Crowd Siren V Facebook

Case 4:16-cv-06232-JSW Document 165 Filed 10/16/18 Page 1 of 28
1 Eric H. Gibbs (SBN 178658) Robert T. Eglet (pro hac vice)

Dylan Hughes (SBN 209113) Robert M. Adams (pro hac vice)
2 David Stein (SBN 257465) Erica D. Entsminger (pro hac vice)
3 Aaron Blumenthal (SBN 310605) Artemus W. Ham (pro hac vice)
GIBBS LAW GROUP LLP EGLET PRINCE
4 505 14th Street, Suite 1110 400 South Seventh Street, Suite 400
Oakland, CA 94612 Las Vegas, NV 89101
5 Telephone: (510) 350-9700 Telephone: (702) 450-5400
Facsimile: (510) 350-9701 Facsimile: (702) 450-5451
6
ehg@classlawgroup.com eservice@egletlaw.com
7 dsh@classlawgroup.com
ab@classlawgroup.com
8
Andrew N. Friedman (pro hac vice)
9 Geoffrey Graber (SBN 211547)
Eric Kafka (pro hac vice)
10
COHEN MILSTEIN SELLERS & TOLL PLLC
11 1100 New York Ave. NW, Fifth Floor
Washington, DC 20005
12 Telephone: (202) 408-4600
Facsimile: (202) 408-4699
13 afriedman@cohenmilstein.com
ggraber@cohenmilstein.com
14
ekafka@cohenmilstein.com
15 [Additional counsel on signature page]
16 Counsel for Plaintiffs and Proposed Class
17
18 UNITED STATES DISTRICT COURT FOR THE

NORTHERN DISTRICT OF CALIFORNIA
19 OAKLAND DIVISION
20 LLE ONE, LLC, d/b/a Crowd Siren and d/b/a Case Nos.: 4:16-cv-06232-JSW
21 Social Media Models, and JONATHAN FOURTH AMENDED CLASS ACTION
MURDOUGH, on behalf of themselves and all
COMPLAINT
22 others similarly situated,
23 DEMAND FOR JURY TRIAL

Plaintiffs,
24 v.
25
FACEBOOK, INC.,
26
Defendant.
27
28
FOURTH AMENDED CLASS ACTION COMPLAINT

Case No.: 4:16-cv-06232-JSW
1 INTRODUCTION
2 1. In September 2016, the Wall Street Journal revealed that, for the past two years,
3 Facebook had been overstating the average time its users spent watching paid video advertisements.
4 Based on information from advertising agencies who had spoken with Facebook, the Wall Street
5 Journal reported that Facebook’s metrics had been overstated by between 60 and 80%.
6 2. In response to the media attention, Facebook admitted it made a mistake, but
7 emphasized that it had only discovered the mistake “about a month ago,” and that “as soon as we
8 discovered [it], we fixed it.” When Plaintiffs filed this class action on behalf of advertisers, they had
9 no reason to believe Facebook was not telling the truth, and limited their claims to contract and
10 statutory claims, which have already been tested and found viable.
11 3. Internal records recently produced in this litigation suggest, however, that Facebook’s
12 action rises to the level of fraud and may warrant punitive damages. Facebook did not discover its
13 mistake one month before its public announcement. Facebook engineers knew for over a year, and
14 multiple advertisers had reported aberrant results caused by the miscalculation (such as 100%
15 average watch times for their video ads). Yet Facebook did nothing to stop its dissemination of false
16 metrics.
17 4. In addition to Facebook knowing about the problem far longer than previously
18 acknowledged, Facebook’s records also show that the impact of its miscalculation was much more
19 severe than reported. The average viewership metrics were not inflated by only 60%-80%; they were
20 inflated by some 150 to 900%.
21 5. To the extent anyone at Facebook did not fully realize how it was calculating average
22 viewership from the beginning (as Facebook’s engineers did), it was only because of Facebook’s
23 longstanding reckless indifference to the metrics’ accuracy. The wide disparity between the actual
24 average viewership and Facebook’s reported metrics should have been corrected immediately. But
25 Facebook severely understaffed the engineering team in charge of fixing errors—employing only as
26 few as two engineers—and this led to long delays before errors were fully investigated and
27 corrected. As the documents produced in this litigation show, Facebook did not take verification of
28 its metrics seriously, did not timely address known problems, and had never conducted an internal
1
1 audit of its metrics.

2 6. Even once Facebook decided to correct the false metrics, it chose not to do so
3 immediately. Instead, Facebook chose to continue disseminating false metrics for several more
4 months while it developed and deployed a “no PR” strategy designed to “obfuscate the fact that we
5 screwed up the math.” All the while, Facebook continued to reap the benefits from the inflated
6 numbers.
7 7. Based on this newly discovered information, Plaintiffs are amending their complaint
8 to add a claim for fraud and are now seeking punitive damages. Plaintiffs’ additional allegations
9 appear in this Introduction, in a New Allegations section set that begins on page 14, and in their
10 Third Cause of Action and revised Prayer, which begins on page 24. (They have also removed the
11 allegations pertaining to former plaintiff, Tyler Barnett PR, LLC.) In all other respects this Fourth
12 Amended Complaint remains unchanged from the Third Amended Complaint, and Plaintiffs are
13 continuing to pursue their previously alleged claims for violation of California’s Unfair Competition
14 Law and breach of Facebook’s contractual duty to perform with reasonable care.
15 JURISDICTION
16 8. This Court has subject matter jurisdiction over this action under 28 U.S.C. §
17 1332(d)(2) because this is a class action wherein the amount in controversy exceeds the sum or value
18 of $5,000,000, exclusive of interest and costs, there are more than 100 members in the proposed
19 class, and at least one member of the class of Plaintiffs is a citizen of a state different from a
20 Defendant.
21 9. This Court has personal jurisdiction over Defendant Facebook, Inc., because
22 Facebook, Inc., is headquartered in California, and conducts business in the state of California.
23 10. Venue is proper in this Court pursuant to 28 U.S.C. §1391(b) because a substantial
24 part of the events or omissions giving rise to the claims occurred in, were directed to, and/or
25 emanated from this District. Venue is also proper because Facebook’s terms of service require that
26 claims are resolved “exclusively in the U.S. District Court for the Northern District of California or a
27
28
2
1 state court located in San Mateo County….”1

2 PARTIES
3 11. Plaintiff LLE One, LLC, d/b/a Crowd Siren and d/b/a Social Media Models is a
4 Nevada limited liability company.
5 12. Plaintiff Jonathan Murdough is a citizen and resident of Pennsylvania.
6 13. Defendant Facebook, Inc., is incorporated in Delaware, and its principal place of
7 business is 1 Hacker Way, Menlo Park, CA 94025.
8 FACTUAL ALLEGATIONS
9 14. Facebook is a Fortune 500 company that operates social media services, including the
10 www.facebook.com and www.instagram.com websites and the WhatsApp Messenger application.
11 15. Facebook.com has 1.79 billion monthly active users.2 Users do not pay Facebook to
12 create a facebook.com account. Once a user opens a Facebook account, the account holder can, at no
13 cost, create a profile page, post content (such as photographs, videos, and links to articles), make
14 friends with other users, and view content posted by other users.
15 16. Instead of charging account holders to access facebook.com, Facebook earns most of
16 its revenue by selling advertising services. For example, in 2015, Facebook’s advertising revenue
17 was over $17.0 billion, more than 95% of its overall revenue.3
18 17. One type of advertising service Facebook sells is video advertisements, where
19 advertisers can pay money to have video displayed to Facebook users.
20 18. Facebook videos (including video ad products) autoplay by default though the audio
21 remains off unless users actively turn it on.4 Facebook allows users to scroll past autoplaying videos
22 (including video advertisements) without ever turning on the sound or watching more than a few
23 seconds of the video.
24
1
25 Facebook, Statement of Rights and Responsibilities, https://www.facebook.com/terms (last
accessed: Jan. 9, 2017).
26 2
Facebook, Company Info, http://newsroom.fb.com/company-info/ (last accessed: Jan. 9, 2017).
3
27 Facebook, Inc., Annual Report (Form 10-K) at 42 (Jan. 28, 2016).
4
Eric Blattberg, Like it or not, autoplay video won (Apr. 21, 2015),
28 http://digiday.com/publishers/autoplay-video-beat-regular-video-sorry-guys/.
3
1 19. As part of Facebook’s video advertising services, advertisers can choose to purchase
2 advertisements with specific objectives, including maximizing “Brand awareness,” “Video views,”
3 or “Conversions.” Facebook says that a “Brand awareness” campaign will “[i]ncrease awareness for
4 your brand,” whereas a “Video views” campaign will “[g]et more people to view your video
5 content,” and a “Conversions” campaign will “[d]rive valuable actions on your website or app.”
6 20. Facebook video advertising services include marketing analytics, which enable
7 purchasers to monitor and evaluate their video advertisements’ performance. Marketing analytics
8 refers to the practice of measuring and analyzing the performance of an advertising or marketing
9 campaign using data from a variety of metrics. One of the main selling points of online advertising is
10 that it offers more detailed and closer to real time marketing analytics than traditional media (such as
11 television or radio), and online advertisement sellers such as Facebook have promoted their
12 marketing analytics as a prime reason that advertisers should purchase advertisements on their
13 platforms.5 Online advertisers do so because advertisers use these analytics to determine where to
14 spend advertising dollars and the effectiveness of the dollars spent. As an article in the Harvard
15 Business Review explains, analytics allows companies to make informed decisions about how to
16 allocate their limited marketing budgets across different mediums (such as television, YouTube, or
17 Facebook), to determine which advertising campaigns to “expand” and which to “kill,” and to
18 “readily adjust or allocate advertising in different markets on a monthly, weekly, or daily basis—
19 and, online, even from one fraction of a second to the next.”6 Accordingly, it has become a standard
20 practice in the industry for online advertisers like Facebook to include marketing analytics as part of
21 their advertising services, and online advertising purchasers expect that any advertising they
22 purchase will include marketing analytics to evaluate the advertising’s performance.
23 21. The importance of analytics is apparent in the competitive online video advertising
24 market. YouTube, LinkedIn, Twitter, and Facebook all have online video advertising offerings and
25 all emphasize the value of their analytics platforms. When Facebook was first entering the online
26
27 5
See, e.g., Yahoo, Yahoo! Gemini, https://gemini.yahoo.com/advertiser/home (“Measure the impact
of your campaigns and drive better results by accessing actionable insights . . .”).
28 6
Id.
4
1 video advertising space, it knew that its analytics would be the key to its success. In a November
2 2013 private presentation to its advertising partners (“November 2013 Presentation”), Facebook
3 provided talking points on how to convince potential customers to purchase Facebook video
4 advertisements instead of other video advertisements, such as YouTube advertisements or television
5 advertisements. The presentation acknowledged that one of the weaknesses of Facebook’s video
6 advertising platform was the relatively basic level of metrics it provided to video purchasers. The
7 presentation stated, “Currently, we only report on video plays, which is a weakness compared to
8 YouTube, which reports on video views, completed views, and average duration of view. We are
9 working on building out our video insights to give advertisers a better sense for how videos are
10 performing. New video insights target launch: Q1 2014.”7
11 22. In May 2014, Facebook began providing video advertising purchasers with more
12 analytics, including video views, completed views, and average duration of view, as part of
13 Facebook’s video advertising services. Facebook told its users that the purpose of the new analytics
14 was to help video advertising purchasers “learn what’s resonating with people and determine how to
15 more effectively create and promote your videos on Facebook.”8 Facebook said its Audience
16 Retention analytics in particular, such as Average Duration of Video Viewed, would assist video
17 advertising purchasers in identifying underperforming videos and finding “the precise moment when
18 most people lost interest and stopped watching.”9 Facebook knew and admitted that “having access
19 to reliable metrics is important to the millions of partners who use our services to grow their
20 businesses.”
21 23. “Average Duration of Video Viewed,” which is the average amount of time that users
22 watched a video, is one of the most important analytics used in evaluating video advertisements’
23 performance. Average Duration of Video Viewed is a measure of “retention,” which advertisers
24
7
Josh Constine, Leaked Facebook Video Ad Pitch Deck Reveals Plans To Steal TV and YouTube
25 Dollars, TechCrunch (Dec. 13, 2013), https://techcrunch.com/2013/12/13/facebook-vs-tv-and-
youtube.
26 8
Facebook, Introducing Video Metrics (May 5, 2014),
27 https://www.facebook.com/business/news/Coming-Soon-Video-Metrics.
9
Id.
28
5
1 care about because the longer people watch an advertisement, the greater the advertisement’s impact
2 on the viewer.10 One study of video advertising campaigns on Facebook found that increasing
3 retention of a viewer from the 3-second mark to the 10-second mark in a video resulted in a 57
4 percent increase in ad recall, a 103 percent increase in brand awareness, and a 64 percent increase in
5 “purchase intent” (the intent to the purchase the advertised product).11 And because advertisers place
6 higher value upon video advertisements that are viewed for longer periods, they are willing to pay
7 more for such advertisements.
8 24. After Facebook announced and released its new video analytics platform in May
9 2014, Plaintiffs and putative class members purchased video advertising services from Facebook
10 with the understanding that video advertising analytics were included in the purchased advertising
11 services.
12 25. Upon introducing the new video metrics in their analytics platform, Facebook didn’t
13 disclose that its new video metrics were not audited or accredited by the Media Rating Council, the
14 marketing industry’s standard-bearer for accurate measurements.
15 26. As acknowledged by Facebook in its private November 2013 Presentation and public
16 statements, Facebook created and disseminated the new video analytics platform and its video
17 metrics to induce users to purchase Facebook’s video advertising services.
18 27. Among the new video metrics provided by Facebook were “Average Duration of
19 Video Viewed” and the “Average Percentage of Video Viewed.” Below is example of how these
20 video advertising metrics would appear on a user’s screen:
21
22
23
24
25
26
10
27 Facebook Business, The Value of Video for Brands (Mar. 17, 2015),
https://www.facebook.com/business/news/value-of-video.
28 11
Id.
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
7
1 28. Facebook told advertising purchasers that the “Average Duration of Video Viewed”
2 was the “total time spent watching a video divided by the total number of people who have played
3 the video”12 (as a reasonable advertising purchaser would expect). Figure 1, below, depicts how
4 Facebook defined “Average Duration of Video Viewed.”
5 Figure 1: Formula for Facebook’s “Definition” of “Average Duration of Video Viewed”
6
9 29. In August 2016, Facebook disclosed in a post in its “Advertising Help Center” that its
10 “Average Duration of Video Viewed” and “Average Percentage of Video Viewed” metrics had been
11 improperly calculated. Facebook admitted that it had erroneously “*calculated* the Average
12 Duration of Video Viewed as ‘the total time spent watching a video divided by *only* the number of
13 people who have viewed a video for three or more seconds’”13 The “Average % of Video Viewed”
14 was also erroneous because Facebook calculated that metric by using the “Average Duration of
15 Video Viewed” as a calculation input. Figure 2 below, depicts how Facebook erroneously calculated
16 “Average Duration of Video Viewed.”14
17 Figure 2: Formula for How Facebook Actually Calculated “Average Duration of Video
18 Viewed” from May 2014 to September 2016
19
20
21
22 30. On or about September 23, 2016, David Fischer, Facebook’s vice president of
23 business and marketing partnerships, admitted that, due to Facebook’s miscalculation, Facebook had
24
25 12
Facebook, How Is the “Average Duration of Video Viewed” Calculated?,
https://www.facebook.com/business/help/community/question/?id=10104227902985423 (late
26 accessed: Jan. 9, 2017).
13
27 Id.
14
Id.
28
8
1 misrepresented its “Average Duration of Video Viewed” metric. He wrote that the “average duration
2 of video viewed … metric should have reflected the total time spent watching a video divided by the
3 total number of people who played the video. But it didn’t – it reflected the total time spent watching
4 a video divided by only the number of ‘views’ of a video (that is, when the video was watched for
5 three or more seconds). And so the miscalculation overstated this metric.”15
6 31. Facebook’s “Average Duration of Video Viewed” metric was inflated because when
7 calculating the average, Facebook included the total time spent by under-3-second viewers in
8 watching the video in the numerator of the fraction, but excluded these under-3-second views from
9 the denominator of the fraction. Thus, Facebook’s “Average Duration of Video Viewed” metric
10 failed to align its criteria for eligible viewing time (the numerator) and its criteria for an eligible
11 view (the denominator).
12 32. The impact of the numerator-denominator mismatch can be seen through the
13 following illustration.
14 Figure 3: Example - How Duration Metric Should Have Been Calculated, According to
15 Facebook’s Definition, For Video Viewed By Four Users for 1, 2, 4, and 9 seconds respectively.
16
17
18 Figure 4: Example - How Duration Metric Was Actually Calculated, From May 2014 to
19 September 2016, For Video Viewed By Four Users for 1, 2, 4, and 9 seconds respectively.
20
21
22 33. Figure 3 demonstrates how Facebook should have measured average duration of
23 video viewed according to its definition. By using the correct method, all four viewers are included
24 in the denominator, yielding an average duration of video viewed of 4 seconds in this hypothetical.
25 Figure 4 demonstrates what Facebook actually did by excluding the video-watches under 3 seconds
26 from the denominator. In this hypothetical, 2 of the 4 video-watches are dropped from the
27
15
David Fischer, Facebook Video Metrics Update, Facebook.com (Sept. 23, 2016)
28 https://www.facebook.com/business/news/facebook-video-metrics-update.
9
1 denominator, causing the average duration of video viewed to incorrectly rise to 8 seconds.
2 34. By including the time spent watching the videos by users who viewed the videos for
3 less than 3 seconds, while excluding those “under 3-second” users from the denominator, the
4 Average Duration of Video Viewed metric not only failed to reflect its stated definition, it also
5 created a highly misleading result that favored Facebook.
6 35. Facebook informed some of its advertisers that it inflated the “Average Duration of
7 Video Viewed” by between 60 and 80%.16 The inflated metric thus made video advertisements
8 appear as if they were performing much better on Facebook than they actually were.
9 36. The 60 to 80% inflation in the “Average Duration of Video Viewed” metric reveals
10 that many users watch Facebook video advertisements for less than 3 seconds. When users scroll
11 right past muted video advertisements, the users have, under Facebook’s reported metric for
12 “Average Duration of Video Viewed,” technically spent time watching the video advertisement.
13 Thus, many of the 1-2 second video-watches (that contributed to the numerator, but not the
14 denominator, of Average Duration of Video Viewed) reflect users quickly scrolling past muted auto-
15 playing video advertisements.
16 37. Video advertising purchasers, including Plaintiffs, viewed the “Average Duration of
17 Video Viewed” and “Average % of Video Viewed” as important metrics because users are more
18 likely to remember a video advertisement and be affected by it if they watch a longer portion of the
19 advertisement.
20 38. Facebook’s misrepresentations induced video advertising purchasers, including
21 Plaintiffs, to purchase video advertisements because purchasers wanted accurate video advertising
22 metrics regarding “Average Duration of Video Viewed” and “Average % of Video Viewed” so that
23 they could monitor their video advertisements’ performance.
25 Plaintiffs, to continue purchasing video advertisements, and to purchase additional video
26
27
16
Suzanne Vranica and Jack Marshall, Facebook Overestimated Key Video Metric for Two Years,
28 Wall Street Journal (Sept. 22, 2016, 7:29 PM), http://www.wsj.com/articles/facebook-overestimated-
key-video-metric-for-two-years-1474586951
10
1 advertisements, because purchasers believed that users were watching their videos, on average, for
2 longer than users were actually watching their videos.
4 Plaintiffs, to pay more for Facebook video advertising than they otherwise would have been willing
5 to pay.
6 41. Facebook’s misrepresentations thereby distorted the market price for its video
7 advertising by artificially increasing the price of Facebook video advertising, causing video
8 advertising purchasers, including Plaintiffs, to pay more than they otherwise would have paid.
9 42. Facebook’s misrepresentations provided Facebook with an unfair competitive
10 advantage over other online video advertising platforms, such as YouTube, LinkedIn, and Twitter.
11 43. Facebook’s misrepresentations interfered with Plaintiffs and other putative class
12 members’ attempts to utilize Facebook’s video advertising analytics, and to run effective video
13 advertising campaigns.
14 44. Facebook’s overstated audience retention metrics resulted directly from Facebook’s
15 failure to properly audit and verify the accuracy of its video-advertising metrics before disseminating
16 them.
17 45. Any reasonable audit or verification process would have alerted Facebook earlier that
18 its metrics for “Average Duration of Video Viewed” and “Average % of Video Viewed” were
19 inaccurate and inflated.
20 46. Facebook’s audit and verification failures have been systemic and continuous, and
21 Facebook has continually refused to reform its conduct. By failing to adopt reasonably reliable
22 auditing or third party verification measures, Facebook has put Plaintiffs and other advertisers at
23 continued risk of being subjected to inaccurate analytics in the future and also made it so they will be
24 unable to rely on the accuracy of Facebook’s analytics.
25 47. In the months surrounding the announcement of the inflated and inaccurate “Average
26 Duration of Video Viewed” and “Average % of Video Viewed” metrics, Facebook’s inadequate
27 audit and verification measures caused approximately eleven additional metrics errors. The errors
28 have been so numerous and unabated that a major marketing publication posted a news article last
11
1 year to keep a running tab on the errors.17 The article begins:

2
To paraphrase J. Cole, Facebook messes up on its math one time; shame on Facebook.
3 Facebook messes up on its math a second time; shame on those of us blindly trusting
Facebook’s math. Facebook messes up on its math a third time; time to make a list.
4
To help myself and anyone else keep track of Facebook’s measurement errors, here’s an
5 itemized list of each error Facebook has announced to date….
6 Each of the 12 reported errors was discovered within a period of just over one year (from September
7 2016 through November 2017). The errors all pertain to the data and analytics that Facebook
8 provides to advertisers about the efficacy of advertising on Facebook, including the reach of
9 Facebook Page posts, video advertisement completion rate (which is a separate error from the one
10 affecting the two video advertising metrics that are the subject of this Complaint), average time spent
11 on Facebook Instant Articles, Facebook’s “referrals” metric, and video carousel ad link click
12 quantities.
13 48. Many individuals with considerable experience and knowledge about advertising on
14 Facebook have attributed the inflated and inaccurate “Average Duration of Video Viewed” and
15 “Average % of Video Viewed” metrics or the pattern of errors to Facebook’s failure to conduct
16 audits and verifications of its data and its advertising metrics.
17 49. Keith Weed, chief marketing officer of Unilever, said in an interview in 2015, that
18 allowing tech companies to block third parties from measuring their platforms is equivalent to
19 “letting them mark their own homework.”18
20 50. After the news first broke about the inflated and inaccurate “Average Duration of
21 Video Viewed” and “Average % of Video Viewed” metrics, Ad buying agency Publicis Media
22 wrote: “This once again illuminates the absolute need to have 3rd party tagging and verification on
23
24 17
Tim Peterson, FAQ: Everything Facebook has admitted about its measurement errors,
Marketing Land (May 17, 2017 12:44 p.m.) https://marketingland.com/heres-itemized-list-
25 facebooks-measurement-errors-date-200663
26
18
Suzanne Vranica and Jack Marshall, Facebook Overestimated Key Video Metric for Two
27 Years, Wall Street Journal (Sept. 22, 2016, 7:29 PM), http://www.wsj.com/articles/facebook-
overestimated-key-video-metric-for-two-years-1474586951
28
12
1 Facebook’s platform. Two years of reporting inflated performance numbers is unacceptable.”19

2 51. Also in reaction to the inflated and inaccurate “Average Duration of Video Viewed”
3 and “Average % of Video Viewed” metrics, Sir Martin Sorrell, the CEO of the world’s biggest
4 advertising group, WPP, said that the error underscored the need for a third party to oversee such
5 metrics: “We have also been calling for a long time for media owners like Facebook … not to mark
6 their own homework and release data to ComScore to enable independent evaluation. The referee
7 and player cannot be the same person.”20
8 52. The New York Times also reacted to the news of the error by writing, “Above all, it
9 reinforces the importance of outside measurement when it comes to data from Facebook and other
10 media companies.”21
11 53. Omnicom, another major advertising agency, had a similar response, saying: “This is
12 why we and many others have been repeatedly saying that third-party verification is a fundamental
13 requirement. Publishers should not ‘grade their own homework.’”22
14 54. Sarah Wood, co-chief executive of Unruly, the social marketing agency to video ad
15 tech company that is owned by News Corp, said that the error would not only “raise questions about
16 the effectiveness of Facebook as a video platform [but] it will raise more fundamental questions
17 around trustworthiness and highlights just why third-party verification is so critical . . . it’s time the
18 walled gardens came crashing down.”23
19 55. Also in response to news of the error, Fortune quoted Shafqat Islam, co-founder of
20
21
19
Id.
22 20
Eric Oster, Sir Martin Sorrell Has Harsh Words for Facebook’s Fake Data in
23 ‘Overstategate’, AgencySpy (Sept. 23, 2016 11:32 a.m.) http://www.adweek.com/agencyspy/sir-
martin-sorrell-has-harsh-words-for-facebooks-fake-data-in-overstategate/117517
24 21
John Herrman & Sapna Maheshwari, Facebook Apologizes for Overstating Video Metrics
(Sept. 23, 2016) https://www.nytimes.com/2016/09/24/business/media/facebook-apologizes-for-
25 overstating-video-metrics.html
22
26 Id.
23
Matthew Garrahan & Lauren Fedor, Ad groups seize on Facebook’s video measuring error,
27 Financial Times (Sept. 23, 2016) https://www.ft.com/content/230ec388-8169-11e6-8e50-
8ec15fb462f4
28
13
1 NewsCred: “Need 3rd party verification everywhere.”24

2 56. After Facebook continued to disclose additional errors, The Wall Street Journal
3 reported that “The lack of fully independent third-party ad verification was a major sticking point
4 following Facebook’s recent video counting mishap…”25 In the same article, it was noted that
5 Facebook was resisting letting third parties use technology to evaluate advertising efficacy. GroupM,
6 which is the ad buying unit of WPP, was quoted as saying that a transition to “full third-party
7 measurement” remained “crucial.” And the article reports that the Association of National
8 Advertisers publicly advocated for an MRC audit.
9 NEW ALLEGATIONS
10 57. Facebook’s internal documents show that Facebook personnel had been alerted to the
11 audience retention metrics mismatch long before news broke in the Wall Street Journal.
12 58. In 2015, several advertisers had inquired about “Average % of Video Viewed”
13 numbers that appeared exorbitant—some were displayed at 100%. As Facebook would put it later,
14 complaints about the inflated metrics had “been filed for a long time and we didn’t fully
15 investigate.” Other Facebook personnel acknowledged internally that one of the things they learned
16 from the fallout was to “[b]e more serious on publishers/advertisers’ reports.”
17 59. In July 2015, in an internal response to one such inquiry, a Facebook engineer
18 discussed the numerator/denominator mismatch: “I remember [another Facebook engineer]
19 mentioned when computing the average, we only consider views greater than 3second [sic], but use
20 the total watch time (including those under 3s).” This mismatch caused the inflation to the metrics
21 that Facebook ultimately disclosed in the fall of 2016. Still, Facebook did not correct the inflated
22 metrics.
23 60. In addition to overt discussion of the discrepancy, Facebook repeatedly and recklessly
24 failed to check its own work. It has long been known that Facebook resisted outside help—such as
25
26
24
Michelle Toh, Why Facebook’s Ad Buyers Are Not Happy Right Now (Sept. 23, 2016)
27 http://fortune.com/2016/09/23/facebook-video-view-metric/
25
Mike Shields, Facebook Says It Found More Miscalculated Metrics (Nov. 16, 2016 10:03
28 a.m.) https://www.wsj.com/articles/facebook-says-it-found-more-miscalculated-metrics-1479303984
14
1 through third party verification or auditing—to ensure the accuracy of the metrics, but Facebook’s
2 internal efforts behind the scenes reflect a company mentality of reckless indifference toward the
3 accuracy of its metrics.
4 61. Facebook’s internal documents show the company had never performed a full audit of
5 its video metrics.
6 62. Facebook’s management also severely understaffed the Video Insights team, which
7 was responsible for investigating and fixing errors. As of September 2016, that team had dwindled
8 to only two engineers.
9 63. In June 2016, a Facebook engineering manager finally followed up on advertiser
10 complaints dating back to early 2015, writing that “[s]omehow there was no progress on the task for
11 a year.” But even once it was decided to take action on the metrics, Facebook did not promptly fix
12 its calculation or disclose that the calculation was wrong. Instead, it continued reporting
13 miscalculated viewership metrics for another several months, as it developed a “no PR” strategy to
14 avoid drawing attention to the error. The company decided to “obfuscate the fact that we screwed up
15 the math” by quietly retiring the erroneous metrics and replacing them with corrected metrics under
16 a new name. For instance, Average Duration of Video Viewed would be replaced with Average
17 Watch Time.
18 64. In August 2016, Facebook began reaching out privately to select, large advertisers,
19 telling them that Facebook had “recently discovered a discrepancy” in the video ad average view
20 metrics. Facebook pushed that message even as personnel internally emphasized that “we didn’t
21 recently discover a discrepancy.” (emphasis added).
22 65. Facebook had successfully used a “no PR” approach before with these same metrics.
23 In January 2015, Facebook discovered it had been over-reporting the Average Duration of Video
24 Viewed and Average % Percentage of Video Viewed metrics by 40% (due to reasons separate from
25 the numerator/denominator mismatch). But rather than correct the metrics immediately in January
26 2015 and tell advertisers, Facebook recognized that “a 40% drop will be too significant and might
27 hurt user trust.” It therefore decided to do a “slow launch to fix” the metric without telling anyone:
28 “so advertisers … won’t notice significant changes.”
15
1 66. This time, the impact of Facebook’s error was much larger. Facebook chose not to
2 correct the Wall Street Journal’s report that its miscalculation had only inflated average viewership
3 metrics by between 60 and 80%. In fact, they had inflated the metrics by some 150 to 900%. For
4 example, when Facebook summarized the impact of the miscalculation internally it presented two
5 typical cases: in one, Facebook had inflated Average Duration of Video Viewed from 2.0 seconds to
6 17.5 seconds (an increase of 775%); in the other, Facebook had inflated Average Duration of Video
7 Viewed from 2.4 seconds to 17.3 seconds (an increase of 621%).
8 67. If Facebook had immediately corrected its miscalculation in a straightforward
9 manner, advertisers would have seen a sudden and precipitous drop in their viewership metrics.
10 They also would have seen that they had spent considerable money on past advertising campaigns
11 that were not enjoying anywhere near the average viewership that Facebook had represented.
12 68. Facebook did not wish to draw scrutiny to its viewership figures because it knows
13 that the majority of video ads on its platform are viewed for very short periods of time — users
14 scroll right past. If advertisers were more widely aware of this fact, and in particular, if they knew
15 that their advertisements were among those that were not drawing viewers’ attention, they would be
16 less likely to continue buying video advertising from Facebook.
17 69. Facebook’s “no PR” approach foundered in September 2016, when Facebook learned
18 that the Wall Street Journal would be publishing an article about its inflated metrics. Facebook then
19 continued to support a misleading narrative. The Journal quoted a Facebook statement claiming that
20 Facebook had only “recently discovered an error in the way we calculate one of our video metrics.”
21 Facebook reiterated that notion in Mr. Fischer’s September 23rd blog post, which stated that
22 Facebook had discovered the error “[a]bout a month ago,” and that “[a]s soon as we discovered the
23 discrepancy, we fixed it.” And of course, Facebook allowed the 60 – 80% viewership inflation
24 range to go uncorrected even though the true level of inflation had been some 10 times higher.
25 PLAINTIFFS’ EXPERIENCES
26 a. LLE One, LLC
27 70. LLE One, LLC, is a social media marketing agency.
28 71. During the period between May 4, 2014, and September 23, 2016, when Facebook
16
1 was providing inflated “Average Duration of Video Viewed” and “Average Percentage of Video
2 Viewed” metrics, LLE One, LLC, purchased video advertising services from Facebook.
3 72. Before LLE One, LLC, purchased video advertising services, LLE One, LLC, knew
4 and expected that as part of the video advertising services, Facebook provides metrics designed to
5 enable advertisers to monitor their video advertisements’ performance.
6 73. After first purchasing video advertising services, LLE One, LLC, began viewing the
7 video metrics provided by Facebook, including the “Average Duration of Video Viewed” and
8 “Average % of Video Viewed” metrics.
9 74. The inflated “Average Duration of Video Viewed” and “Average % of Video
10 Viewed” metrics were material to LLE One, LLC, and LLE One, LLC, relied on the metrics,
11 purchasing more video advertising services from Facebook than it otherwise would have.
12 75. Since September 23, 2016, LLE One, LLC, has continued to purchase Facebook
13 video advertising services and it intends to continue purchasing Facebook video advertising services.
14 b. Jonathan Murdough
15 76. Jonathan Murdough (“Murdough”) is a citizen and resident of Pennsylvania.
16 77. During the period between May 4, 2014, and September 23, 2016, when Facebook
17 was providing inflated “Average Duration of Video Viewed” and “Average Percentage of Video
18 Viewed” metrics, Murdough purchased video advertising services from Facebook.
19 78. Before Murdough purchased video advertising services for himself, he viewed video
20 metrics provided by Facebook to a company that he worked for, including the “Average Duration of
21 Video Viewed” and “Average % of Video Viewed” metrics.
22 79. Before Murdough purchased video advertising services, Murdough knew and
23 expected that as part of the video advertising services, Facebook provides metrics designed to enable
24 advertisers to monitor their video advertisements’ performance.
25 80. After purchasing video advertising services for himself, Murdough viewed video
26 metrics provided by Facebook, including the “Average Duration of Video Viewed” and “Average %
27 of Video Viewed” metrics.
28 81. The inflated “Average Duration of Video Viewed” and “Average % of Video
17
1 Viewed” metrics were material to Murdough, and Murdough relied on the metrics, purchasing more
2 video advertising services from Facebook than he otherwise would have.
3 82. Murdough intends to continue purchasing Facebook video advertising services.
4
5 CLASS ALLEGATIONS
6 83. Plaintiffs re-allege and incorporate by reference herein all of the allegations contained
7 above.
8 84. Pursuant to the Federal Rule of Civil P. 23(b)(3), Plaintiffs assert claims on behalf of
9 the following “Class”: All persons or entities who, from May 4, 2014, to September 23, 2016 (“Class
10 Period”), had an account with Facebook, Inc., and who paid for placement of video advertisements
11 on a Facebook-owned website. Excluded from the Class are Defendant, any entity in which
12 Defendant has a controlling interest, and Defendant’s officers, directors, legal representatives,
13 successors, subsidiaries, and assigns. Also excluded from the Class is any judge, justice, or judicial
14 officer presiding over this matter and the members of their immediate families and judicial staff.
15 85. This action has been brought and may properly be maintained as a class action as it
16 satisfies the numerosity, commonality, typicality, adequacy, and superiority requirements of Rule
17 23(b)(3). Plaintiffs seek to represent an ascertainable Class, as determining inclusion in the class can
18 be done through Facebook’s own records.
19 86. Plaintiffs reserve the right to amend the Class definition if discovery and further
20 investigation reveal that the Class should be expanded, divided into subclasses, or modified in any
21 other way.
22 87. Although the precise number of Class members is unknown and can only be
23 determined through appropriate discovery, Plaintiffs believe, and on that basis allege, that the
24 proposed Class is so numerous that joinder of all members would be impracticable as Facebook sells
25 millions of advertisements annually, and a significant portion of those advertisements are video
26 advertisements.
27 88. Questions of law and fact common to the putative Class exist that predominate over
28 questions affecting only individual members, including inter alia:
18
1 a. Whether Facebook made material misrepresentations about its video advertising

2 services, including misrepresenting the “Average Duration of Video Viewed” and
3 “Average % of Video Viewed” metrics;
4 b. Whether Facebook’s use of inaccurate, unaudited, and unverified video metrics was
5 likely to deceive members of the public and thus constituted a fraudulent business
6 practice under California’s Unfair Competition Law (Cal. Bus. & Prof. Code §
7 17200);
8 c. Whether Facebook’s failure to properly audit and verify its video metrics was
9 unethical, unscrupulous, or substantially injurious to video advertising purchasers and
10 thus constituted an unfair business practice under California’s Unfair Competition
11 Law; and
12 d. Whether Facebook breached its contractual duty to perform competently and with
13 reasonable care by reporting erroneous video metrics.
14 89. Plaintiffs are members of the putative Class. The claims asserted by the Plaintiffs in
15 this action are typical of the claims of the members of the putative Class, as the claims arise from the
16 same course of conduct by the Defendant and the relief sought is common.
17 90. Plaintiffs will fairly and adequately represent and protect the interests of the members
18 of the putative Class, as their interests are coincident with, not antagonistic to, the other members of
19 the Class.
20 91. Plaintiffs have retained counsel competent and experienced in both consumer
21 protection and class action litigation. Plaintiffs’ counsel specifically has experience litigating some
22 of the largest and most complex consumer class actions, including numerous consumer class actions
23 in the Northern District of California.
24 92. Certification of the Class is appropriate pursuant to Fed. R. C. P. 23(b)(3) because
25 questions of law or fact common to the respective members of the Class predominate over questions
26 of law or fact affecting only individual members. This predominance makes class litigation superior
27 to any other method available for the fair and efficient adjudication of these claims including
28 consistency of adjudications. Absent a class action it would be highly unlikely that the members of
19
1 the Class would be able to protect their own interests because the cost of litigation through
2 individual lawsuits might exceed the expected recovery.
3 93. A class action is a superior method for the adjudication of the controversy in that it
4 will permit a large number of claims to be resolved in a single forum simultaneously, efficiently, and
5 without the unnecessary hardship that would result from the prosecution of numerous individual
6 actions and the duplication of discovery, effort, expense, and the burden of the courts that individual
7 actions would create.
8 94. In the alternative, the Class should be certified because:
9 a. The prosecution of separate actions by the individual members of the proposed class
10 would create a risk of inconsistent adjudications, which could establish incompatible
11 standards of conduct for Facebook;
12 b. The prosecution of individual actions could result in adjudications, which as a
13 practical matter, would be dispositive of the interests of non-party class members or
14 which would substantially impair their ability to protect their interests; and
15 c. Facebook has acted or refused to act on grounds generally applicable to the proposed
16 Class, thereby making appropriate final and injunctive relief with respect to the
17 members of the proposed Class as a whole.
18
FIRST CAUSE OF ACTION
19 (On behalf of the Class)26
CALIFORNIA UNFAIR COMPETITION LAW,
20 CAL. BUS. & PROF. CODE § 17200, et seq.
22 above.
23 96. Facebook violated California’s Unfair Competition Law (UCL), Cal. Bus. & Prof.
24 Code §17200 et seq., by engaging in the fraudulent and unfair business acts and practices alleged
25 previously, and as further specified below.
26
27 26
Pursuant to Facebook’s terms of service, the laws of the State of California govern “any claim”
that arises between Facebook and its users, “without regard to conflict of law provisions.” Facebook,
28 Statement of Rights and Responsibilities, https://www.facebook.com/terms (last accessed: Jan. 9,
2017).
20
1 97. Facebook’s dissemination of inaccurate and inflated video-advertising metrics

2 constitutes a fraudulent practice under the UCL, as it is likely to deceive Class members into
3 believing that their video advertisements were viewed, on an average, for a longer duration than they
4 actually were viewed.
5 98. Facebook’s failure to properly audit and verify the accuracy of its video-advertising
6 metrics before disseminating them to Class members is unethical, unscrupulous, and substantially
7 injurious to video-advertising purchasers, and thus constitutes an unfair practice under the UCL.
8 Facebook’s practice was also contrary to legislatively declared and public policies that seek to
9 protect consumers from misleading statements, as reflected by laws like the Federal Trade
10 Commission Act (15 U.S.C. § 45), Consumers Legal Remedies Act (Cal. Civ. Code § 1750 et seq.),
11 and California False Advertising Law (Cal Bus. & Prof. Code § 17500). The harm these practices
12 caused to Plaintiffs and the Class members outweigh their utility, if any.
13 99. Facebook should have known that its metrics for “Average Duration of Video
14 Viewed” and “Average % of Video Viewed” were inaccurate and inflated, and had Facebook
15 properly audited and verified its video-advertising metrics it would have known that those metrics
16 were inaccurate and inflated. The calculation errors that Facebook allowed to persist for over two
17 years were obvious errors that would have been discovered by a reasonable auditing and verification
18 process.
19 100. Facebook’s failure to employ reasonable auditing and verification procedures gave it
20 an unfair competitive advantage, as it allowed Facebook to provide video-advertising services at a
21 lower cost and made those advertising services appear to be more effective than they were.
22 101. Plaintiffs have standing to bring these claims under the UCL because they were
23 injured and lost money or property, including but not limited to money paid for Facebook video
24 advertisements, as a result of Facebook’s fraudulent and unfair business practices. Among other
25 things, Plaintiffs would not have bought as much video-advertising services if Facebook had not
26 disseminated inflated metrics and would have paid a lower price for the video-advertising services
27 they did purchase.
28 102. Pursuant to Cal. Bus. & Prof. Code § 17203, Plaintiffs seek equitable relief to prevent
21
1 the continued use of Facebook’s unfair and fraudulent practices and to restore to the Class all money
2 Facebook may have acquired by means of its fraudulent and unfair business practices.
3
4 SECOND CAUSE OF ACTION

(On behalf of the Class)
5 BREACH OF IMPLIED DUTY TO PERFORM WITH REASONABLE CARE
7 above.
8 104. Plaintiffs and Class members contracted with Facebook to provide them with video-
9 advertising services. They did so through one or more of Facebook’s advertising interfaces, where
10 Class members can submit video advertisements for approval, set a daily advertising budget that can
11 be adjusted at any time, set or adjust their target audience, and set or adjust the placement of the
12 advertising.
13 105. Plaintiffs and Class members met all or substantially all of their contractual
14 obligations, including submitting their advertising for Facebook’s approval and paying for
15 Facebook’s advertising services.
16 106. There is not one integrated contract that spells out Facebook’s obligations to the
17 Class, but those obligations can be determined by reference to Facebook’s course of dealing with the
18 Class, industry practice, and from various webpages created by Facebook, including Facebook’s
19 Statement of Rights and Responsibilities (https://www.facebook.com/legal/terms); Facebook’s Self-
20 Serve Ad Terms (https://www.facebook.com/legal/self_service_ads_terms); Facebook’s Advertising
21 Guidelines (https://www.facebook.com/policies/ads); Facebook’s Advertiser Help Center
22 (https://www.facebook.com/business/help); Facebook’s Community Payments Terms
23 (https://www.facebook.com/payments_terms); and Facebook’s ad-creation, editing, and management
24 interfaces, which may be accessed through Facebook’s Ads Manager
25 (https://www.facebook.com/ads/manager/creation/), Facebook’s Power Editor
26 (https://www.facebook.com/ads/manage/powereditor), Facebook’s Ads Manager App, Facebook
27 Pages, and Facebook Ads API.
28 107. One of Facebook’s obligations is to provide Class members with an advertising
22
1 interface: a central portal where Class members can create, edit, and monitor advertisements for
2 placement on Facebook’s websites. This obligation is implied by Facebook’s course of dealing with
3 the Class, industry practice, Facebook’s Self-Serve Ad Terms, and Facebook’s Advertising Help
4 Center. For example, the Self-Serve Ad Terms were drafted by Facebook to “apply to [Class
5 members’] use of the self-service advertising interfaces and APIs … (collectively, the ‘Self-Serve
6 Ad Interfaces’),” and those terms state that “you can cancel your Order at any time through our
7 online portal”—the implication being that there would in fact be advertising interfaces for Class
8 members to use. Facebook’s Advertiser Help Center also refers to Facebook’s advertising interfaces
9 on several occasions.
10 108. Another of Facebook’s obligations is to track the advertising it places for Plaintiffs
11 and Class members and report metrics by which Plaintiffs and Class members can evaluate the
12 performance of their advertisements. This obligation is implied by Facebook’s course of dealing
13 with the Class, industry practice, statements made by Facebook in the Advertiser Help Center and in
14 standardized communications with the Class. For example, Facebook has stated in its Advertiser
15 Help Center: “You can always go to Ads Manager to see how your ads are performing,” and “When
16 you create a Video Views campaign, we’ll measure both impressions and engagement with your
17 videos.” And when Facebook approves an advertisement, it sends a standardized email that states:
18 “Your ad is approved and should begin delivering shortly. Click the ad name below to manage it or
19 view its performance.”
20 109. Under California law, Facebook was required to perform its contractual obligations
21 competently and with reasonable care. Facebook breached that duty by incorrectly measuring
22 viewer engagement with the advertising it placed for Plaintiffs and Class members; by including
23 inaccurate data in the advertising interface that it provided to Plaintiffs and Class members; and by
24 reporting inaccurate advertising-performance metrics to Plaintiffs and Class members.
25 110. Had Facebook used reasonable care, it would have calculated the video-viewing
26 metrics correctly, it would have discovered any calculation errors through routine auditing and
27 verification, and it would not have used an incorrect formula for over two years.
28 111. As a result of Facebook’s failure to provide its agreed advertising services
23
1 competently and using reasonable care, Plaintiffs and Class members purchased advertising services
2 they would not otherwise have purchased and failed to receive the benefit of their bargain. They are
3 entitled to damages in an amount to be proven at trial.
4 112. On behalf of the proposed Class, Plaintiffs have notified Facebook that it is in breach
5 and given it an opportunity to remedy the harm suffered by the Class, but Facebook has refused.
6 THIRD CAUSE OF ACTION
(On behalf of the Class)
7 FRAUD
9 above.
10 114. Facebook falsely represented the average time that Facebook users spent watching
11 Plaintiffs’ and Class members’ video advertisements. The Average Duration of Video Viewed and
12 Average Percentage of Video Viewed metrics that Facebook reported to Plaintiffs and Class
13 members was typically inflated by between 150% and 900%.
14 115. Facebook either knew that the average viewership metrics it was reporting to
15 Plaintiffs and Class members was false or reported those metrics recklessly and without regard for
16 their truth. Facebook engineers knew about the miscalculation by July 2015. Facebook received
17 numerous reports from advertisers (such as nonsensical 100% viewership rates) over the course of
18 many months, and still continued to report false metrics. And even after Facebook ultimately
19 decided to correct its metrics, it still knowingly disseminated false metrics for several additional
20 months. The persistence of Facebook’s false metrics was possible only because Facebook did not
21 take verification of its metrics seriously, severely understaffed the engineering team in charge of
22 fixing errors, did not fully investigate or correct errors that were reported to it, and refused to allow
23 third-party verification of its metrics.
24 116. Facebook intended that Plaintiffs and Class members rely on its average viewership
25 metrics. Facebook promoted the metrics to users as useful tools to identify which video
26 advertisements were effective; Facebook knew that the average viewership metrics were considered
27 key metrics by advertisers; and Facebook concealed the fact that its recklessness had caused the
28 metrics to be inflated because it knew that would hurt user trust and result in advertisers purchasing
24
1 fewer video ads.

2 117. Plaintiffs and Class members did rely on Facebook’s inflated viewership metrics
3 when deciding whether and how to purchase video advertising from Facebook. As a result of the
4 inflated metrics, Plaintiffs and Class members purchased more video advertising from Facebook
5 than they otherwise would have and paid a higher price than they otherwise would have.
6 118. Plaintiffs seek an award of compensatory and punitive damages. Facebook’s conduct
7 as previously described constitutes oppression, fraud, or malice, and was authorized or ratified by
8 Facebook’s officers.
9 PRAYER FOR RELIEF
10 WHEREFORE, Plaintiffs LLE One, LLC, d/b/a Crowd Siren and d/b/a Social Media Models,
11 and Jonathan Murdough, on behalf of themselves and the Class, seek the following relief:
12 A. An order certifying this action as a class action under Fed. R. Civ. P. 23, defining the
13 Class as requested herein, appointing Cohen Milstein Sellers & Toll PLLC, Gibbs Law Group LLP,
14 and Eglet Prince as Class Counsel, and finding that Plaintiffs are proper representatives of the Class
15 requested herein.
16 B. Plaintiffs request injunctive relief. Awarding injunctive and other equitable relief as
17 is necessary to protect the interests of the Class, including: (i) an order prohibiting Facebook from
18 engaging in the wrongful acts described herein; (ii) requiring Facebook to engage third-party
19 auditors to conduct audits and evaluations of Facebook’s advertising metrics on a periodic basis and
20 ordering them to promptly correct any problems or issues detected by these auditors, and (iii)
21 requiring Facebook to disclose any further inaccurate advertising metrics in a timely and accurate
22 manner.
23 C. Plaintiffs also request damages (including punitive damages), restitution, attorneys’
24 fees, statutory costs, and such other and further relief as is just and proper. Plaintiffs seek attorneys’
25 fees under California Code of Civil Procedure 1021.5.
26 JURY TRIAL DEMAND
27 Plaintiffs hereby request a jury trial for all issues so triable.
28
25
1 DATED: August 3, 2018 Respectfully submitted,

2
By: /s/ Eric Gibbs
3
4 Eric Gibbs
Dylan Hughes
5 David Stein
Aaron Blumenthal
6
GIBBS LAW GROUP LLP
7 505 14th Street, Suite 1110
Oakland, CA 94612
8 Telephone: (510) 350-9700
Facsimile: (510) 350-9701
9 ehg@classlawgroup.com
10 dsh@classlawgroup.com
ds@classlawgroup.com
11 ab@classlawgroup.com
12 Andrew N. Friedman
Geoffrey Graber
13 Eric Kafka
14 COHEN MILSTEIN SELLERS & TOLL PLLC
1100 New York Ave. NW, Fifth Floor
15 Washington, DC 20005
Telephone: (202) 408-4600
16 Facsimile: (202) 408-4699
afriedman@cohenmilstein.com
17 ggraber@cohenmilstein.com
18 ekafka@cohenmilstein.com
19 Michael Eisenkraft
COHEN MILSTEIN SELLERS & TOLL PLLC
20 88 Pine Street, 14th Floor
New York, NY 10005
21
Telephone: (212) 838-7797
22 Facsimile: (212) 838-7745
meisenkraft@cohenmilstein.com
23
Robert T. Eglet
24 Robert M. Adams
Erica D. Entsminger
25
Artemus W. Ham
26 EGLET PRINCE
400 South Seventh Street, Suite 400
27 Las Vegs, NV 89101
Telephone: (702) 450-5400
28 Facsimile: (702) 450-5451
26
1 eservice@egletlaw.com
2
Aisha Christian (pro hac vice forthcoming)
3 129 West 27th Street, 11th Floor
New York, NY 10001
4 Telephone: (646) 285-2029
5 Charles Reichmann (SBN 206699)

LAW OFFICES OF CHARLES REICHMANN
6
16 Yale Circle
7 Kensington, CA 94708-1015
Telephone: (415) 373-8849
8 charles.reichmann@gmail.com
9 Joseph A. Motta (SBN 133531)
RUEB & MOTTA
10
1401 Willow Pass Road, Suite 880
11 Concord, CA 94520
Telephone: (925) 602-3400
12 Facsimile: (925) 602-0622
joe@rmmprolaw.com
13
14 Counsel for Plaintiffs and Proposed Class
15
16
17
18
19
20
21
22
23
24
25
26
27
28
27

Crowd Siren V Facebook

Uploaded by

Copyright:

Available Formats

Crowd Siren V Facebook

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Crowd Siren V Facebook

Uploaded by

Copyright:

Available Formats

Case 4:16-cv-06232-JSW Document 165 Filed 10/16/18 Page 1 of 28

1 Eric H. Gibbs (SBN 178658) Robert T. Eglet (pro hac vice)

18 UNITED STATES DISTRICT COURT FOR THE

23 DEMAND FOR JURY TRIAL

FOURTH AMENDED CLASS ACTION COMPLAINT

1 audit of its metrics.

1 state court located in San Mateo County….”1

1 year to keep a running tab on the errors.17 The article begins:

1 Facebook’s platform. Two years of reporting inflated performance numbers is unacceptable.”19

1 NewsCred: “Need 3rd party verification everywhere.”24

1 a. Whether Facebook made material misrepresentations about its video advertising

1 97. Facebook’s dissemination of inaccurate and inflated video-advertising metrics

4 SECOND CAUSE OF ACTION

15 Facebook’s advertising services.

19 Statement of Rights and Responsibilities (https://www.facebook.com/legal/terms); Facebook’s Self-

20 Serve Ad Terms (https://www.facebook.com/legal/self_service_ads_terms); Facebook’s Advertising

21 Guidelines (https://www.facebook.com/policies/ads); Facebook’s Advertiser Help Center

22 (https://www.facebook.com/business/help); Facebook’s Community Payments Terms

23 (https://www.facebook.com/payments_terms); and Facebook’s ad-creation, editing, and management

24 interfaces, which may be accessed through Facebook’s Ads Manager

25 (https://www.facebook.com/ads/manager/creation/), Facebook’s Power Editor

26 (https://www.facebook.com/ads/manage/powereditor), Facebook’s Ads Manager App, Facebook

27 Pages, and Facebook Ads API.

28 107. One of Facebook’s obligations is to provide Class members with an advertising

1 fewer video ads.

1 DATED: August 3, 2018 Respectfully submitted,

5 Charles Reichmann (SBN 206699)

14 Counsel for Plaintiffs and Proposed Class

You might also like