Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 1 of 33
`
`Andrew Friedman (pro hac vice)
`Geoffrey Graber (SBN 211547)
`Julia Horwitz (pro hac vice)
`Karina G. Puttieva (SBN 317702)
`COHEN MILSTEIN SELLERS & TOLL PLLC
`1100 New York Ave. NW, Fifth Floor
`Washington, DC 2000
`Telephone: (202) 408-4600
`Facsimile: (202) 408-4699
`afriedman@cohenmilstein.com
`ggraber@cohenmilstein.com
`jhorwitz@cohenmilstein.com
`kputtieva@cohenmilstein.com
`
`Eric Kafka (pro hac vice)
`COHEN MILSTEIN SELLERS & TOLL PLLC
`88 Pine Street, 14th Floor
`New York, NY 10005
`Telephone: (212) 838-7797
`Facsimile: (212) 838-7745
`ekafka@cohenmilstein.com
`
`Counsel for Plaintiffs and
`Proposed Class
`
`[Additional counsel on signature page]
`
`UNITED STATES DISTRICT COURT FOR THE
`NORTHERN DISTRICT OF CALIFORNIA
`
`DZ Reserve and Cain Maxwell (d/b/a Max
`Martialis) individually and on behalf of others
`similarly situated,
`
`Case No.: 3:18-cv-04978
`
`THIRD AMENDED CONSOLIDATED
`CLASS ACTION COMPLAINT
`
`Plaintiffs,
`
`DEMAND FOR JURY TRIAL
`
`v.
`
`FACEBOOK, INC.,
`
`Defendant.
`
`CLASS ACTION
`
`REDACTED VERSION OF
`DOCUMENT SOUGHT TO BE
`SEALED
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 2 of 33
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`Plaintiffs DZ Reserve and Cain Maxwell (d/b/a Max Martialis), individually and on behalf all
`
`others similarly situated, hereby file suit against Facebook, Inc., and allege the following:
`
`INTRODUCTION
`
`1.
`
`Facebook, Inc. (“Facebook”) is a social media company that “generate[s] substantially
`
`all of [its] revenue from advertising.”1 In 2017, Facebook earned approximately $40 billion from
`
`advertising revenue.2
`
`2.
`
`The core of Facebook’s business is its large purported user base, which ostensibly
`
`enables advertisements placed on Facebook.com to reach a large number of people. Facebook claims
`
`to have 2.13 billion monthly active users globally, with over 240 million monthly active users in the
`
`U.S. alone.3
`
`3.
`
`Before advertisers make a purchase, Facebook represents that their advertisements can
`
`potentially reach a specified number of people (“Potential Reach”). Facebook defines “potential reach”
`
`as “an estimation of how many people are in an ad set’s target audience.”4 Depending on the
`
`demographic targeting selected by the advertiser, the Potential Reach is often millions of people.
`
`Facebook also represents that the advertisement will reach an estimated number of people per day
`
`(“Estimated Daily Reach”). The Estimated Daily Reach is based, in part, on the audience size or
`
`Potential Reach. According to Facebook, Estimated Daily Reach “gives you an idea of how many of
`
`the people in your target audience [or Potential Reach] you may be able to reach on a given day.”5
`
`4.
`
`These foundational representations are false. Based on publicly available research and
`
`Plaintiffs’ own analysis, Facebook overstates the Potential Reach of its advertisements. For example,
`
`based on publicly available data, Facebook’s purported Potential Reach among the key 18-34 year-old
`
`demographic in every state exceeds the actual population of 18-34 year-olds. Based on a combination
`
`
`1 Facebook, Inc., Annual Report (Form 10-K) for 2017 Fiscal Year, at 41 (Feb. 1, 2018).
`2 Id., at 43 (Feb. 1, 2018).
`3 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018)
`4 See “About potential Reach”, available at https://www.facebook.com/adsmanager/creation
`5 See What are “Potential Reach” and “Estimated Daily Reach?”, available at
`https://www.facebook.com/business/help/717368264947302/?ref=u2u (last accessed August 1, 2018)
`
`1
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 3 of 33
`
`
`
`of publicly available research and Plaintiffs’ own analysis, among 18-34-year-olds in Chicago, for
`
`example, Facebook asserted its Potential Reach was approximately 4 times (400%) higher than the
`
`number of real 18-34 year-olds with Facebook accounts in Chicago. Based on a combination of
`
`publicly available research and Plaintiffs’ own analysis, Facebook’s asserted Potential Reach in Kansas
`
`City was approximately 200% higher than the number of actual 18-54 year-olds with Facebook
`
`accounts in Kansas City. This inflation is apparent in other age categories as well.
`
`5.
`
`Documents produced by Facebook in this litigation confirm that senior executives at the
`
`company knew for years that its Potential Reach metric was inflated – yet they failed to do anything,
`
`and even took steps to cover up the problem. As explained below, in late 2017 and throughout 2018,
`
`Facebook executives repeatedly acknowledged Potential Reach was inflated and misleading due to,
`
`among other reasons, the fact that Potential Reach includes duplicate and fake accounts. Facebook
`
`failed to fix the problem, noting that
`
` Yet, as the product
`
`manager for Potential Reach put it:
`
`
`
` Facebook’s conduct was – and remains – inexcusable. Because Facebook inflated its
`
`Potential Reach, Plaintiffs and putative class members purchased more advertisements from Facebook
`
`and paid a higher price for advertisements than they otherwise would have. Plaintiffs and putative class
`
`members accordingly seek compensation and injunctive relief for violations of California’s UCL,
`
`breach of the covenant of good faith and fair dealing, fraudulent misrepresentation, fraudulent
`
`concealment, and for restitution and punitive damages.
`
`JURISDICTION
`
`6.
`
`This Court has subject matter jurisdiction over this action under 28 U.S.C. § 1332(d)(2)
`
`because this is a class action wherein the amount in controversy exceeds the sum or value of
`
`$5,000,000, exclusive of interest and costs, there are more than 100 members in the proposed class,
`
`and at least one member of the class of Plaintiffs is a citizen of a state different from a Defendant.
`
`7.
`
`This Court has personal jurisdiction over Defendant Facebook, Inc., because Facebook,
`
`Inc., is headquartered in California, and conducts business in the state of California.
`
`8.
`
`Venue is proper in this Court pursuant to 28 U.S.C. §1391(b) because a substantial part
`
`of the events or omissions giving rise to the claims occurred in, were directed to, and/or emanated from
`
`2
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 4 of 33
`
`
`
`this District. Venue is also proper because Facebook’s terms of service require that claims be resolved
`
`“exclusively in the U.S. District Court for the Northern District of California or a state court located in
`
`San Mateo County….”6
`
`PARTIES
`
`9.
`
`Plaintiff DZ Reserve is a company incorporated and headquartered in the state of
`
`Colorado.
`
`10.
`
`Plaintiff Cain Maxwell (d/b/a Max Martialis) is a citizen and resident of the state of
`
`Ohio.
`
`11.
`
`Defendant Facebook, Inc., is incorporated in Delaware, and its principal place of
`
`business is 1 Hacker Way, Menlo Park, CA 94025.
`
`FACTUAL ALLEGATIONS
`
`12.
`
`Facebook is one of the largest social media companies in the world. It owns and operates
`
`Facebook.com, as well as Instagram and the WhatsApp Messenger service.
`
`13.
`
`In 2017, Facebook saw its revenues from online advertisements reach approximately
`
`$40 billion.7
`
`14.
`
`In marketing its online advertisement services, Facebook claims that more than 2 billion
`
`people use Facebook every month.8 Additionally, in its Ads Manager, Facebook has claimed that 240
`
`million of those active users are located in the United States.9
`
`15.
`
`Until approximately March 12, 2019, Facebook’s Potential Reach was based, at least in
`
`part, on estimates of the number of people who were active users in the past month.10 After March 12,
`
`
`6 Facebook, Statement of Rights and Responsibilities, https://www.facebook.com/terms (last
`accessed: June 17, 2019).
`7 Facebook, Inc., Annual Report (Form 10-K) for 2017 Fiscal Year, at 41 (Feb. 1, 2018), at 43.
`8 Facebook, Inc., Annual Report (Form 10-K) for 2017 Fiscal Year, at 41 (Feb. 1, 2018), at 34.
`9 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018).
`10 See Updates to Potential Reach, available at
`https://www.facebook.com/business/help/567031670465069?helpref=faq_content (last accessed June
`14, 2019)
`
`
`
`3
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 5 of 33
`
`
`
`2019, Potential Reach is based, at least in part, on how many people have been shown an ad on a
`
`Facebook Product in the past 30 days who match the respective advertiser’s desired audience and
`
`placement criteria.11
`
`16.
`
`Facebook defines a Monthly Active User (MAU) as a “registered Facebook user who
`
`logged in and visited Facebook through [Facebook’s] website or a mobile device, or used [Facebook’s]
`
`Messenger application (and is also a registered Facebook user), in the last 30 days as of the date of
`
`measurement. MAUs are a measure of the size of our global active user community.”12
`
`17.
`
`Audience size is an important factor when advertisers determine where to spend
`
`marketing dollars. Indeed, Potential Reach and Estimated Daily Reach are the only information
`
`Facebook provides to advertisers regarding the anticipated performance of the ad campaign prior
`
`purchasing an advertisement. Moreover, user inflation can skew an advertiser’s decision making,
`
`which is frequently based on the anticipated reach of the advertising campaign, or “Potential Reach.”13
`
`18.
`
`Reach inflation can have “real consequences for an advertiser’s overall communications
`
`plain.”14 For advertisers, “Facebook Ads Manager functions as a tool for an advertiser to plan, budget,
`
`buy and optimize their own campaigns across Facebook platforms.”15
`
`19.
`
`Facebook’s large purported potential reach is widely acknowledged as one of the main
`
`reasons that advertisers choose to purchase advertisements from Facebook.
`
`20.
`
`For example, an online advertising commentator recently noted that the large number of
`
`people that can potentially be reached on Facebook is one of the four “reasons why you can’t ignore
`
`Facebook advertising.”16
`
`
`
`11 Id.
`12 Facebook, Inc., Annual Report (Form 10-K) for 2017 Fiscal Year, at 41 (Feb. 1, 2018), at 36
`13 “Facebook Audience Inflation a Global Issue-Ad News Study”, AdNews (September 8, 2017)
`available at http://www.adnews.com.au/news/facebook-audience-inflation-a-global-issue-adnews-
`study (visited July 16, 2018).
`14 Video Advertising Bureau, “Facebook’s Reach (on Reach), Miscalculations in the Age of
`Precision”, p. 19, September 2017, available at https://www.thevab.com/wp-content/uploads/2017
`/09/Facebooks-Reach.pdf (accessed August 6, 2018).
`15 Id.
`16 “4 Reasons Why You Can’t Ignore Facebook Advertising,” InstaPage (September 10, 2018)
`
`
`
`4
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 6 of 33
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`21.
`
`Similarly, Aleksandar Radonjic, the Founder and Chief Growth Strategist at Evolving
`
`Digital (a digital marketing agency) explained in his blog post, “Why Facebook Has To Be Included in
`
`Your Marketing Plan” that “big brands understand the value of Facebook and are actively engaged with
`
`their customers. Small and medium-sized organizations have to do the same, even if you are a small
`
`local business because the potential reach is unmatched by any other social media platform.” (emphasis
`
`added).17
`
`22. Moreover, Facebook itself encourages advertisers to refer to the Potential Reach after
`
`they have started an advertising campaign to determine the percentage of their target audience that they
`
`have reached.18
`
`I.
`
`Facebook’s Representations to Advertisers
`
`23.
`
`In describing the advantages of advertising on its website, Facebook tells advertisers
`
`that “[t]wo billion people use Facebook every month.”19
`
`24.
`
`Facebook touts the reach of its platform, telling advertisers “Show your ads to more
`
`people in more places. Improve your reach by running your ads across Facebook, Instagram, and
`
`Audience Network.”20
`
`25.
`
`Facebook makes these representations when advertisers purchase advertisements from
`
`Facebook, which they do through Facebook’s “Ads Manager.”
`
`26.
`
`27.
`
`The terms on the “Ads Manager” are part of Facebook’s contract with advertisers.
`
`The “Ads Manager” provides only two data points related to audience size, “Potential
`
`Reach” and “Estimated Daily Results Reach.” The Estimated Daily Results Reach is derived in part
`
`from the Potential Reach audience size. Facebook’s Estimated Daily Reach “gives you an idea of how
`
`
`https://instapage.com/blog/4-reasons-facebook-advertising-cant-be-ignored (last accessed December
`19, 2018).
`17 “Why Facebook Has to Be Included in Your Marketing Plan,” Aleksandar Radonjic,
`http://www.evolving-digital.com/resources/Facebook-Marketing (last accessed December 19, 2018).
`18 https://www.facebook.com/business/help/1639908612985580 (last accessed December 19,
`2018).
`19 Quoted from https://www.facebook.com/business/products/ads (accessed August 13, 2018).
`20 Id.
`
`
`
`5
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 7 of 33
`
`
`
`many of the people in your target audience [or Potential Reach] you may be able to reach on a given
`
`day.”21
`
`28.
`
`The Potential Reach and the Estimated Daily Results Reach are provided to every
`
`advertiser who purchases advertisements through Ads Manager, and are part of Facebook’s contract
`
`with advertisers.
`
`29.
`
` On the “Ads Manager,” advertisers can target their advertisement to users of different
`
`demographics. See Figure 1. Using the demographic criteria on “Ads Manager,” advertisers can target
`
`people in various locations, age ranges, or genders.
`
`Figure 1
`
`30.
`
`Regardless of the demographic criteria used to target the advertisements, the “Ads
`
`Manger” shows a graphic with the audience size based on the criteria that advertisers selected. See
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`Figure 2.
`
`25
`
`26
`
`27
`
`28
`
`
`
`
`
`
`21 See What are “Potential Reach” and “Estimated Daily Reach?”, available at
`https://www.facebook.com/business/help/717368264947302/?ref=u2u (last accessed August 1, 2018)
`
`6
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 8 of 33
`
`
`
`
`
`Figure 2
`
`
`
`
`
`
`
`
`
`
`
`
`
`31.
`
`Below the audience size graphic, the “Ads Manager” states “Potential Reach: ________
`
`people.” See Figure 2. The “Ads Manager” displays the “Potential Reach” statistic before the advertiser
`
`purchasers the advertisement.
`
`32.
`
`Facebook consistently represents that the Potential Reach is a measurement of actual
`
`people, not just accounts.
`
`7
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 9 of 33
`
`
`
`33.
`
`The Ads Manager graphic displaying the Potential Reach and Estimated Daily Results
`
`Reach is shown on multiple consecutive pages on Ads Manager where advertisers input information
`
`before purchasing an advertisement. See Exhibit 1, Facebook’s Ads Manager.22 The Potential Reach
`
`and Estimated Daily Reach are included on the same page where advertisers set the budget and schedule
`
`for their advertisements:
`
`34. When an advertiser clicks on the “i” icon next to the Potential Reach, Facebook’s
`
`website states, “Estimates are based on the placements and targeting criteria you select and include
`
`
`22 Exhibit 1 to Plaintiffs’ Second Amended Consolidated Class Action Complaint is identical to
`ECF No. 66-3 (Exhibit C to Facebook’s Motion to Dismiss Consolidated Class Action Complaint).
`
`8
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 10 of 33
`
`
`
`factors like Facebook user behaviors, user demographics and location data. They’re designed to
`
`estimate how many people in a given area could see an ad a business might run. They’re not designed
`
`to match population or census estimates. Numbers may vary due to performance reasons.” (emphasis
`
`added) See Figure 3.
`
`35.
`
`Yet, Facebook fails to disclose to advertisers that its Potential Reach statistic is inflated.
`
`
`
`Figure 3
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`36. When an advertiser clicks on “Learn More” for potential reach, Facebook’s website
`
`states, “Potential reach is an estimation of how many people are in an ad set’s target audience…. It
`
`updates in real time as you create or edit your ad set to help you understand how your targeting and
`
`placement choices affect the number of people you could reach.” (emphasis added) See Figure 4.
`
`9
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 11 of 33
`
`
`
`
`
`Figure 4
`
`37.
`
`Facebook’s Advertiser Help Center states that the “Potential Reach is an estimation of
`
`the potential number of people your ads could reach.”23
`
`II.
`
`Facebook’s Potential Reach is Inflated
`
`38.
`
`According to Census data, the U.S. population numbered over 320 million individuals
`
`in 2017, with just over 250 million adults over 18 years old.24
`
`39.
`
`Facebook represents to advertisers a Potential Reach of 230 million monthly active users
`
`over the age of 18 in the U.S.25
`
`40.
`
`According to a 2018 survey from the Pew Research Center, 68% of U.S. adults, aged
`
`18 and over, use Facebook.26 Based on this 68% figure, only 170 million U.S. adults have a Facebook
`
`account. Thus, Facebook represents to advertisers that it has approximately 35% more users in the
`
`United States than it actually does.
`
`41.
`
`Using the usage rate estimate provided by Pew Research Center, Facebook’s Potential
`
`Reach is inflated with respect to the 18 years and over demographic in the four largest states, as set
`
`
`23 Quoted from https://www.facebook.com/business/help (accessed August 13, 2018).
`24 See United States Census Bureau’s “American Fact Finder” available at
`https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=PEP_2017_PEPA
`GESEX&prodType=table (last accessed August 1, 2018).
`25 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018)
`26 Pew Research Center, March 2018, “Social Media Use in 2018”.
`
`
`
`10
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 12 of 33
`
`
`
`forth below:
`
`
`
`State
`
`Estimated
`Actual Facebook Users27
`
`Facebook’s
`Claimed Reached28
`
`
`Inflation Rate
`
`California
`
`20.7 million
`
`Texas
`
`14.2 million
`
`Florida
`
`11.4 million
`
`New York
`
`10.6 million
`
`
`
`30 million
`
`21 million
`
`16 million
`
`16 million
`
`45%
`
`48%
`
`40%
`
`50%
`
`42.
`
`Among the most important – if not the most important – demographic for advertisers is
`
`the 18-34 year-old audience. This is because younger audiences are generally believed to have less
`
`brand loyalty and more disposable income.29
`
`43.
`
`Even more inflated is the key marketing demographic of young adults aged 18-34, where
`
`the U.S. population in 2017 numbered just under 76 million.30 Facebook represents to advertisers that
`
`it can reach 100 million people in this age category.31 Accordingly, Facebook claims that it can reach
`
`24 million more people in this demographic than actually exist. Additionally, according to Pew
`
`Research only 80% of this demographic actually use Facebook.32 Thus, only 61 million 18-34 year-
`
`olds are actually using Facebook, resulting in a nationwide inflation rate of 64%.
`
`44.
`
`Nor can this discrepancy be explained by non-resident travelers. Facebook indicates that
`
`its Potential Reach among 18-34 year olds in the United States includes less than 2 million non-resident
`
`travelers. See Figure 5.
`
`
`
`
`27 These figures are based on the U.S. Census Bureau’s estimated demographic size and
`accounting for the national average of Facebook participation as determined by Pew Research’s
`polling data of 68%.
`28 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018)
`29 See e.g., Weinman, Jaime J. (2012). “Television's mid-life crisis.” Maclean's. 125 (27): 72
`30 See United States Census Bureau’s “American Fact Finder”, available at
`https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=PEP_2017_PEPSY
`ASEXN&prodType=table (last accessed August 1, 2018).
`31 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018)
`32 Pew Research Center, March 2018, “Social Media Use in 2018”.
`
`11
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 13 of 33
`
`
`
`Figure 5
`
`45.
`
`Publicly available research has shown that the Potential Reach for 18-34 year-olds is
`
`not only overstated at a national level but exceeds the U.S. Census Population Data in every state. See
`
`Figure 6.33
`
`
`
`33 Video Advertising Bureau Report, at 13.
`
`12
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 14 of 33
`
`
`
`Figure 6
`
`46.
`
`The difference between Facebook’s Potential Reach and U.S. Census Bureau statistics
`
`can be illustrated at the metropolitan level as well. For example, the following chart provides examples
`
`of the massive inflation of Facebook’s claimed Potential Reach in selected cities:
`
`47.
`
`Other examples34 from September 2017 for the 18-34 year-old demographic include:
`
`18-34 year-old Census Pop. Facebook’s Claimed Reach Inflation Rate
`
`City
`379,567
`
`
`1,200,000
`
`
`216%
`Dallas
`645,229
`
`
`1,800,000
`
`
`179%
`Houston
`Los Angeles 1,125,300
`
`
`2,700,000
`
`
`140%
`New York
`2,305,171
`
`
`4,100,000
`
`
`78%
`Philadelphia 459,386
`
`
`1,100,000
`
`
`139%
`Phoenix
`410,220
`
`
`720,000
`
`
`76%
`San Antonio 401,801
`
`
`690,000
`
`
`72%
`San Diego
`434,646
`
`
`760,000
`
`
`75%
`San Jose
`256,343
`
`
`490,000
`
`
`91%
`
`
`48.
`
`Because not every 18-34 year-old has a Facebook Account, the discrepancy between
`
`
`34 See Video Advertising Bureau Report, at 14.
`
`13
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 15 of 33
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`Facebook’s Potential Reach for the 18-34 year-old demographic and the Census Population does not
`
`capture the full extent to which Facebook’s Potential Reach is overstated.
`
`49.
`
`For example, Plaintiffs commissioned their own survey to determine the percentage of
`
`18-34 years old in Chicago who have a Facebook account. The survey found that 59% of 18-34 year-
`
`olds in Chicago have Facebook accounts. In September 2017 in Chicago, Facebook’s Potential Reach
`
`was 1,900,000. However, the Census Population for 18-34 year-olds in Chicago was only 808,785.35
`
`Thus, based on this comparison alone, Facebook’s Potential Reach for 18-34 years old in September
`
`2017 was 2.35 times the Census Population.
`
`50.
`
` The inflation of Facebook’s Potential Reach is even more dramatic when compared to
`
`the survey. In 2017, based on the survey described above, approximately 485,000 actual Chicago
`
`residents between 18-34 had Facebook accounts. Therefore, when Facebook represented that the
`
`“Potential Reach” for this demographic was 1,900,000, it was almost 4 times that true number of people
`
`potentially reached.
`
`51.
`
`A similar analysis reveals dramatic inflation of the Potential Reach number in Kansas
`
`City. Among the 18-54-year-olds demographic in the Kansas City metropolitan area, Facebook claims
`
`advertisements placed on its marketplace have the potential to reach 1.4 million people.36
`
`52.
`
`But the 18-54 demographic in the Kansas City metropolitan area only consists of
`
`approximately 1 million people.37 And even then, based on survey data commissioned by the Plaintiff
`
`only 70% of that demographic has Facebook accounts. This puts the inflation rate in Kansas City at
`
`100%.
`
`III.
`
`Facebook’s User Base Inflation Is Material
`
`53.
`
`Advertising purchasers, including Plaintiffs DZ Reserve and Cain Maxwell, viewed the
`
`“Potential Reach” as an important statistic because it means that more individuals could potentially
`
`
`
`35 Id.
`36 See Facebook’s Ads Manager, available at https://www.facebook.com/adsmanager/creation
`(last accessed August 1, 2018)
`37 See United States Census Bureau’s “American Fact Finder”, available at
`https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=DEC_10_SF1_QT
`P2&prodType=table (last accessed August 1, 2018).
`
`14
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 16 of 33
`
`view their advertisement.
`
`54.
`
`55.
`
`Facebook fails to disclose to advertisers that its Potential Reach statistic is inflated.
`
`Plaintiffs DZ Reserve and Cain Maxwell read Facebook’s “Potential Reach” prior to
`
`purchasing advertisements from Facebook, and they relied on Facebook’s “Potential Reach” in making
`
`their decisions to purchase advertisements from Facebook.
`
`56.
`
`Facebook’s misrepresentation of the Potential Reach of its advertisements induced
`
`advertising purchasers (including Plaintiffs DZ Reserve and Cain Maxwell)
`
`to purchase
`
`advertisements, which they otherwise would not have purchased, because purchasers (including
`
`Plaintiffs DZ Reserve and Cain Maxwell) believed that more people could potentially be reached by
`
`their advertisements than possibly could have been.
`
`57.
`
`Facebook’s misrepresentation of the Potential Reach of its advertisements induced
`
`advertising purchasers (including DZ Reserve and Cain Maxwell) to pay more for Facebook advertising
`
`than they otherwise would have been willing to pay.
`
`58.
`
`Facebook’s misrepresentation of the Potential Reach of its advertisements thereby
`
`distorted the market price for its advertising by artificially increasing the price of Facebook advertising,
`
`causing advertising purchasers (including Plaintiffs DZ Reserve and Cain Maxwell) to pay more than
`
`they otherwise would have paid.
`
`59.
`
`Facebook’s misrepresentation of the Potential Reach of its advertisements provided
`
`Facebook with an unfair competitive advantage over other advertising platforms including other online
`
`advertising platforms, such as YouTube, LinkedIn, and Twitter.
`
`NEW FACTUAL ALLEGATIONS
`
`60.
`
`Facebook’s internal documents show that Facebook personnel knew for years that the
`
`Potential Reach metric that it provides to Facebook advertisers on its advertisement purchasing
`
`interfaces (including on Ads Manager and Power Editor) was inflated and misleading.
`
`61.
`
`Facebook’s internal documents show that Facebook personnel knew its Potential Reach
`
`metric is a material representation that advertisers rely on. Internally, Facebook acknowledged in its
`
`own official documents that
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`15
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 17 of 33
`
`
`
`
`
`
`
`
`
`62.
`
`Facebook employees acknowledged in internal documents that complaints about the
`
`Potential Reach being misleading have been made since approximately September 2015.
`
`63.
`
`For example, in December 2016, the UK Advertising Standards Authority (UK ASA)
`
`complained to Facebook that the Potential Reach was misleading.
`
`64.
`
`In the fall of 2017, the Video Advertising Bureau (VAB) published a report that alleged
`
`that Facebook’s Potential Reach was inflated and exceeded the Census numbers. This report was
`
`referenced in Plaintiffs’ Second Amended Complaint, and cited in ¶¶ 45-48 of the Third Amended
`
`Complaint as the VAB Report (“Report.”)
`
`65.
`
`Senior Facebook officials struggled to respond to the publication of the Report.
`
`Facebook employees conducted analysis
`
`
`
` Among
`
`the inflation identified by the Report, the Report found that Facebook’s Potential Reach exceeds the
`
`number of 18-34 residents in every state. Cf. ¶ 46, Figure 6.
`
`66.
`
`In the fall of 2017, Facebook employees, including Facebook Chief Operating Officer
`
`Sheryl Sandberg,
`
`
`
`Ms. Sandberg wrote:
`
`
`
`
`
`
`
`67.
`
`In response to Ms. Sandberg’s e-mail, a Facebook Vice President wrote to Ms. Sandberg
`
`and Dave Wehner (Facebook’s Chief Financial Officer) that
`
`
`
`
`
`
`
`68.
`
`Facebook’s failure to remove duplicate and fake accounts from its Potential Reach
`
`metric makes the metric fundamentally misleading; Facebook represents that Potential Reach is
`
`measurement of people (see ¶ 32) when Potential Reach is, at best, a measurement of accounts.
`
`16
`
`THIRD AMENDED CONSOLIDATED CLASS ACTION COMPLAINT
`Case No.: 3:18-cv-04978
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`

`

`Case 3:18-cv-04978-JD Document 166 Filed 04/15/20 Page 18 of 33
`
`
`
`69.
`
`Facebook executives repeatedly acknowledged
`
`
`
`
`
`70.
`
`Notwithstanding their own internal acknowledgement that Potential Reach was inflated,
`
`Facebook executives did nothing. Instead, Facebook took steps to obfuscate the issue, and even to
`
`cover it up.
`
`71.
`
`First, instead of disclosing the inflation that it internally acknowledged, Facebook
`
`developed a set of talking points, by claiming, inter alia, that discrepancies between the census and its
`
`Potential Reach metric were due to travelers, and the fact that Potential Reach is not designed to match
`
`the census. Facebook never acknowledged externally what it admitted internally: that its Potential
`
`Reach metric was inflated, and that the inflation was partly due to duplicate and fake accounts.
`
`72.
`
`Second, Facebook executives took steps to cover up the problem.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`73.
`
`Facebook CFO Dave Wehner never disclosed, on the earnings call or elsewhere, the
`
`impact of duplicate and fake accoun