terms. Google regularly updated the language of these terms and conditions. Until July 2011, the “Fees” provision of that contract generally provided that while Google offered its services, it would provide a version of that service to the user free of charge. Google promised to provide a version of its services that was “substantially the same” as the service in place on the date that the user enrolled.

The contract in place until July 2011 also allowed Google to change or modify the terms and conditions of the agreement, except as provided in the “Fees” section. In July 2011, Google amended the contract’s modification clause to reserve to Google the right to change or modify any term or condition contained in the agreement. Simultaneously, Google omitted language agreeing to provide access to the Standard Edition for free so long as Google continued to offer that service. On December 7, 2011, Google amended the “Fees” provision to state that “Google may, at any time, discontinue the non-charge version of the Service and only offer a premium version[,]” provided that Google gave its users adequate notice under the agreement and allowed those users to upgrade to a premium version of the service.

From 2012 through 2022, Google continued to provide legacy users with access to the Standard Edition at no cost. But in January 2022, Google announced its plans to eliminate the Standard Edition entirely, even for legacy users. Google set an August 2022 deadline for legacy users to either transition to a paid Workspace subscription or download their data and terminate their accounts. Those who decided to sign up for Workspace agreed to a user agreement, one clause of which provided that the agreement “terminates and supersedes any and all other agreements between the parties relating to its subject matter, including any prior versions of this Agreement.” Users who transitioned to Workspace began paying a monthly subscription fee. Google changed course in April 2022, announcing that it would allow non-commercial legacy users to “opt out” of the transition to Workspace and to continue using the free Standard Edition. Google did not make the opt-out available to every user. Instead, only those users who self-identified as “non-commercial” could opt out. Google also allowed non-commercial users who had already transitioned to Workspace to “undo” their transition and take advantage of a retroactive opt-out. Google offered this retroactive opt-out until August 23, 2023, about one year after Google’s deadline for its Standard Edition commercial users to transition to Workspace. Approximately 120,000 users opted out and nearly 60,000 other users “undid” their transition after signing up for Workspace.

The Stratford Company, LLC filed this complaint on August 5, 2022 on behalf of a purported class of persons and entities who had lost access to the Standard Edition. On October 26, 2022, plaintiffs filed an amended complaint adding Steve Rabin and Ian Graves as additional named plaintiffs. The Stratford Company voluntarily dismissed its claims against Google without prejudice on November 14, 2022.

Plaintiff Steve Rabin is a sole-proprietor CPA based in California. He signed up for the Standard Edition using a third-party service, CPA Site Solutions, in 2009. Mr. Rabin experienced problems, glitches, and bugs with Google’s offerings but continued to use the Standard Edition through 2022 because many of his clients use Google products. Google sent Mr. Rabin more than fifteen emails notifying him that he would need to transition to Workspace to avoid losing access to his account. Although Mr. Rabin read those emails, he testified that he believed they were fraudulent. When Google suspended his account in September 2022, he entered his credit card details and accepted the Workspace Agreement, unlocking his account. There are seven users associated with Mr. Rabin’s account and he pays between $17 and $50 every month for continued access to Workspace. Plaintiff Ian Graves signed up for the Standard Edition in 2012 when he was a software engineering student at the Rochester Institute of Technology. He alleges that he read about Google’s promise of free access to its Standard Edition by way of a Google press release. Mr. Graves also experienced glitches and defects with the Standard Edition in its early years, but eventually became a strong proponent of Google tools. Google sent Mr. Graves multiple emails about its requirement to transition to Workspace, but Mr. Graves stated that he did not read these emails and instead learned about the transition from various news sources. He agreed to the Workspace agreement on June 23, 2022. As a non-commercial user, Mr. Graves was eligible for the opt-out but did not take advantage of either that option the retroactive opt-out. He testified that he was unaware of both options. There are three users associated with Mr. Graves’s account and he pays between $2 and $22 every month for continued access to Workspace.

The operative complaint, filed on July 17, 2023, asserts three claims against Google: (1) breach of contract, (2) breach of the implied covenant of good faith and fair dealing, and (3) a violation of California’s Unfair Competition Law (UCL) for “unfair” or “unlawful” conduct. The Court dismissed plaintiffs’ good faith and fair dealing claim. Plaintiffs now move for class certification as to the remaining claims.

LEGAL STANDARDS

Federal Rule of Civil Procedure 23 governs class certification. The party seeking class certification must first satisfy the requirements of Rule 23(a) by demonstrating that:

(1) the class is so numerous that joinder of all members is impracticable;

(2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.

Fed. R. Civ. P. 23(a). In addition to making this showing, “the proposed class must satisfy at least one of the three requirements listed in Rule 23(b).” Wal-Mart Stores, Inc. v. Dukes , 564 U.S. 338, 345 (2011). Plaintiffs seek certification under Rule 23(b)(3) or, alternatively, Rule 23(b)(2). Rule 23(b)(3) requires that common questions of law or fact “predominate over any questions affecting only individual members, and that a class action [be] superior to other available methods for fairly and efficiently adjudicating the controversy.” Under Rule 23(b)(2), certification is appropriate where the “party opposing the class has acted or refused to act on grounds that apply generally to the class.” In such a scenario, “final injunctive relief or corresponding declaratory relief is appropriate [as to] the class as a whole.” Id.

The party seeking class certification “must affirmatively demonstrate [its] compliance with Rule 23.” White v. Symetra Assigned Benefits Serv. Co. , 104 F.4th 1182, 1192 (9th Cir. 2024) (citing Wal-Mart Stores , 564 U.S. at 345; Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC , 31 F.4th 651, 665 (9th Cir. 2022) (cleaned up)). Thus, plaintiffs “must actually prove —not simply plead—that their proposed class satisfies each requirement of Rule 23, including … the predominance requirement of Rule 23(b)(3).” Halliburton Co. v. Erica P. John Fund, Inc. , 573 U.S. 258, 275 (2014) (emphasis in original).

ANALYSIS

Plaintiffs move for class certification and ask the Court to certify the following class: All persons or entities in the United States who: (a) signed up for the free version of Google Apps between August 1, 2006 and December 6, 2012; (b) were still Legacy Free customers as of January 19, 2022; (c) had at least one active user on their account during the 180 days prior to January 19, 2022; and (d) did not “opt-out”* prior to being charged for Workspace. * “opt-out” includes customers who affirmatively opted out as non-commercial customers or were migrated to Workspace for Education Fundamentals or Workspace for Nonprofits.

For the reasons that follow, the Court grants plaintiffs’ motion in part. Because the claims of non-commercial class members who were eligible to opt out but did not do so do not satisfy Rule 23(a)’s commonality requirement, the Court redefines the class as provided below. See Armstrong v. Davis , 275 F.3d 849, (9th Cir. 2001) (“Where appropriate, the district court may redefine the class”). The Court further concludes that plaintiff Ian Graves fails to satisfy Rule 23(a)’s typicality requirement as to the redefined class. The Court therefore appoints Steven Rabin as a representative of the following certified class: All persons or entities in the United States who: (a) signed up for the free version of Google Apps between August 1, 2006 and December 6, 2012; (b) were still Legacy Free customers as of January 19, 2022; (c) had at least one active user on their account during the 180 days prior to January 19, 2022; and (d) were did not eligible to “opt-out”* prior to being charged for Workspace.

* Customers eligible to “opt-out” included customers who identified as affirmatively opted out as non-commercial customers or were migrated to Workspace for Education Fundamentals or Workspace for Nonprofits. The Court declines to certify a Rule 23(b)(2) class.

I. Rule 23(a)

Plaintiffs must first demonstrate that their proposed class satisfies the numerosity, commonality, typicality, and adequacy requirements of Rule 23(a). Google does not contest that the purported class satisfies the numerosity requirement. Although neither side provides an exact figure for the size of this purported class, counsel stated at the motion hearing that the class likely includes over 400,000 members. Generally, the numerosity requirement is met where plaintiffs’ class contains forty or more members. See Corley v. Google, Inc. , 316 F.R.D. 277, 290 (N.D. Cal. 2016) (collecting cases). The Court therefore agrees that the numerosity requirement is satisfied.

The parties’ Rule 23(a) dispute instead centers around commonality, typicality, and—to a lesser extent—adequacy. For the following reasons, the Court concludes that there are commonality issues with the class as proposed by plaintiffs but that a more-narrowly defined class satisfies Rule 23(a). The Court further concludes that Ian Graves’s claims are not typical of the class and that he may be an inadequate class representative.

A. Commonality

Rule 23(a) requires that there be a question or law or fact that is common to the class. “What matters to class certification … is not the raising of common ‘questions’—even in droves—but rather, the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation.” Wal-Mart Stores , 564 U.S. at 350. “[E]ven a single common question will do.” Id. at 359 (cleaned up and citations omitted). The putative class asserts three claims against Google, the first of which is a breach of contract claim. They allege that Google breached its contractual promise to provide indefinite free access to the Standard Edition when it terminated that service in 2022. The second claim— essentially a derivative of the first—alleges that this breach of contract violates the “unlawful” prong of California’s Unfair Competition Law (UCL). For these two claims, plaintiffs have identified two common questions:

(1) When, if ever, did Google cease “offer[ing] the Service to the Customer” so as to relieve Google of its obligation to continue to provide Class members access to the free version under the pre-July 19, 2011 contract language?

(2) When, if ever, did Google’s obligation to provide continuing access to the free version cease under the “Fees provision” language applicable to Class members who signed up on or after July 19, 2011?

Although these questions technically apply differently to class members based on whether they signed up before or after July 19, 2011, the answer to both questions will likely be the same and will therefore determine questions of liability in “one stroke.” Olean Wholesale , 31 F.4th at 664. Both questions are a variation of the same common inquiry: When, if ever, did Google’s obligation to provide Standard Edition access to legacy users terminate? If the answer to that question is a date prior to 2022, Google is likely not in breach. But a different answer to that question might instead mean that Google discontinued the Standard Edition before its contractual obligation had terminated. That would likely mean that Google breached its contract, thus requiring an assessment of Google’s affirmative defenses. For this reason, the answers to these common questions will drive the litigation towards resolution.

Google argues that plaintiffs have not satisfied the commonality requirement because they have not shown that they can rely on versions of the user agreement that pre-date the agreement in place when Google terminated the Standard Edition. That version provided that “Google may, at any time, discontinue the no-charge version of the Service and only offer a premium version.” According to Google, the common questions that plaintiffs have identified may be irrelevant to liability because Google was not bound by the contracts addressed in those questions. But the 2022 agreement was operative when Google terminated the Standard Edition and is thus relevant to the claims of every purported class member. If anything, this argument highlights not a lack of common questions but a potential flaw common to the claims of every purported class member. If Google is correct, then the breach of contract claims must necessarily fail because the 2022 contract gave it the express option to terminate the service at any time. Put differently, Google’s “concern about the proposed class is not that it exhibits some fatal dissimilarity but, rather, a fatal similarity—[an alleged] failure of proof as to an element of the plaintiffs’ cause of action[.]” Tyson Foods, Inc. v. Bouaphakeo , 577 U.S. 442, 457 (2016) (citation omitted). Because Google’s argument goes to the merits of the class claims, the Court will “engage that question as a matter of summary judgment, not class certification.” Id. (citation omitted).

Plaintiffs’ third claim is that Google’s actions were “unfair” and thus violate California’s Unfair Competition Law. To prevail on this claim, plaintiffs must demonstrate that Google engaged in some “unfair” act or practice. See Lozano v. AT&T Wireless Servs., Inc. , 504 F.3d 718, 731 (9th Cir. 2007). For this claim, they have identified the following common question: “Was Google’s action here objectively unfair?” Google does not dispute that this is a common question. Although the Court agrees that the commercial members of the proposed class satisfy Rule 23(a)’s commonality requirement, the two common questions plaintiffs have identified will not provide answers driving resolution of the claims of non-commercial members of the purported class. Plaintiffs’ theory of the case is that Google breached its contract by discontinuing the Standard Edition. But for a large subset of the purported class—self-identified non-commercial users—Google offered an opt-out that allowed those purported class members to receive continued service. To be sure, several months passed between Google’s termination announcement and its implementation of the opt-out. But every self-identified non-commercial user could have retained access to the Standard Edition either by opting out or by “undoing” their transition. Indeed, over 180,000 users exercised one of these two options and retained access to the Standard Edition. For those non-commercial users who exercised neither option, the common questions identified above will not resolve their claims. Instead, their claims will depend upon why they did not exercise either option and whether Google properly required them to do so. Plaintiffs argue that Google’s requirement that they opt out of the transition was itself a breach because nothing in the contract permitted Google to condition its obligations on users taking further action. Plaintiffs may be correct that whether Google’s offer of an opt-out constituted a contractual breach presents a common question as to those users. But that claim is not before the Court. The operative complaint asserts a breach theory premised on Google’s decision to terminate access to the Standard Edition, not a theory premised on the opt-out. It addresses the opt-out theory only in connection with plaintiffs’ breach of the implied covenant of good faith and fair dealing claim, which has been dismissed. So even if there are common questions bearing on the claims of this class of plaintiffs, those questions do not relate to the theory of plaintiffs’ current case. ^[1]

B. Typicality

Rule 23(a) also requires that the named plaintiffs’ claims be typical of those of the class. “The test of typicality is whether other members have the same or similar injury, whether the action is based on conduct which is not unique to the named plaintiffs, and whether other class members have been injured by the same course of conduct.” Ellis v. Costco Wholesale Corp ., 657 F.3d 970, 984 (9th Cir. 2011) (citing Hanon v. Dataproducts Corp. , 976 F.2d 497, 508 (9th Cir. 1992)) (cleaned up). “Typicality refers to the nature of the claim or defense of the class representative, and not to the specific facts from which it arose or the relief sought.” Id. (Citing Hanon , 976 F.2d at 508).

Google argues that neither Mr. Graves nor Mr. Rabin satisfy the typicality requirement. The Court concludes that Mr. Rabin’s claims are typical of the class but that Mr. Graves’s are not. Mr. Rabin is a commercial user. Google suspended his account and he signed up for Workspace to regain access. He was never eligible to opt out of the transition. In all of these respects, he is typical of the other commercial users within the purported class. Google nonetheless argues that factual circumstances unique to Mr. Rabin make him an atypical representative. Specifically, it notes that Mr. Rabin signed up for the Standard Edition through a third-party service called CPA Site Solutions, which Google argues subjects him to a unique, third-party beneficiary defense. But from the briefing on this motion, the partes appear to agree that Mr. Rabin was always the primary account holder notwithstanding that he used a third- party to activate his account initially. Although Google may believe it has a unique defense as to Mr. Rabin’s claims, it does not seriously dispute that Mr. Rabin’s core claim—breach of contract—mirrors that of other commercial users in the purported class.

Nor does it appear that, in terminating his account, Google engaged with Mr. Rabin differently from the way it engaged with other purported class members. Although there may be facts specific to Mr. Rabin that he does not share with other class members, that does not make his legal claims atypical of the class. What matters at class certification is that Mr. Rabin’s claim arises from the “same course of events” and that he makes “similar legal arguments to prove the defendant’s liability.” Rodriguez v. Hayes , 591 F.3d 1105, 1124 (9th Cir. 2010) (cleaned up). His experience as a commercial user, ineligible for an opt-out, who lost access to the Standard Edition gives rise to UCL and breach of contract claims identical to those of other members of the class. That renders him a typical plaintiff well-suited to serve as class representative.

Unlike Mr. Rabin, Mr. Graves does not satisfy Rule 23(a)’s typicality requirement as to the members of the class for which commonality is satisfied. Mr. Graves asserts that Google is liable to him because it failed to properly notify him of the opt-out provision. But to be a typical representative of the class, Mr. Graves must demonstrate that he has “the same or similar injury, … [which] is based on conduct [that] is not unique to [him].” Ellis , 657 F.3d at 984. His failure-to- notify theory involves a form of harm different from the compulsory loss of access to the Standard Edition. Any injury Mr. Graves suffered is not solely traceable to Google’s termination of his access to Standard Edition because he could have retained access by opting out. Assessing whether Google is liable to Mr. Graves will depend on the adequacy of Google’s notice to Mr. Graves under the agreement, rather than whether termination of access to the Standard Edition constituted a breach as to Mr. Graves. He therefore cannot show that he has “been injured by the same course of conduct” that injured the absentee class members. Id. C. Adequacy Rule 23(a)’s adequacy prerequisite requires plaintiffs to show that they “will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). The Court must assess two questions: “(1) do the named plaintiffs and their counsel have any conflicts of interest with other class members and (2) will the named plaintiffs and their counsel prosecute the action vigorously

on behalf of the class?” Ellis , 657 F.3d at 985 (citation omitted). This depends largely on the “absence of antagonism between representatives and absentees, and a sharing of interest” between the two. Id.

Google does not explicitly contest Mr. Rabin’s adequacy as a class representative, and the Court concludes that he will adequately represent the interests of the class. His claims arise from the same underlying facts as those of the other class members. As noted above, the fact that he signed up for the Standard Edition through a third-party service does not undermine the fact that his legal claims mirror those of the other class members. Google has not raised any potential conflicts of interest between Mr. Rabin and the absentee class members, and the Court cannot identify any such conflicts. The fact that Mr. Rabin uses Google’s services for his own business and has multiple users associated with his account signals that he has a clear interest in the outcome of this litigation and will therefore vigorously prosecute the action on behalf of the class. He has also selected well-experienced and seasoned counsel to represent him in this matter. Both firms have experience representing plaintiffs in similar class action litigation, and Google has not argued that they would fail to adequately represent the class here. The Court therefore concludes that Mr. Rabin and the proposed class counsel will adequately represent the interests of the class.

At hearing on this motion, Google’s counsel argued that “lapsed users” lack an adequate representative. These are legacy users who did not do anything when Google announced that it would terminate the Standard Edition. They neither upgraded to Workspace nor downloaded their data and terminated their accounts; their accounts simply expired. Google has not demonstrated why Mr. Rabin could not adequately represent these class members. There are no clear conflicts of interest between Mr. Rabin and them. If Mr. Rabin prevails, both he and the lapsed users would likely receive some form of expectation damages. Further, Mr. Rabin—as somebody actively paying for a service he once received for free—is well suited to vigorously prosecute this action on behalf of the class, perhaps more so than an entity that allowed its service to lapse and took no further action to vindicate its rights. Although Google did not argue that Mr. Graves would be an inadequate representative, the Court has concerns about his ability to vigorously represent the interests of absentee class members. Mr. Graves joined this litigation as a named plaintiff in October 2022 after having signed up for Workspace. At that time, he was unquestionably eligible to take advantage of Google’s retroactive opt-out to undo his transition to Workspace and regain access to the Standard Edition. This option remained available to him until August 2023, a full ten months after he joined this litigation. Yet Mr. Graves never exercised that option. His counsel has stated that Mr. Graves was unaware that the retroactive opt-out was available to him. Yet approximately 60,000 non- commercial users exercised this option. It does not appear that Mr. Graves ever reached out to Google to inquire about regaining access. Given his inability or lack of willingness to inquire into ways that he might have retained access to the Standard Edition, the Court has concerns about his ability to vigorously pursue the claims of absentee class members. This, coupled with his atypical claims, means that Mr. Graves cannot serve as an adequate representative of the absent non- commercial class members.

II. Rule 23(b)(3)

Rule 23(b)(3) requires that the common questions of law or fact identified under Rule 23(a) “predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy.” The Court concludes that both requirements are satisfied as to commercial users within the proposed class.

A. Predominance Rule 23(b)(3)’s predominance inquiry “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” Amchem Prods., Inc. v. Windsor , 521 U.S. 591, 623 (1997). “The commonality and predominance inquiries overlap.” White , 104 F.4th at 1191. After separating questions common to the legal claims of the class from those that “present individualized issues,” courts “analyze whether the common questions predominate over the individual questions.” DZ Reserve v. Meta Platforms , 96 F.4th 1223, 1233 (9th Cir. 2024). In essence, this inquiry “asks whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues.” Tyson Foods , 577 U.S. at 453 (2016). To prevail, plaintiffs must demonstrate that the common questions “can be determined in one stroke.” Olean Wholesale , 31 F.4th at 664.

Google argues that the purported class cannot be certified because any common questions that may arise across their causes of action do not predominate over necessary individualized inquiries. Google also argues that if plaintiffs are successful, it would be impossible to adjudicate relief on a class-wide basis.

1. Breach of Contract & UCL Unlawful Claim Plaintiffs’ breach of contract claim and their derivative UCL “unlawful” claim both implicate the common questions of whether Google’s obligation to provide access to the Standard Edition had terminated and whether the class has a right to rely on prior versions of the contract. Google argues that two issues require individualized assessments that will predominate over these common issues: (1) whether class members validly assented to the Workspace agreement, and (2) whether class members were eligible to opt out of the transition to Workspace.

Google first argues that issues presented by its voluntary payment defense defeat Rule 23(b)(3)’s predominance requirement. After transitioning to Workspace, class members began making monthly subscription payments to access their new services. Google previously moved to dismiss plaintiffs’ breach of contract claim in so far as plaintiffs sought to recover those payments, contending that such a recovery is barred by the voluntary payment doctrine, which provides that “[p]ayments voluntarily made, with knowledge of the facts, cannot be recovered.” Steinman v. Malamed , 185 Cal. App. 4th 1550, 1557 (Cal. Ct. App. 2010) (citation omitted). Plaintiffs then amended their complaint to add new allegations showing that class members who converted to Workspace made their payments to Google involuntarily and under duress. See, e.g. , W. Gulf Oil Co. v. Title Ins. & Tr. Co. , 92 Cal. App. 2d 257, 264 (Cal. Ct. App. 1949) (“Payments of illegal claims enforced by duress … will be deemed to have been made involuntarily and may be recovered”). To establish duress sufficient to defeat the voluntary payment defense, plaintiffs will have to demonstrate that “by reason of the peculiar facts a reasonably prudent man [would have found] that in order to preserve his property or protect his business interests it [was] necessary to make a payment of money” Steinman , 185 Cal. App. 4th at 1558 (citation omitted). Google argues that this duress defense is a fact-intensive, individualized issue that will predominate over the litigation-resolving common questions identified by plaintiffs. But even if evaluating duress may require some individualized assessments, that does not mean that plaintiffs’ claims necessarily fail Rule 23(b)(3) because any such evaluation will not resolve core issues of liability. As the Court previously held, Google’s voluntary payment defense is a limitation on “recovery rather than culpability.” Rabin v. Google , 725 F. Supp. 3d 1028, 1031 (N.D. Cal. 2024). Moreover, plaintiffs have plausibly alleged an entitlement to specific performance, see id. , and the voluntary payment doctrine does not apply to a specific performance remedy because “that remedy does not involve any return of past payments,” id. Thus, to the extent Google’s voluntary payment defense is relevant (and that the duress defense thereto is in turn relevant), it will operate solely as a limitation upon damages. As discussed further below, individualized “damage calculations alone cannot defeat certification.” Yokoyama v. Midland Nat’l Life Ins. Co. , 594 F.3d 1087, 1094 (9th Cir. 2010).

Nor will duress-related issues predominate over the common questions arising from plaintiffs’ claim that Google violated the UCL’s “unlawful” prong. Although neither the Ninth Circuit nor the California Supreme Court has addressed whether the voluntary payment defense applies to a consumer UCL claim, state and federal courts routinely hold that it does not. See Bautista v. Valero Mktg. and Supply Co. , No. 15-cv-05557-RS, 2018 WL 11356583, at *3 n.2 (N.D. Cal. Dec. 4, 2018) (collecting cases); Torliatt v. Ocwen Loan Servicing, LLC , 570 F. Supp. 3d 781, 800 (N.D. Cal. 2021) (collecting cases). This Court agrees that applying the voluntary payments doctrine to a UCL claim would likely “run contrary to public policy[.]” Bautista , 2018 WL 11356583, at *4. Further, as the Court previously held, plaintiffs have adequately alleged that they lack adequate legal remedies under their predicate breach of contract claim. If they are correct, the UCL may entitle them to equitable restitution or injunctive relief. See Cal-Tech Comms., Inc. v. L.A. Cellular Tel. Co. , 20 Cal. 4th 16, 179 (1999). Given that Google’s voluntary payment defense does not defeat the predominance of common questions as to plaintiffs’ breach of contract claim, it certainly does not do so with respect to their UCL claim, which expressly provides for injunctive relief. Conversely, the Court agrees with Google that individualized questions related to non- commercial, opt-out-eligible users predominate over any common questions their claims might present. As noted already, approximately 180,000 self-identified non-commercial users either opted out of the transition to Workspace or undid that transition. Determining why the remaining non-commercial users failed to exercise either option would require individualized inquiries regarding the notice provided to those users and their reasons for not accepting Google’s offer of continued free access to the Standard Edition. Accordingly, Rule 23(b)(3)’s predominance requirement provides an independent basis for excluding non-commercial users who were eligible to opt out of the Workspace transition from the certified class. ^[2]

2. UCL Unfair Claim

Google next argues that individualized inquiries will predominate over the common questions presented by plaintiffs’ claim that Google violated the UCL’s “unfair” prong.

To determine whether a business practice is “unfair,” California Courts apply one of two tests, either a “public policy test,” under which a plaintiff must show “that a practice violates public policy as declared by ‘specific constitutional, statutory or regulatory provisions,’” Rubio v. Cap. One Bank , 613 F.3d 1195, 1205 (9th Cir. 2010) (citing Gregory v. Albertson’s, Inc. , 104 Cal. App. 4th 845 (Cal. Ct. App. 2002)), or a balancing test under which plaintiffs must “prove facts showing that ‘the harm to the consumer’ … outweighed the [action’s] utility,’” Id. (citing Lozano , 504 F.3d at 735–36).

The plaintiffs here invoke the balancing test, and Google argues that assessing whether the harm of terminating the Standard Edition outweighs the utility of that action is necessarily an individualized inquiry. In Google’s view, many users chose to upgrade to a paid service with new features and the benefits certain users derived from the transition will vary based on their individual circumstances. The problem with Google’s argument is that it suggests its uniform conduct towards the putative class might, for purposes of the UCL, be “fair” as to some users but “unfair” as to others. “But what is challenged here is one uniform practice” and “[t]he assessment of the legality of this singular practice under the UCL is measured in the aggregate.” Brooks v. Thomson Reuters Corp. , No. 21-cv-01418-EMC, 2023 WL 9316647, at *8 (N.D. Cal. Aug. 10, 2023) (collecting cases). “Where a singular systemic practice is being challenged, that balancing test looks at the harm and benefits in the aggregate.” Id. Here, the plaintiffs allege that Google engaged in a uniform practice by unilaterally terminating all commercial users’ access to the Standard Edition despite years of promising them continued free access. That some class members may have received benefits from the forced migration to Workspace does not create an individualized issue that will predominate over common questions arising from plaintiffs’ UCL “unfairness” claim. Instead, “those members to self-certify their commercial user status, much in the way that Google allowed users to self-identify as non-commercial users and thus opt out of the transition. The parties can then cross reference such certifications against Google’s user data. intermittent benefits must [simply] be weighed, in the aggregate, at large against the harms caused to the entire class as a whole.” Id.

3. Entitlement to Relief Finally, Google argues that plaintiffs have failed to demonstrate that they can litigate relief on a class-wide basis. But Google’s argument is at odds with Ninth Circuit precedent and, in any event, plaintiffs have met their burden to show that relief can be determined on a class-wide basis.

At the damages stage of Rule 23(b)(3)’s predominance analysis, plaintiffs must show that “damages are capable of measurement on a classwide basis.” Comcast Corp. v. Behrend , 569 U.S. 27, 34 (2013). This requires a “translation of the legal theory of the harmful event into an analysis of the economic impact of that event.” Id. (emphasis omitted). In other words, plaintiffs must show “that the whole class suffered damages traceable to the same injurious course of conduct underlying [their] legal theory.” Just Film, Inc. v. Buono , 847 F.3d 1108, 1120 (9th Cir. 2017). “[T]he presence of individualized damages cannot, by itself, defeat class certification under Rule 23(b)(3).” Leyva v. Medline Indus. Inc. , 716 F.3d 510, 514 (9th Cir. 2013). At the class certification stage, plaintiffs need not produce a perfect damages model or expert testimony, but “need only show that such damages can be determined without excessive difficulty and attributed to their theory of liability.” Just Film , 847 F.3d at 1121. Plaintiffs have met their burden of demonstrating that relief on their UCL claims can be litigated on a class-wide basis. Should plaintiffs prevail on their UCL claims, the Court can issue any order or judgment that “may be necessary to restore to any person in interest any money or property, real or personal, which may have been acquired by means of … unfair competition.” Cal. Bus. & Prof. Code § 17203. Under the UCL, “prevailing plaintiffs are generally limited to injunctive relief and restitution.” Korea Supply Co. v. Lockheed Martin Corp. , 29 Cal. 4th 1134, 1144 (2003) (cleaned up and citation omitted). The California Supreme Court has defined UCL restitution as “compelling a UCL defendant to return money obtained through an unfair business practice to those persons in interest from whom the property was taken[.]” Id. at 1144 (citation omitted).

Plaintiffs propose adjudicating restitution by calculating the amount that Google has received from each class member in return for access to Workspace since the termination of the Standard Edition in 2022. This directly relates to their UCL theory. They posit that Google violated the UCL in terminating access to the Standard Edition despite promising class members that they could retain free access in perpetuity. An order requiring Google to return money received since that termination thus would thus “compel[ Google] to return money obtained through an unfair business practice” to class members that had an interest in retaining access to the Standard Edition. Id. Plaintiffs’ expert has confirmed that Google’s historical data shows the subscription payments each class member has made to Google since that time. Google contests neither the availability nor reliability of this data. To account for users that upgraded their account to a more expensive service than Workspace, plaintiffs further propose discounting damages based on the share of users who upgraded to a higher-tiered version. Plaintiffs’ expert has also confirmed that this data exists. Google may be correct that some users would have signed up for Workspace regardless of its termination of the Standard Edition. But Google cannot defeat class certification simply by showing that its alleged “wrongdoing makes it more difficult for the plaintiff to establish the precise amount of its injury.” Dolphin Tours, Inc. v. Pacifico Creative Serv., Inc. , 773 F.2d 1506, 1511 (9th Cir. 1985). To be sure, Google might use this argument to “attempt to pick off the occasional class member here or there through individualized rebuttal,” but this argument alone does not mean that individualized counterfactuals undermine the predominance of questions common to the class. Olean Wholesale , 31 F.4th at 668.

Nor does the fact that class members may have derived a benefit from Workspace defeat classification on predominance grounds. In support of this contention, Google cites a series of cases for the proposition that restitution must be calculated by assessing the “difference between what the plaintiff paid and the value of what the plaintiff received.” In re Vioxx Class Cases , 180 Cal. App. 4th 116, 131 (Cal. Ct. App. 2009); see also In re POM Wonderful LLC , No. ML 10- 02199-DPP, 2014 WL 1225184, at *3 (C.D. Cal. Mar. 25, 2014) (reasoning that restitution “requires evidence of the actual value of what the plaintiff received.”). Although it might be appropriate for the Court to offset damages based on additional features contained in Workspace that were not available in the Standard Edition, Google’s objection is not especially relevant at certification because plaintiffs have demonstrated a viable method of calculating restitution damages should the Court later require such an offset. That method borrows from plaintiffs’ proposed methodology for assessing their breach of contact claim damages (discussed further below) by assessing the market value of Workspace’s additional features and reducing damages accordingly. Plaintiffs have thus satisfied their burden under either measurement to provide class- wide measurements of restitution that are “attributed to their [UCL] theory of liability.” Just Film , 847 F.3d at 1121.

Plaintiffs have also met their burden to demonstrate that relief on their breach of contract claim can be litigated on a class-wide basis. Should plaintiffs prevail on their breach of contract claim, they will be entitled to expectation damages. Such damages are generally equal to the difference between the value of performance absent breach and the value of the performance actually rendered. See Williams v. Apple, Inc. , 338 F.R.D. 629, 654 (N.D. Cal. May 28, 2021) (citing Restatement (Second) of Contracts § 347; Postal Instant Press, Inc. v. Sealy , 43 Cal. App. 4th 1704, 1709 (Cal. Ct. App. 1996)). Plaintiffs propose calculating class-wide damages by presenting the jury with four categories of evidence, including: (1) Google’s prices for paid versions of similar features, (2) documents and testimony describing core features and apps, (3) documents detailing competitors’ similar offerings and prices, (4) internal Google records that reference Standard Edition features as core offerings in the paid versions of Google’s services. Although they have not done so here, plaintiffs note that they may introduce expert analysis within some of these categories. Google has not established that this evidence is unavailable or unreliable. Plaintiffs propose that the jury use this evidence to determine the per-user monthly value of the Standard Edition. Plaintiffs then propose multiplying that figure by the number of class members and the number of “user months” in the damages period to determine class-wide damages.

The Court agrees that this is a straightforward way of litigating expectation damages on a class-wide basis. Google’s primary response is that the value of the contractual benefit will vary based on when a user signed up for the Standard Edition because Google’s user agreement promised to provide class members access only to a service substantially similar to the service that was in place on the date that the user signed up. Between 2006 and 2012, Google updated its service about 662 times, and Google argues the jury will need to calculate the value of Google’s services at each of these 662 times in order to determine the value of the service to each class member at the time they enrolled.

While Google’s service evolved over time and it is possible that each time Google introduced an entirely new app—like Docs or Calendar—their service transformed into something substantially different, that does not mean that all 662 updates Google made to its service rendered the service substantially different. ^[3] Indeed, many of the service updates that Google points to were simply tweaks to existing products rather than the introduction of an entirely new app or service. See, e.g. , Dkt. No. 137-9, at 71 (“Optional attendees now in Google Calendar”); id. at 86 (“Filtering is now available in Google Spreadsheets”); id. at 96 (“You can now categorize your events using colors.”). Further, Google does not dispute that it generally made the same version of the Standard Edition available to all legacy users. The only way that Google ever differentiated the services offered to legacy users was that Google allowed earlier account holders to retain a higher number of users than newer account holders, and plaintiffs’ damages model explicitly accounts for this difference by using “user months” rather than “account months.” While the jury may need to determine the value of the service at a subset of discrete moments in time, that inquiry is not enough to defeat predominance.

B. Superiority Rule 23(b)(3)’s superiority requirement “assure[s] that the class action is the most efficient and effective means of resolving the controversy.” Wolin v. Jaguar Land Rover N. Am., LLC , 617 F.3d 1168, 1175 (9th Cir. 2010) (cleaned up). Factors relevant to assessing superiority include: (1) class members’ interest in individually controlling the litigation, (2) the presence of other similar litigation from class members, (3) the desirability of aggregating the claims in a given forum, and (4) the feasibility of managing the class action. See Fed. R. Civ. P. 23(b)(3)(A–D). “Where recovery on an individual basis would be dwarfed by the cost of litigation on an individual basis, this factor weighs in favor of class certification.” Wolin , 617 F.3d at 1175. The superiority test rests on “concern[s] for judicial economy” and thus “requires the court to determine whether maintenance of [the] litigation as a class action is efficient and … fair.” Id. at 1175–76. The putative class, as modified, satisfies the superiority requirement. Neither party has alerted the Court to any other litigation concerning Google’s termination of the Standard Edition, and there is not a clear reason why one terminated user would or should desire greater control over this litigation. Aggregating the claims in the Northern District of California makes sense given Google’s headquarters and the proximity to Silicon Valley. Aggregation is also logical given the amount of damages each plaintiff might stand to gain were they to proceed on an individual basis. Mr. Rabin pays between $17 and $50 monthly for Workspace. Even assuming he is entitled to receive a full refund of everything he has paid to Google since termination of his Standard Edition account, that figure likely amounts to the value of two to three hours of attorney time billed at prevailing hourly rates. Pursuing the claims of the modified class may therefore provide the only viable method for providing relief. Aggregation here will greatly further judicial economy. Plaintiffs have therefore satisfied Rule 23(b)(3)’s superiority requirement. III. Rule 23(b)(2) Certification of a Rule 23(b)(2) class is only “appropriate where the primary relief sought is declaratory or injunctive.” Ellis , 657 F.3d at 968. “[C]laims for monetary relief” may not be certified under that rule “at least where … the monetary relief is not incidental to the injunctive or declaratory relief.” Wal-Mart Stores , 564 U.S. at 360. Here, the primary relief sought is monetary damages or restitution, and specific performance or other injunctive relief can be addressed under Rule 23(b)(3). Certification under Rule 23(b)(2) is therefore denied.

CONCLUSION

The Court certifies the following class under Rule 23(b)(3): All persons or entities in the United States who: (a) signed up for the free version of Google Apps between August 1, 2006 and December 6, 2012; (b) were still Legacy Free customers as of January 19, 2022; (c) had at least one active user on their account during the 180 days prior to January 19, 2022; and (d) were not eligible to “opt-out”* prior to being charged for Workspace.

* Customers eligible to “opt-out” included customers who identified as non- commercial customers or were migrated to Workspace for Education Fundamentals or Workspace for Nonprofits.

Plaintiff Steve Rabin is appointed as class representative and Lieff Cabraser Heimann & Bernstein, LLP and Webb, Klase & Lemond, LLC are appointed as class counsel. Certification under Rule 23(b)(2) is denied, as is the motion to appoint plaintiff Ian Graves as class representative.

The parties shall jointly file within 30 days of this order a proposed plan to notify the certified class and provide an opportunity to opt out.

IT IS SO ORDERED. Dated: June 20, 2025 P. Casey Pitts United States District Judge

Notes

[1] Even were the Court to assume that plaintiffs’ conditioned access breach claim was properly before the Court, the class would nevertheless fail Rule 23(b)(3)’s predominance inquiry for the reasons discussed below.

[2] To the extent Google believes that distinguishing opt-out-eligible non-commercial users from 28 opt-out-ineligible commercial users creates a predominance problem, the parties can allow class

[3] In his deposition, Google’s employee Dave Girouard stated that updating a feature within the 28 Standard Edition did not necessarily convert Google’s offerings into a new service.

Case Details

Case Name: Rabin, CPA v. Google LLC

Court Name: District Court, N.D. California

Date Published: Jun 20, 2025

Citation: 5:22-cv-04547

Docket Number: 5:22-cv-04547

Court Abbreviation: N.D. Cal.