Trade Regulation Rule on the Use of Consumer Reviews and Testimonials

FEDERAL TRADE COMMISSION

16 CFR Part 465

RIN 3084-AB76

Trade Regulation Rule on the Use of Consumer Reviews and Testimonials

AGENCY:

Federal Trade Commission.

ACTION:

Notice of proposed rulemaking; request for public comment.

SUMMARY:

The Federal Trade Commission (“FTC or “Commission”) commences a rulemaking to promulgate a trade regulation rule entitled “Rule on the Use of Consumer Reviews and Testimonials,” which would prohibit certain specified unfair or deceptive acts or practices involving consumer reviews or testimonials. The Commission finds such practices to be prevalent based on the comments it received in response to an advance notice of proposed rulemaking and other information discussed in this publication. The Commission now solicits written comment, data, and arguments concerning the utility and scope of the proposed trade regulation rule to prohibit the specified unfair or deceptive acts or practices.

DATES:

Comments must be received on or before September 29, 2023.

ADDRESSES:

Interested parties may file a comment online or on paper by following the instructions in the Comment Submissions part of the

SUPPLEMENTARY INFORMATION

section below. Write “Reviews and Testimonials NPRM, R311003” on your comment and file your comment online at

https://www.regulations.gov.

If you prefer to file your comment on paper, mail your comment to the following address: Federal Trade Commission, Office of the Secretary, 600 Pennsylvania Avenue NW, Suite CC-5610 (Annex F), Washington, DC 20580.

FOR FURTHER INFORMATION CONTACT:

Michael Ostheimer, Attorney, Federal Trade Commission, Bureau of Consumer Protection, Advertising Practices Division, (202) 326-2699,

mostheimer@ftc.gov.

SUPPLEMENTARY INFORMATION:

The Commission invites interested parties to submit data, views, and arguments on the proposed Rule on the Use of Consumer Reviews and Testimonials (“proposed Rule”) and, specifically, on the questions set forth in Section X of this notice of proposed rulemaking (“NPRM”). The comment period will remain open until September 29, 2023.

1

To the extent practicable, all comments will be available on the public record and posted at the docket for this rulemaking on

https://www.regulations.gov.

If interested parties request to present their position orally, the Commission will hold an informal hearing, as specified in Section 18(c) of the FTC Act, 15 U.S.C. 57a(c). Persons interested in making a presentation at an informal hearing must file a comment expressly requesting a hearing in response to this publication, containing a statement identifying their interests in the proceeding and any proposals to add disputed issues of material fact necessary to be resolved during an informal hearing. The comment should describe why the person thinks the informal hearing is warranted and how they would participate, and include a summary of their expected testimony. Interested persons' comments may also, without requesting an informal hearing, expressly request to speak at any informal hearing that is held, which may happen if another commenter requests an informal hearing or if the Commission on its own elects to hold one. If an informal hearing is held, the Commission will publish a separate notice in accordance with 16 CFR 1.12(a) (“initial notice of informal hearing”).

1

The Commission elects not to provide a separate, second comment period for rebuttal comments.

See

16 CFR 1.11(e) (“The Commission may in its discretion provide for a separate rebuttal period following the comment period.”).

I. Background

The Commission published, on November 8, 2022, an advance notice of proposed rulemaking (“ANPR”) under the authority of Section 18(a)(1)(B) of the FTC Act, 15 U.S.C. 57a(a)(1)(B);

2

which authorizes the Commission to promulgate, modify, or repeal trade regulation rules that define with specificity acts or practices that are unfair or deceptive in or affecting commerce within the meaning of Section 5(a)(1) of the FTC Act, 15 U.S.C. 45(a)(1).

2

Fed. Trade Comm'n, ANPR: Trade Regulation Rule on the Use of Reviews and Endorsements (“ANPR”), 87 FR 67424 (Nov. 8, 2022),

https://www.federalregister.gov/documents/2022/11/08/2022-24139/trade-regulation-rule-on-the-use-of-reviews-and-endorsements.

The ANPR described the Commission's history of educating industry and consumers about the use of deceptive reviews and testimonials and of taking law enforcement action against certain unfair or deceptive acts or practices involving consumer reviews or testimonials.

3

Specifically, the ANPR discussed: (a) the use of reviews or endorsements by people who do not exist, who did not actually use or test the product or service, or who were misrepresenting their experience with it; (b) review hijacking, where a seller steals or repurposes reviews of another product; (c) marketers offering compensation or other incentives in exchange for, or conditioned on, the writing of positive or negative consumer reviews; (d) owners, officers, or managers of a company (i) writing reviews or testimonials of their own products or services, or publishing testimonials by their employees or

family members, which fail to provide clear and conspicuous disclosures of those relationships, or (ii) soliciting reviews from employees or relatives without instructing them to disclose their relationships; (e) the creation or operation of websites, organizations, or entities that purportedly provide independent reviews or opinions of products or services but are, in fact, created and controlled by the companies offering the products or services; (f) misrepresenting that the consumer reviews displayed represent most or all of the reviews submitted when, in fact, reviews are being suppressed based upon their negativity; (g) the suppression of customer reviews by physical threat or unjustified legal threat; and (h) selling, distributing, or buying followers, subscribers, views, and other indicators of social media influence. The ANPR also asked a series of questions to inform the Commission's determination about whether it has reason to believe that such practices are prevalent and, if so, whether and how to proceed with an NPRM.

4

During the 60-day comment period, the Commission received 42 responsive comments.

3

The ANPR was entitled “Trade Regulation Rule Concerning Reviews and Endorsements.” The Commission has decided to change the name of the proposed rule to “Trade Regulation Rule on the Use of Consumer Reviews and Testimonials,” to better reflect its content.

4

ANPR, 87 FR at 67427.

Based on the substance of these comments, as well as the Commission's history of enforcement and other information discussed below, the Commission is now exercising its authority under Section 18(a)(1)(B) of the FTC Act to propose a trade regulation rule that defines conduct that, in the context of consumer reviews or testimonials, constitutes unfair or deceptive acts or practices. The Commission has reason to believe that certain unfair or deceptive practices involving consumer reviews or testimonials are prevalent

5

and that proceeding with this rulemaking is in the public interest.

5

See

15 U.S.C. 57a(b)(3) (“The Commission shall issue a notice of proposed rulemaking pursuant to paragraph (1)(A) only where it has reason to believe that the unfair or deceptive acts or practices which are the subject of the proposed rulemaking are prevalent.”).

After reviewing the comments and because the Commission believes it would be in the public interest to move forward expeditiously with this rulemaking proceeding, the Commission has decided to issue this NPRM without holding the public workshops originally contemplated in the ANPR. Upon reviewing the ANPR comments, the Commission determined that conducting public workshops at that stage of the proceeding would not provide additional unique viewpoints or issues. Instead, by issuing this NPRM and analyzing the comments submitted in response, the Commission will be able to further develop the record, receive comments on potential alternatives, and decide whether additional events or methods are needed to facilitate public participation in the rulemaking process.

Below, after discussing the comments, setting out the evidence of prevalence, and explaining its considerations in developing the proposed Rule, the Commission poses specific questions for comment and provides the text of the proposed Rule.

II. Summary of Comments to ANPR

The Commission received 42 responsive comments in response to the ANPR.

6

Twenty-nine comments supported the Commission proceeding with a rulemaking. Four comments expressed the view that a rulemaking was unnecessary, premature, or should not apply to the commenter's constituents. One commenter expressed skepticism about the utility of a rulemaking. The remaining commenters did not express a clear view on the merits of proceeding or did not address the question. Fifteen comments came from individual consumers. Seven comments were submitted by trade associations, five by review platform operators and one by an employee of one such operator, three by small businesses and one by a small business employee, three by consumer advocacy organizations, three by entities dedicated to fighting fake reviews, one by a public interest research center, one by a think tank, one by academic researchers, and one by an insurance marketing organization.

6

The comments are publicly available on this rulemaking's docket at

https://www.regulations.gov/docket/FTC-2022-0070/comments.

The 15 individual consumers expressed significant concerns about fake consumer reviews and testimonials.

7

One consumer comment declared: “this rule to extend the FTC power over fraudulent and paid for testimonies and reviews is a necessity. I think . . . protection against these types of scams is an integral need to the people of the United States.”

8

Consumer commenters wrote about the difficulty that many consumers have in identifying fake reviews.

9

One consumer who selected an auto repair shop based upon misleading reviews written by the shop's employees or their spouses spoke of having been personally harmed by deceptive reviews.

10

Two consumer comments said that truthful negative reviews are valuable and should not be suppressed.

11

Two additional consumer commenters spoke of the need to punish and deter bad actors.

12

7

Anonymous Cmt. on Trade Regulation Rule on the Use of Reviews and Endorsements (“Cmt. on ANPR”) (Nov. 15, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0006

(“Anonymous Consumer A Cmt.”); Mahzer Zaim, Cmt. on ANPR (Nov. 21, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0008

(“Zaim Cmt.”); Jill Monday, Cmt. on ANPR (Dec. 3, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0010

(“Monday Cmt.”); Donald Kelly, Cmt. on ANPR, (Dec. 7, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0012

(“Kelly Cmt.”); Heather Earl, Cmt. on ANPR (Dec. 11, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0013;

Andrea Sliger, Cmt. on ANPR (Dec. 11, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0014;

Merrill Ahrens, Cmt. on ANPR (Dec. 11, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0015;

Diane Dauite, Cmt. on ANPR (Dec. 11, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0016

(“Dauite Cmt.”); Stephanie Smith, Cmt. on ANPR (Dec. 12, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0017

(“Smith Cmt.”); Anonymous, Cmt. on ANPR (Dec. 13, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0018

(“Anonymous Consumer B Cmt.”); Jim Zevely, Cmt. on ANPR (Dec. 17, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0019

(“Zevely Cmt.”); Frank Evelhoch II, Cmt. on ANPR (Dec. 17, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0021

(“Evelhoch Cmt.”); Anonymous, Cmt. on ANPR (Dec. 26, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0022

(“Anonymous Consumer C Cmt.”); Judy Draper, Cmt. on ANPR (Dec. 28, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0023;

Anonymous, Cmt. on ANPR (Dec. 31, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0025

(“Anonymous Consumer D Cmt.”).

8

Anonymous Consumer A Cmt. at 1.

9

Kelly Cmt. at 1; Smith Cmt. at 1; Zevely Cmt. at 1; Evelhoch Cmt. at 1; Anonymous Consumer D Cmt. at 1.

10

Monday Cmt. at 1.

11

Dauite Cmt. at 1; Anonymous Consumer B Cmt. at 1.

12

Zaim Cmt. at 2; Anonymous Consumer C Cmt. at 1.

The four comments from small businesses or a small business employee were from Ubiquitous Advising, LLC, (“Ubiquitous Advising”), Patrick's Pet Care, an anonymous small business that sells products through a particular online marketplace, and Tammy Provencal, who is a small business employee.

13

Ubiquitous Advising supports the rulemaking, and commented that fake reviews cause more damage than anyone can imagine.

14

It also said that review

suppression is just as bad, with businesses threatening, bullying, or suing consumers who are trying to warn other consumers, even when there is zero chance of those businesses winning such a lawsuit.

15

Ubiquitous Advising described a company in its local area that is constantly threatening and bullying reviewers.

16

Patrick's Pet Care did not indicate clearly whether it supports the rulemaking but complained about being attacked with negative reviews.

17

It suggested that people should not be able to post anonymous, non-traceable reviews and that platforms should disclose the names of reviewers.

18

The anonymous small business that submitted a comment did not address the proposed rulemaking and asserted that a particular online marketplace was manipulating the placement of negative reviews.

19

The small business employee supports the rulemaking and stated that a competitor is giving incentives for 5-star reviews.

20

13

Ubiquitous Advising, LLC, Cmt. on ANPR (Dec. 29, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0024

(“Ubiquitous Advising Cmt.”); Patrick's Pet Care, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0032

(“Patrick's Pet Care Cmt.”); Anonymous, Cmt. on ANPR (Nov. 28, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0009

(“Anonymous Business Cmt.”); Tammy Provencal, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0042

(“Provencal Cmt.”).

14

Ubiquitous Advising Cmt. at 1-2.

15

Id.

at 1.

16

Id.

17

Patrick's Pet Care Cmt. at 1-2.

18

Id.

at 2.

19

Anonymous Business Cmt. at 1-2.

20

Provencal Cmt. at 1.

The five review platforms that submitted comments, Yelp, Inc. (“Yelp”), Trustpilot A/S (“Trustpilot”), Google LLC (“Google”), Tripadvisor LLC (“Tripadvisor”), and Amazon.com, Inc. (“Amazon”), wrote of the importance of reviews to consumers and the lengths to which they go to stop and combat fake reviews.

21

These comments conveyed information both about the prevalence and harm caused by fake review practices.

21

Yelp, Inc., Cmt. on ANPR (Jan. 6, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0028

(“Yelp Cmt.”); Trustpilot A/S, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0031

(“Trustpilot Cmt.”); Google LLC, Cmt. on ANPR, (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0034

(“Google Cmt.”); Tripadvisor LLC, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0036

(“Tripadvisor Cmt.”); Amazon.com, Inc., Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0041

(“Amazon Cmt.”).

Yelp, which supports civil penalties for “businesses and individuals who author, arrange for or pay for deceptive reviews,”

22

said that an overwhelming majority of consumers who read reviews (83 percent) say they trust online reviews about local businesses.

23

In one Yelp survey, 71 percent of respondents said they would no longer visit a business if they learned the business has fake or compensated online reviews.

24

As a first line of defense, Yelp uses automated software systems in order to detect biased reviews and “flags a significant percentage of reviews—about 19% based on Yelp's most recent . . . figures—as `not recommended.' ”

25

Yelp said that groups to facilitate the buying, selling, or exchange of fake reviews exist on various online platforms (

e.g.,

Facebook, Instagram, Twitter).

26

In 2021, Yelp made more than 1,000 reports to online platforms warning them of nearly 950 suspicious groups, posts, or individuals found on their sites.

27

Yelp also wrote that “abusive and questionable or unjustified legal threats are another form of review suppression that Yelp constantly confronts” and that its 2021 data shows a majority of such threat alerts “stemmed from beauty and health categories—businesses consumers often turn to when making critical life decisions or that can otherwise be sensitive in nature.”

28

22

Yelp Cmt. at 12. At the same time, Yelp said that because “such deceptive review practices are already illegal under Section 5 of the FTC Act,” it “recommends against additional rulemaking that is specifically directed toward liability for deceptive reviews.” The Commission has difficulty reconciling these two comments as the Commission could not adopt new civil penalties without rulemaking.

23

Id.

at 4.

24

Id.

25

Id.

at 6.

26

Id.

at 8.

27

Id.

28

Id.

at 11-12.

Trustpilot, a Danish company operating a website that hosts reviews of businesses worldwide, did not appear to support or oppose the rulemaking. It said that of the 46.7 million Trustpilot reviews written globally in 2021, it removed 2.7 million fake reviews.

29

In 2021, Trustpilot identified and took action against more than 60,000 reviews about United States businesses that were submitted by accounts it deemed to be companies or individuals who offered fake reviews for sale online.

30

It noted that it is appropriate for consumers to review a service provider with which they have had an experience even if they did not make a purchase.

31

In 2021, Trustpilot detected and removed as biased just over 8,000 reviews for United States businesses written by owners, officers, or employees of the company reviewed, or their family members.

32

Trustpilot stated that such behavior does not necessarily reflect intentional fraud.

33

It commented that it is aware of cases outside of Trustpilot in which the suppression of negative reviews has occurred on retailer or business websites.

34

It has seen some cases, mostly outside of the US, in which businesses have threatened reviewers if they do not delete a negative review.

35

In response to the ANPR, Trustpilot said it is possible that, before moving to regulation, there may be benefits in seeking to maximize the effects of other steps, such as educating businesses and consumers or developing codes of conduct.

36

It noted that while regulation could send a strong signal, it may face the challenge of being quite static in a dynamic and fast-paced environment.

37

29

Trustpilot Cmt. at 2. Trustpilot noted that it defines “fake reviews” more broadly than was used in the FTC's ANPR.

30

Id.

at 3.

31

Id.

at 3-4.

32

Id.

at 7.

33

Id.

34

Id.

at 8.

35

Id.

at 9.

36

Id.

at 16.

37

Id.

at 17.

Google supports the rulemaking.

38

It said that fake reviews undermine users' confidence in the information available on its platform.

39

Google uses both automated systems and human operators to monitor compliance with its policies and identify and remove fake reviews.

40

Spammers constantly evolve their tactics, so distinguishing between fake and authentic reviews is an ongoing battle.

41

For example, in response to advances in Google's detection and mitigation capabilities, bad actors have adapted, such as by using Virtual Private Networks (“VPNs”) to evade routine detection.

42

Google said that businesses may also have strong incentives to buy positive reviews, which exacerbates the problem.

43

In addition, many reviews displayed on its platform are sourced or surfaced from third parties (

e.g.,

from the merchant website where consumers purchased the product or service), and it can be more difficult to detect when such reviews are fake because Google lacks access to some signals of inauthentic activity, such as the account that created the review being used to post duplicate content.

44

In 2022, Google removed millions of reviews from Google Play that it determined to be fake, inorganic, or otherwise malicious.

45

In 2021, users submitted around one billion Google Maps reviews and Google blocked or removed more than 95 million of them for violating its policies.

46

Google also removed another one million reviews that were reported

directly to it, and it disabled more than one million user accounts due to policy-violating activity.

47

Google urged the Commission to focus on those posting fake reviews rather than on the platforms.

48

38

Google Cmt. at 9.

39

Id.

at 1, 2, 9.

40

Id.

at 1.

41

Id.

at 3.

42

Id.

43

Id.

44

Id.

at 3, 6.

45

Id.

at 8.

46

Id.

47

Id.

48

Id.

at 9.

Tripadvisor agrees that deceptive actions by bad actors harm consumers and honest businesses.

49

In 2021, of the 26 million reviews submitted to Tripadvisor, it identified 3.6 percent as violating its fraud guidelines.

50

It said that in certain scenarios it can be difficult to distinguish authentic reviews from fake.

51

Tripadvisor also said that efforts to suppress negative reviews, including by threatening reviewers, is one of the problems that plague the online consumer review ecosystem.

52

Finally, it believes that targeted authority for the FTC to impose financial penalties on bad actors can be an element of a comprehensive effort to improve the consumer information ecosystem, but that any provision that authorizes the assessment of a financial penalty must be appropriately targeted in both design and enforcement at those who knowingly engage in clearly deceptive and fraudulent practices.

53

49

Tripadvisor Cmt. at 7.

50

Id.

51

Id.

at 5-6.

52

Id.

at 11.

53

Id.

at 10.

Amazon did not state support for or opposition to the rulemaking. Amazon said that in 2021 alone, it invested more than $900 million and employed more than 12,000 people who were dedicated to protecting customers and its store from fraud and other forms of abuse.

54

Amazon stated that it proactively stopped more than 200 million suspected fake reviews in 2020 alone.

55

Amazon also noted that fraudsters approach its customers through their own websites and on social media and solicit them to write misleading reviews in exchange for money, free products, or other incentives.

56

In 2021, Amazon reported more than 16,000 social media groups that were buying or exchanging misleading reviews to the social media sites that hosted them, including Facebook, Twitter, and Instagram, resulting in the removal of groups with more than 11 million members.

57

In July 2022, Amazon sued more than 10,000 such Facebook groups.

58

Amazon encouraged the FTC to increase the use of its existing authority to pursue fake review brokers, collaborate with other regulators to combat bad actors who facilitate review abuse, continue to provide guidance to legitimate businesses, and educate consumers about how to identify and report fake reviews.

59

54

Amazon Cmt. at 2.

55

Id.

56

Id.

at 2.

57

Id.

58

Id.

at 3.

59

Id.

An individual Amazon employee working in the Amazon Risk department for the past 10 years submitted a comment.

60

The commenter personally reviewed thousands of seller and buyer accounts for review abuse and said there is no dispute that deceptive reviews are widespread and harmful to customers.

61

The commenter is “skeptical about whether the regulation will be effective” because most online platforms and shopping websites do not require customers to register using real identities in order to leave reviews and because Section 230 of the Communications Decency Act (47 U.S.C. 230) immunizes internet service providers, like Google and Facebook, from lawsuits “based on claims related to content published by third-parties using their service[s].”

62

60

Miao Zhao, Cmt. on ANPR (Dec. 19, 2022),

https://www.regulations.gov/comment/FTC-2022-0070-0020

(“Zhao Cmt.”) at 1.

61

Id.

62

Id.

at 1-2.

The academic researchers who submitted a comment were Rajvardhan Oak and Zubair Shafiq from the University of California Davis, who had examined reviews on online marketplaces.

63

They infiltrated an “incentivized review service geared towards Amazon.com” and discovered solicitations for incentivized five-star reviews for 242,000 products.

64

They found more than 250 groups on Facebook in which reviews were brokered, the largest of which had around 550,000 members.

65

Over the six weeks that they tracked products for which incentivized reviews were sought, no reviews were removed from nearly 50 percent of those products.

66

Although Amazon delists products suspected of seeking incentivized reviews, only 25 of the 1,600 products they were tracking were removed by Amazon during the six-week period.

67

They also said that, in response to Amazon's lawsuits against Facebook groups, group administrators and agents simply created alternate communication channels, such as Signal/Telegram groups, and circulated the details of the alternatives.

68

63

Rajvardhan Oak and Zubair Shafiq, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0030

(“Oak & Shafiq Cmt.”) at 1.

64

Id.

at 3-4.

65

Id.

at 4.

66

Id.

at 7-8.

67

Id.

at 8.

68

Id.

The Commission received comments from three entities dedicated to fighting fake reviews: the Transparency Company, Fake Review Watch, and Fakespot, Inc.

69

All three commenters asserted that the strategies that are currently being used by review platforms are insufficient.

70

69

The Transparency Company, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0044

(“Transparency Company Cmt.”); Fake Review Watch, Cmt. on ANPR (Jan. 5, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0026

(“Fake Review Watch Cmt.”); Fakespot, Inc., Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0035

(“Fakespot Cmt.”).

70

Transparency Company Cmt. at 9, 18; Fake Review Watch Cmt. at 2; Fakespot Cmt. at 1-2.

The Transparency Company, which supports a rulemaking, said that its research suggests that the major review websites are unable to detect a majority of fake reviews online.

71

It estimated that 8.5 percent of published reviews—for all industries—are fake, and provided a link to its fake review research, which asserted that 10.7 percent of Google reviews, 7.1 percent of Yelp reviews, and 5.2 percent of Tripadvisor reviews were fake.

72

The comment noted that 54 percent of consumers say that they would not buy a product if they suspected it to have fake reviews and estimated that consumer injury from fake reviews is approximately $5 billion per year.

73

It documented over 1,000 examples of fake negative reviews causing injury to competition and it estimates that thousands of lawyers are hired each year to send demand letters to and intimidate the authors of negative consumer reviews.

74

The comment identified platform actions that have been effective in reducing consumer harm associated with fake reviews but said that ending online review fraud would require, among other things, the authentication of consumer reviewers.

75

71

Transparency Company Cmt. at 9.

72

Id.;

Uberall,

The State of Online Review Fraud: An Analysis of 4 Million Reviews on Google, Facebook, Yelp and Tripadvisor

at 15,

https://join.momentfeed.com/hubfs/2021%20Fake%20Reviews/FakeReviews_Report.pdf.

73

Transparency Company Cmt. at 16, 18.

74

Transparency Company Cmt. at 16.

75

Id.

at 18-20.

Fake Review Watch, which supports a rulemaking, said that there is a robust black market for paid for (or traded for) reviews on Google, Yelp, Facebook, Trustpilot, and numerous other review sites and that many of the transactions are conducted on social media.

76

It

stated that fake reviews are commonplace and often difficult to detect without examining review profile histories across multiple businesses.

77

Fake Review Watch has observed over 100 Facebook groups operating as review exchanges, with hundreds or thousands of members each.

78

The comment also asserted that Google: (a) often allows profiles that posted fake reviews to remain active even after it removes those reviews, (b) provides no alerts to consumers about businesses with fake reviews, and (c) makes fake review detection more difficult by allowing profiles to choose not to display all of their reviews and by not displaying the dates of reviews.

79

The comment also complained about reviews by Yelp Elite members,

80

which Fake Review Watch asserted are automatically recommended and not subject to evaluation by Yelp's recommendation software, and about the inadequacy of Yelp's consumer alerts.

81

Fake Review Watch said that regulators should require review sites to tell consumers everything they know about a business's reviews and to post notices reminding consumers that the site cannot guarantee the truthfulness and accuracy of any review.

82

76

Fake Review Watch Cmt. at 1, 9.

77

Id.

at 1.

78

Id.

79

Id.

at 3.

80

According to Yelp, Yelp Elite members are chosen “based on a number of things, including well-written reviews, high quality photos, a detailed personal profile, and a history of playing well with others.”

https://www.yelp-support.com/article/What-is-Yelps-Elite-Squad?l=en_US.

81

Id.

at 4-5.

82

Id.

at 9.

Fakespot did not state support for or opposition to a rulemaking. In its comment, Fakespot opined that sellers posting fake reviews and fake review farms are among the malicious actors generating fake online content that, in the last five years, has led to a “dramatic deterioration” of trust between sellers, platforms, and consumers.

83

83

Fakespot Cmt. at 1.

The three consumer advocacy organizations that submitted comments, Truth in Advertising, Inc. (“TINA”), the U.S. Public Interest Research Group (“US PIRG”), and the National Consumers League (“NCL”), all advocated for a rulemaking.

84

84

Truth in Advertising, Inc., Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0029

(“TINA Cmt.”) at 1; U.S. Public Interest Research Group, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0045

(“US PIRG Cmt.”) at 2; National Consumers League, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0039

(“NCL Cmt.”) at 1. Electronic Privacy Information Center (“EPIC”), which is a public interest research center, submitted a comment supporting a rulemaking. Its comment focused mainly on endorsements by police organizations of one product. Electronic Privacy Information Center, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0043.

TINA said that fake reviews are an insidious problem, primarily because consumers have come to rely heavily on reviews in making their online purchasing decisions, and it provided numerous citations to publications regarding the importance of reviews to consumer decision making.

85

It stated that incentives to generate early, positive reviews have led to a proliferation of false and fake reviews—a deceptive marketing tactic that will only continue to flourish if not effectively reined in by regulators.

86

TINA said that, given the Supreme Court's

AMG Capital Management

decision, a rule would substantially improve the agency's ability to combat and deter deception and unfairness in this area.

87

85

TINA Cmt. at 2-3.

86

Id.

at 3.

87

Id.

at 3.

US PIRG cited findings by industry observers that 30 to 40 percent of online reviews are not genuine, and stated that consumers have no way of knowing which reviews are legitimate.

88

It asserted that fake reviews harm both consumers who are trying to make informed buying decisions and honest businesses, and that, when consumers lose confidence in reviews, legitimate positive reviews do not mean as much.

89

It said that the marketplace is poisoned by outright fake reviews, reviews written in exchange for free items, fake negative reviews written about competitors, review suppression, reviews or endorsements written for consideration, and misrepresentations that a website or a certification or a seal is independent.

90

88

US PIRG Cmt. at 1.

89

Id.

90

Id.

at 1-2.

NCL said that millions of consumers use reviews every day to inform billions of dollars in purchasing decisions involving both online and offline businesses.

91

It cited an estimate that in 2021, fraudulent reviews cost U.S. consumers $28 billion.

92

NCL also cited a different study which said that, by deceiving buyers into purchasing lower quality and potentially unsafe products, fake reviews lead to $0.12 of consumer welfare lost for every $1 spent online.

93

It said that the practices outlined in the ANPR were all unfair and deceptive, that the sellers and service providers that do not use fake reviews are at a competitive disadvantage, that the effects of a fake review may last up to a month after its deletion or detection, and that the threat of fake negative reviews is being used to extort honest businesses.

94

NCL also asked the Commission to require platforms to implement measures to combat the unfair and deceptive uses of reviews, endorsements, and indicators of social media influence, possibly requiring purchase verification before allowing a user to leave a review and the active policing of reviews.

95

Finally, NCL suggested that the FTC explore options for holding platforms accountable for allowing organized review fraud to flourish.

96

91

NCL Cmt. at 1.

92

Id.

at 2.

93

Id.

94

Id.

at 3-4.

95

Id.

at 4.

96

Id.

The seven trade associations that submitted comments, the North American Insulation Manufacturers Association (“NAIMA”), the American Dental Association (“ADA”), the Computer & Communications Industry Association (“CCIA”), the Travel Technology Association (“Travel Tech”), the National Automobile Dealers Association (“NADA”), the National Retail Federation (“NRF”), and the Association of National Advertisers (“ANA”),

97

took widely divergent positions on a rulemaking.

97

North American Insulation Manufacturers Association, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0037

(“NAIMA Cmt.”); American Dental Association, Cmt. on ANPR (Jan. 5, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0027

(“ADA Cmt.”); Computer & Communications Industry Association, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0047

(“CCIA Cmt.”); Travel Technology Association, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0046

(“Travel Tech Cmt.”); National Automobile Dealers Association, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0038

(“NADA Cmt.”); National Retail Federation, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0039

(“NRF Cmt.”); Association of National Advertisers, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0040

(“ANA Cmt.”).

NAIMA and ADA both support a rulemaking.

98

With respect to reviews or other endorsements by nonexistent individuals, NAIMA said that it has challenged misleading claims that were “supported by avatars or entities that there was no chance of making real contact with.”

99

It also asserted that testimonials by those misrepresenting their experiences with products are plentiful.

100

Finally, NAIMA stated that it regularly challenges statements about

its members' products “that appear on standalone websites which falsely claim to be independent reviewers.”

101

ADA wants the FTC to allow dentists to disclose patient information in responding to reviews and to require that reviewers identify themselves.

102

98

NAIMA Cmt. at 1; ADA Cmt. at 1.

99

NAIMA Cmt. at 2.

100

Id.

101

Id.

102

ADA Cmt. at 1.

It is unclear whether Travel Tech or CCIA support a rulemaking. Travel Tech commented that the integrity of reviews is essential to maintain the trust and confidence of the customers of Travel Tech members.

103

It stated that the overwhelming majority of reviews are legitimate and that Travel Tech members have systems in place to address the minority of reviews that can be harmful to consumers or travel-related operators and providers.

104

Travel Tech recommended that the Commission utilize its existing authority to combat nefarious paid review-generation sites, referred to as “click farms.”

105

CCIA said that any proposed rulemaking should focus on bad actors engaging in fraudulent behavior, not legitimate endorsements that happen to occur through social media.

106

103

Travel Tech Cmt. at 1.

104

Id.

at 3.

105

Id.

at 4.

106

CCIA Cmt. at 4.

NADA commented that rulemaking is unnecessary because the Commission did not identify any harmful market conduct for which remedies to protect consumers do not exist under current Federal and state law and because monetary penalty authority alone is not reason enough to issue a rule.

107

Its comment continued that, if the rulemaking proceeds, the Commission should stick to its stated goal of addressing “certain types of clear Section 5 violations involving reviews and endorsements” to “benefit consumers, help level the playing field, and not burden legitimate marketers.”

108

With respect to any potential rule provision addressing businesses writing, soliciting, or publishing reviews by their employees or family members, NADA asked that the FTC make clear that a violation “only arises when the business, and not another entity, affirmatively writes, solicits, and publishes reviews that fail to provide clear and conspicuous disclosures of those relationships” and that the FTC define the term “relative.”

109

The comment asserted that businesses may legitimately “seek to remove reviews or comments that are off topic or include false statements, advertisements, inappropriate language, or confidential or personal identification information” or to “remove comments or review functions on their own websites or certain social media posts.”

110

NADA also posited other practices that they considered legitimate and did not want prohibited under a possible rule: (a) responding on a comment thread to each negative review, offering an explanation, making customers whole, and asking any successfully satisfied customers to respond on the thread with their satisfaction or update their previously negative review; (b) surveying customer satisfaction and prompting only satisfied customers to leave reviews; (c) reaching out to consumers in an effort to change reviews by addressing their issues, sometimes giving customers something of value in satisfaction of their problems; or (d) highlighting five-star reviews from satisfied customers on a dealer's websites.

111

NADA said it understood that some third-party review websites promoted their services to businesses and if a business did not purchase those services, it would have a negative effect on the consumer reviews shown for the business.

112

Finally, NADA said that the FTC should directly engage with review websites, e-commerce sites, and consumer brands through public workshop conferences.

113

107

NADA Cmt. at 1-2.

108

Id.

at 3.

109

Id.

110

Id.

111

Id.

at 3-4.

112

Id.

at 4 n.12.

113

Id.

at 5.

NRF opposed additional regulation of retailers but not of fake review brokers.

114

It believes that the issue of fake and misleading reviews is important but that fake review brokers are much more likely to mislead consumers and create issues for retailers given the potential for brokers to submit fake reviews in volume.

115

NRF said that the fraudulent tactics employed by review brokers can include: (a) using “bots” and artificial intelligence tools to generate reviews on behalf of nonexistent consumers; (b) posting identical, or substantially identical, reviews for multiple different products and/or under multiple consumer accounts; (c) flooding social media platforms such as Twitter, Instagram, and Facebook with false review content, whether as standalone posts or as comments or replies to genuine reviews or consumer questions; (d) creating and operating social media groups or standalone websites that purport to offer benefits like refunds or coupons in exchange for specified types of reviews or ratings; and (e) reimbursing consumers for what would otherwise appear to be bona-fide purchases in exchange for positive 5-star reviews and ratings.

116

NRF opposed requiring retailers to restrict consumer reviews to verified purchasers.

117

It also opposed blanket approaches such as “requiring manual review of every consumer review and the poster's profile” or approaches that “risk inadvertent discriminatory or disparate deletion of reviews based on implicit biases towards certain consumer classes.”

118

NRF said that if a retailer is actually acting in bad faith (whether by itself or by intentionally engaging a fake review broker to act on its behalf), the FTC can take the step of “filing a complaint and bringing formal enforcement action seeking monetary damages as it has done several times this year alone.”

119

It accordingly believes that no new enforcement mechanism is necessary for the Commission to ensure retailers comply with existing law, or to hold them accountable for violations.

120

114

NRF Cmt. at 1.

115

Id.

at 2.

116

Id.

at 2-3.

117

Id.

at 6.

118

Id.

119

Id.

120

Id.

at 7.

ANA asserted that a rulemaking is premature, while making clear that “ANA does not take the position that fake reviews may not produce economic injury.”

121

It asserted that the “FTC has not demonstrated evidence of prevalence and has not identified a particular industry that would justify embarking upon rulemaking that would be sufficient, clear, narrowly tailored, easy to enforce, and not burdensome to legitimate marketers.”

122

ANA appeared to agree that some of the practices challenged in past FTC cases involving the offering of compensation or other incentives in exchange for, or conditioned on, the writing of positive consumer reviews are problematic and deceptive.

123

It sought to distinguish such practices from other practices that, according to ANA, do not obviously cause consumer harm, such as review gating or the “mere solicitation of positive reviews.”

124

121

ANA Cmt. at 2, 4.

122

Id.

at 1-2.

123

Id.

at 6.

124

Id.

at 7.

The Commission also received a comment from an insurance marketing organization, Family First Life LLC (“Family First Life”), which supported “a narrowly tailored rule [that] would benefit consumers, help level the playing field, and not burden legitimate

marketers.”

125

It recommended that any regulation “be tailored to exclude situations where an employee or independent contractor is leaving a review of their experience working with their employer or principal.”

126

Family First Life pointed out that when someone “writes a review of her own personal experience working with a company on workplace-review platforms, such as Glassdoor or Indeed,” concerns about the reviewer's undisclosed relationship to the company are absent because, on such platforms, “there is an obvious and assumed relationship between the reviewer and the company.”

127

Family First Life commented that the “FTC should not write a rule that sweeps in and penalizes any review just because the reviewer was offered an incentive to write it—without otherwise dictating what the review says.”

128

Family First Life also stated that the “FTC should include in any proposed regulation it promulgates a safe harbor for truthful reviews that are incentivized but not influenced, controlled, or conditioned by the entity offering the incentive.”

129

Finally, it asserted that the FTC should not treat platforms' determinations of policy violations as evidence of rule breaking.

130

125

Family First Life LLC, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0049

(“Family First Life Cmt.”) at 1-2.

126

Id.

at 9.

127

Id.

at 9-10.

128

Id.

at 12-13.

129

Id.

at 14.

130

Id.

at 18.

The Commission also received a comment from a non-partisan think tank, the Center for Data Innovation (“CDI”).

131

As part of its comment, CDI asserted that regulation is premature because there are no widely accepted best practices for platforms and platforms are still experimenting with solutions.

132

CDI acknowledged that researchers studying deceptive reviews found that fake reviews do have a large presence online and a significant impact on commerce, citing research and reports that included the following findings, among others: (a) “around five percent of reviews left for a private-label apparel company were posted by individuals who did not purchase products”; (b) “around 4 percent of online reviews [we]re fake in 2021”; (c) fake reviews impact nearly $152 billion in global e-commerce revenue; (d) Yelp flagged and filtered out around 16 percent of reviews in 2016; and (e) 20 percent of 41,572 reviews on Tripadvisor were suspicious.

133

It noted that an artificial intelligence (“AI”) system is able to write reviews that are nearly indistinguishable from reviews written by people.

134

CDI commented that fake review brokers help facilitate the creation of fake reviews by connecting bad actors with reviewers, often using “large groups on websites such as Facebook to find reviewers willing to write reviews in exchange for free products or compensation.”

135

The comment asserted that the review broker dictates the rating and what the review should say and then pays the reviewer only once the review is accepted and posted.

136

CDI proposed that, instead of engaging in a rulemaking, the FTC should establish partnerships with review companies, e-commerce platforms, and social media companies to establish voluntary best practices to detect and prevent fake reviews.

137

131

Center for Data Innovation, Cmt. on ANPR (Jan. 9, 2023),

https://www.regulations.gov/comment/FTC-2022-0070-0048

(“CDI Cmt.”).

132

Id.

at 5.

133

Id.

at 3.

134

Id.

at 3-4.

135

Id.

at 5.

136

Id.

137

Id.

at 5-6.

III. Prevalence of the Consumer Review and Testimonial Practices at Issue

A. Fake or False Consumer Reviews or Testimonials

Comments from the platforms support a finding that fake consumer reviews are prevalent. In 2020, Amazon asserted it proactively stopped more than 200 million suspected fake reviews.

138

In 2021, according to the company, Google blocked or removed more than 95 million Google Maps reviews for policy violations; in 2022, it removed millions of fake, inorganic, or otherwise malicious Google Play reviews.

139

Yelp commented that, in 2021, its recommendation software identified about 19 percent of reviews as “not recommended.”

140

In 2021, Tripadvisor reportedly flagged 3.6 percent of reviews submitted (or about one million reviews) as fraudulent.

141

Trustpilot stated that in 2021, accounts deemed to be review sellers submitted more than 60,000 reviews of U.S. businesses; it identified and filtered the reviews and blocked the accounts associated with them.

142

138

Amazon Cmt. at 2.

139

Google Cmt. at 8.

140

Yelp Cmt. at 6.

141

Tripadvisor Cmt. at 7.

142

Trustpilot Cmt. at 3.

Several comments spoke about the prevalence of consumer review rings on various online platforms (

e.g.,

Facebook, Instagram, Twitter) that facilitate the buying, selling, or exchange of fake reviews. In 2021, Amazon reported more than 16,000 abusive review-related groups to social media sites, leading to the removal of groups with more than 11 million members.

143

In July 2022, Amazon sued administrators of more than 10,000 Facebook groups that attempted to orchestrate fake reviews on Amazon.com in exchange for money or free products.

144

In 2021, Yelp reported almost 950 suspicious groups, posts, or individuals to online platforms.

145

Fake Review Watch has accessed more than 100 Facebook review exchange groups, each with hundreds or thousands of participants.

146

143

Amazon Cmt. at 2.

144

Id.

at 3.

145

Yelp Cmt. at 8.

146

Fake Review Watch Cmt. at 1.

The comment from the Amazon employee who reviewed thousands of accounts for review abuse said that deceptive reviews are widespread.

147

147

Zhao Cmt. at 1.

Other comments suggest the platforms may be underestimating the extent of the fake review problem. The Transparency Company estimated 8.5 percent of published consumer reviews are fake.

148

The Fake Review Watch comment explained reviews written by Yelp Elite members are not subject to evaluation by Yelp's automatic software and there is a robust market for Yelp Elite Reviews.

149

US PIRG asserted 30 to 40 percent of online reviews are fabricated or otherwise not genuine.

150

The UC Davis researchers found that nearly 50 percent of the products sold on

Amazon.com

by those seeking incentivized reviews did not have any of their reviews removed during the six-week period the researchers tracked them.

151

CDI cited research regarding the prevalence of fake reviews, including findings that “around five percent of reviews left for a private-label apparel company were posted by individuals who did not purchase products,” “around 4 percent of online reviews [we]re fake in 2021,” and, based on a third-party analysis of 41,572 reviews, around 20 percent of Tripadvisor reviews were suspicious.

152

148

Transparency Company Cmt. at 9.

149

Fake Review Watch Cmt. at 4.

150

US PIRG Cmt. at 1.

151

Oak & Shafiq Cmt. at 7-8.

152

CDI Cmt. at 3.

Numerous research reports, several of which are cited in the comments, further establish the prevalence of fake reviews. For example, in 2020, the Department of Homeland Security issued a report that focused on counterfeit and pirated goods but also found that “the ratings systems across platforms have been gamed, and the

proliferation of fake reviews and counterfeit goods on third-party marketplaces now threatens the trust mechanism itself.”

153

An Uberall report from 2021 estimated 10.7 percent of Google reviews, 7.1 percent of Yelp reviews, and 5.2 percent of Tripadvisor reviews were fake.

154

A 2021 joint report by the University of Baltimore and CHEQ AI Technologies Ltd., a company that provides online security services, described the “booming market” for fake reviews and estimated, based on self-reporting from several major platforms, four percent of global reviews are fake.

155

Also in 2021, Fakespot released a report finding that, in 2020, nearly 37.6 percent of reviews on

Walmart.com

were unreliable, with the figure at 27.6 percent for

Amazon.com.

156

Further, in its most recent annual local consumer review survey, BrightLocal reported 54 percent of consumers were confident they saw fake reviews on

Amazon.com

in 2022, with the figures being 50 percent for Google and 42 percent for Facebook.

157

153

See, e.g.,

Department of Homeland Security, Combating Trafficking in Counterfeit and Pirated Goods (2020),

https://www.dhs.gov/sites/default/files/publications/20_0124_plcy_counterfeit-pirated-goods-report_01.pdf.

154

See

“Fake Reviews: How Big a Problem Exactly?,” Oct. 28, 2021,

https://uberall.com/en-us/resources/blog/how-big-a-problem-are-fake-reviews.

Notably, these percentages refer to reviews that were not blocked by these platforms before publication.

155

See

University of Baltimore and CHEQ, “The Economic Cost of Bad Actors on the Internet: Fake Online Reviews 2021,”

https://f.hubspotusercontent00.net/hubfs/5228455/Research/Fake%20Online%20Reviews%202021.pdf.

156

See

Fakespot, “2021 Fakespot US Online Shopping, Ratings & Reviews Analysis Report,”

https://www.fakespot.com/2021holidayreport.

157

See

Sammy Paget, “Local Consumer Review Survey 2023,” Feb. 7, 2023,

https://www.brightlocal.com/research/local-consumer-review-survey/.

Academic research—some of which, again, is cited in the comments—has also repeatedly confirmed the prevalence of fake reviews.

158

Numerous journalists, including from The Washington Post, The Wall Street Journal, CBC News, and CNBC, have also reported on such prevalence, sometimes having undertaken their own investigations.

159

Further, Which?, a consumer advocacy group based in the United Kingdom, has issued several reports documenting fake and manipulated reviews across multiple platforms.

160

158

See, e.g.,

Jesper Akesson et al., “The Impact of Fake Reviews on Demand and Welfare,” National Bureau of Economic Research Conference, July 20, 2022,

https://conference.nber.org/conf_papers/f166391.pdf;

Sherry He et al., “The Market for Fake Reviews,” 2021,

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3664992;

Devesh Raval, “Do Bad Businesses Get Good Reviews? Evidence from Online Review Platforms,” 2020,

https://deveshraval.github.io/reviews.pdf;

Renee DiResta, “Manipulating Consumption,” 2018,

https://medium.com/@noupside/manipulating-consumption-42f2e9013d0b;

Ted Lappas et al., “The Impact of Fake Reviews on Online Visibility: A Vulnerability Assessment of the Hotel Industry,” 2016,

https://pubsonline.informs.org/doi/abs/10.1287/isre.2016.0674;

Michael Luca and Georgios Zervas, “Fake It Till You Make It: Reputation, Competition, and Yelp Review Fraud,” 62(12) Mgmt. Sci. Dec. 3412-27 (2016),

https://dash.harvard.edu/handle/1/22836596.

159

See, e.g.,

Bob Segall, “Millions of those 5-star online reviews are fake; Here's how to spot them,” WTHR, Feb. 15, 2022,

https://www.wthr.com/article/news/investigations/13-investigates/many-of-those-5-star-reviews-you-see-online-are-totally-fake-yelp-google-facebook-false-accounts/531-f175843b-1316-494a-a746-5bdfcada43fa;

Nicole Nguyen, “Fake Reviews and Inflated Ratings Are Still a Problem for Amazon,” Wall St. J., June 13, 2021,

https://www.wsj.com/articles/fake-reviews-and-inflated-ratings-are-still-a-problem-for-amazon-11623587313;

Laura Sydell, “Fake patient reviews are making it increasingly hard to seek medical help on Google, Yelp and other directory sites,” Wash. Post, June 5, 2021,

https://www.washingtonpost.com/business/2021/06/04/fake-medical-reviews-google-zocdoc-trustpilot/;

Matthew Pierce et al., “Black market in Google reviews means you can't believe everything you read,” CBC News, May 4, 2021 (finding that sale of reviews is a growing and widespread problem),

https://www.cbc.ca/news/investigates/fake-reviews-on-google-1.6033859;

Natasha Lomas, “Apple urged to root out rating scams as developer highlights ugly cost of enforcement failure,” Tech Crunch, Feb. 3, 2021 (finding that selling fake App Store reviews “is a booming business”),

https://techcrunch.com/2021/02/03/apple-urged-to-root-out-rating-scams-as-developer-highlights-ugly-cost-of-enforcement-failure/;

Katie Tarasov, “Amazon is filled with fake reviews and it's getting harder to spot them,” CNBC, Sep. 6, 2020,

https://www.cnbc.com/2020/09/06/amazon-reviews-thousands-are-fake-heres-how-to-spot-them.html;

Greg Sterling, “Fake reviews problem is much worse than people know,” Search Engine Land, Apr. 22, 2020,

https://searchengineland.com/fake-reviews-problem-is-much-worse-than-people-know-333331;

Nick Fernandez, “It's 2020 and the Google Play Store still has a major fake review problem,” Android Authority, Feb. 23, 2020,

https://www.androidauthority.com/play-store-fake-review-problem-1082191/;

Eric Griffith, “39 Percent of Online Reviews Are Totally Unreliable,” PCMag, Nov. 7, 2019,

https://www.pcmag.com/news/39-percent-of-online-reviews-are-totally-unreliable;

Elizabeth Dwoskin and Craig Timberg, “How merchants use Facebook to flood Amazon with fake reviews,” Wash. Post, Apr. 23, 2018,

https://www.washingtonpost.com/business/economy/how-merchants-secretly-use-facebook-to-flood-amazon-with-fake-reviews/2018/04/23/5dad1e30-4392-11e8-8569-26fda6b404c7_story.html.

160

See, e.g.,

Hannah Walsh, “Apple App store and Google Play flooded with fake reviews,” WHICH?, Mar. 9, 2023,

https://www.which.co.uk/news/article/apple-app-store-and-google-play-flooded-with-fake-reviews-aEA138U8bUw6;

Sara Spary, “How Facebook fuels Amazon's fake reviews,” WHICH?, Jan. 13, 2022 (finding Facebook groups with more than 200,000 members facilitating the sale of fake Amazon reviews),

https://www.which.co.uk/news/2022/01/how-facebook-fuels-amazons-fake-reviews/;

Hannah Walsh, “How a thriving fake review industry is gaming Amazon marketplace,” WHICH?, Feb. 16, 2021 (finding a “thriving industry of review manipulation businesses” targeting the Amazon marketplace and trading on a “massive scale”),

https://www.which.co.uk/news/2021/02/how-a-thriving-fake-review-industry-is-gaming-amazon-marketplace/.

More recently, concerns have been raised that generative artificial intelligence (“AI”) tools can be used to write product reviews.

161

It has been reported that an AI chatbot is being used to create fake reviews.

162

As the reporting notes, the widespread emergence of AI chatbots is likely to make it easier for bad actors to write fake reviews.

161

See, e.g.,

Chat GPT, “Reader Beware: This Gear Review Was Written by an AI Bot,” GearJunkie, Dec. 7, 2022,

https://gearjunkie.com/news/chat-gpt-ai-gear-review-msr-pocket-rocket.

162

See

Annie Palmer, “People are using A.I. chatbots to write Amazon reviews,” CNBC, Apr. 25, 2023,

https://www.cnbc.com/2023/04/25/amazon-reviews-are-being-written-by-ai-chatbots.html.

The Commission has brought numerous cases involving allegedly fabricated consumer reviews. S

ee, e.g.,

Complaint at 9-17,

FTC

v.

Roomster Corp.,

No. 1:22-CV-07389 (S.D.N.Y. Aug. 30, 2022) (alleged purchase and sale of fake app store and other reviews for room and roommate finder app and platform); Complaint at 2-4,

Sunday Riley Modern Skincare, LLC,

No. C-4729 (Nov. 6, 2020) (company personnel allegedly created fake accounts to write fake reviews of company's products on third-party retailer's website);

Shop Tutors, Inc.,

169 F.T.C. 476, 487-89 (2020) (reviews of LendEDU were allegedly fabricated by its employees, other associates, or their friends and published on a third-party website); Complaint at 20,

FTC

v.

Cure Encapsulations, Inc.,

No. 1:19-cv-00982 (E.D.N.Y. Feb. 26, 2019) (Amazon.com reviews of defendants' product were allegedly fabricated by one or more third parties whom defendants had paid to generate reviews); Complaint at 19,

FTC

v.

Genesis Today, Inc.,

1:15-cv-00062 (W.D. Tex. Jan. 26, 2015) (Amazon.com product reviews allegedly purchased by defendants); Complaint at 5, 8,

FTC

v.

Dunlevy,

No. 1:11-cv-01226-TWT (N.D. Ga. Apr. 4, 2011) (alleged fake consumer comments).

State Attorneys General have also brought cases challenging allegedly fabricated consumer reviews.

See, e.g.,

Complaint at 4,

Washington

v.

Alderwood Surgical Ctr., LLC,

No. 2:22-cv-01835 (W.D. Wash. Dec. 29, 2022) (creating allegedly fake positive reviews on Google, Yelp, and other review sites); Complaint at 17-22,

State

v.

Amazon Home Warranty LLC,

No. CV2021-007632 (Ariz. Sup. Ct. Maricopa Cnty. May 10, 2021) (disseminated or caused the dissemination of allegedly fake favorable consumer reviews on third-party review websites, including on the

BBB's website); Assurance of Voluntary Compliance at 3-5,

State

v.

Unified Holding Grp., LLC,

No. 2020-06785 (C.P. Cumberland Cnty., Pa. Dec. 16, 2020) (alleged fabricated reviews on the BBB website); Complaint at 15,

State

v.

US Air Ducts & Sky Builders, Inc.,

No. 19-2-24757-6-SEA (Wash. Sup. Ct. Kings Cnty., Sept. 20, 2019) (allegedly created fake Google reviews); Complaint at 8-9,

State

v.

Mechs. Heating & Air Conditioning, LLC,

No. 13108809 (Ga. Sup. Ct. Cobb Cnty., Oct. 11, 2013) (alleged fake favorable customer reviews, including on Yelp.com, Kudzu.com, and Google+Local.com). In September 2013, the New York Attorney General's office announced settlements with 19 companies that allegedly either purchased fake reviews or arranged to have fake reviews posted for their clients.

163

163

Press Release,

A.G. Schneiderman Announces Agreement With 19 Companies To Stop Writing Fake Online Reviews And Pay More Than $350,000 In Fines,

Sept. 23, 2013,

https://ag.ny.gov/press-release/2013/ag-schneiderman-announces-agreement-19-companies-stop-writing-fake-online-reviews.

Numerous private lawsuits have involved purportedly fake consumer reviews.

See, e.g., BHRS Grp., LLC

v.

Brio Water Tech., Inc.,

553 F. Supp. 3d 793, 797 (C.D. Cal. 2021) (defendant allegedly enlisted individuals to purchase products for the purpose of leaving positive Amazon.com reviews of its products and negative Amazon.com reviews of plaintiff's competing products);

Marksman Sec. Corp.

v.

P.G. Sec.,

No. 19-62467-CIV-CAN, 2021 U.S. Dist. LEXIS 196580, at *43 (S.D. Fla. Oct. 12, 2021) (denying plaintiff's motion for default and granting in part its motion for summary judgment in a case in which defendants paid for positive Google reviews from at least three individuals who never lived in a building that a defendant serviced);

Rubinstein

v.

Ourian,

No. 20-21948-CIV-MORE, 2021 U.S. Dist. LEXIS 171799, at *3-4 (S.D. Fla. Sep. 10, 2021) (order granting motions for summary judgment on claims and counterclaims in a case in which defendant allegedly purchased negative reviews of plaintiff plastic surgeon);

RingCentral, Inc.

v.

Nextiva, Inc.,

No. 19-cv-02626-NC, 2021 U.S. Dist. LEXIS 114042, at *7-8 (N.D. Cal. June 17, 2021) (order denying plaintiff's motion for summary judgment, and granting in part defendant's motion for summary judgment in a case in which plaintiff alleged defendant posted fake positive reviews for itself and fake negative reviews of the plaintiff, and defendant made similar allegations about plaintiff);

AlphaCard Sys. LLC

v.

Fery LLC,

Civil Action No. 19-20110 (MAS) (TJB), 2020 U.S. Dist. LEXIS 147059, at *2 (D.N.J. Aug. 14, 2020) (denying defendant's motion to dismiss in a case in which defendant allegedly “placed” hundreds of phony Amazon.com customer reviews on defendant's products);

Stonecoat of Tex., LLC

v.

Procal Stone Design, LLC,

Civil Action No. 4:17CV303, 2019 U.S. Dist. LEXIS 153115, at *7-8 (E.D. Tex. July 25, 2019) (denying motions for summary judgment on claims and counterclaims, and denying motion to strike attachments in a case in which plaintiffs allegedly directed employees and/or representatives to submit fake complaints/negative reviews about defendant and post fake positive reviews about plaintiff);

Super Mario Plumbing

v.

Belodedov,

No. 2:17-cv-02545-TLN-AC, 2018 U.S. Dist. LEXIS 24514, at *1-3 (E.D. Cal. Feb. 14, 2018) (denying motion for preliminary injunction in a case in which defendant allegedly posted fake negative reviews about competitor plaintiff);

SA Luxury Expeditions LLC

v.

Latin Am. for Less, LLC,

No. C 14-04085 WHA, 2014 U.S. Dist. LEXIS 159520, at *1-2 (N.D. Cal. Nov. 12, 2014) (motion to dismiss held in abeyance in a case in which defendant allegedly posted fake negative consumer reviews about competitor plaintiff).

The problem of fake reviews is not limited to the United States. Regulators in other countries, including Canada, the United Kingdom, and Germany, as well as international bodies like the Organisation for Economic Co-operation and Development (OECD), have all stated fake reviews are a growing, thriving, or substantial marketplace problem.

164

The extent of fake reviews outside of the United States lends additional support to the conclusion that fake reviews are prevalent, but the Commission is not determining prevalence based upon such facts.

164

See, e.g.,

Competition Bureau Canada, “Honest Advertising in the Digital Age,” Jan. 22, 2020,

https://www.canada.ca/en/competition-bureau/news/2020/01/honest-advertising-in-the-digital-age.html;

UK Competition and Markets Authority, “CMA expects Facebook and eBay to tackle sale of fake reviews,” June 21, 2019,

https://www.gov.uk/government/news/cma-expects-facebook-and-ebay-to-tackle-sale-of-fake-reviews;

Germany Federal Cartel Office, “Bundeskartellamt launches sector inquiry into user reviews,” May 23, 2019,

https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2019/23_05_2019_SU_Nutzerbwertungen.html;

OECD, “Understanding Online Ratings and Reviews” at 14-15 (2019),

https://www.oecd-ilibrary.org/science-and-technology/understanding-online-consumer-ratings-and-reviews_eb018587-en;

OECD, “Good Practice Guide on Online Consumer Ratings and Reviews” at 6 (2019) (noting evidence that some businesses post fake reviews “on a large scale”),

http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=DSTI/CP(2019)5/FINAL&docLanguage=En.

The Commission has also challenged allegedly fictitious consumer testimonials that appear in advertising.

See, e.g.,

Complaint at 15, 17-18,

FTC

v.

Wellco, Inc.,

No. 1:21-cv-02081 (S.D.N.Y. Mar. 10, 2021) (testimonials allegedly copied from competitors' websites);

Shop Tutors, Inc.,

169 F.T.C. 476, 488-89 (2020) (allegedly fabricated testimonials); Complaint at 14, 19,

FTC

v.

A.S. Resch., LLC (Synovia),

No. 1:19-cv-3423 (D. Colo. Dec. 5, 2019) (allegedly fake testimonials); Complaint at 20-22, 31,

FTC

v.

Global Cmty. Innovations LLC,

No. 5:19-CV-00788 (N.D. Ohio Apr. 10, 2019) (allegedly fake testimonials); Complaint at 12, 18,

FTC

v.

Fat Giraffe Mktg. Grp. LLC,

No. 2:19-cv-00063-CW (D. Utah Jan. 29, 2019) (the people featured in testimonials allegedly were not real customers);

FTC

v.

Cardiff,

No. ED 18-cv-02104-DMG (PLAx), 2020 U.S. Dist. LEXIS 210930, at *15-16 (C.D. Cal. Oct. 9, 2020) (granting in part FTC motion for summary judgment and finding testimonialists in infomercial had not used the product); Complaint at 12-13, 20,

FTC

v.

Mktg. Architects, Inc.,

No. 2:18-cv-00050-NT (D. Me. Feb. 5, 2018) (allegedly fake testimonials); Complaint at 14, 21,

FTC

v.

Health Res. Labs., LLC,

No. 2:17-cv-00467-JDL (D. Me. Nov. 30, 2017) (allegedly fake testimonials); Complaint at 13, 18, 28,

FTC

v.

XXL Impressions LLC,

No. 1:17-cv-00067-NT (D. Me. Feb. 22, 2017) (defendants allegedly did not know whether consumer endorsers of their products who appeared in their ads actually exist); Complaint at 5, 7, 12-13,

FTC

v.

Anthony Dill,

No. 2:16-cv-00023-GZS (D. Me. Jan. 19, 2016) (allegedly fake testimonials); First Amended Complaint at 75,

FTC

v.

Jeremy Johnson,

No. 10-cv-2203-RLH (GWF) (D. Nev. Feb. 25, 2013) (defendants allegedly hired third parties to post fake positive online articles and web pages purportedly by consumers who had successfully used defendants' product to find government grants);

FTC

v.

Grant Connect, LLC,

827 F. Supp. 2d 1199, 1228 (D. Nev. 2011) (granting summary judgment on FTC's deception count where defendants presented no evidence showing certain testimonials were genuine),

aff'd in part and vacated in part on other grounds,

763 F.3d 1094 (9th Cir. 2014);

Buckingham Prods., Inc.,

106 F.T.C. 116 (1985) (testimonials allegedly do not represent actual and genuine testimonials from customers);

Technobrands, Inc.,

133 F.T.C. 647, 650, 654-55 (2002) (purported consumer endorsers allegedly did not exist);

Plaza

Club, Inc.,

80 F.T.C. 62 (1972) (testimonialist allegedly was not a member of respondents' physical fitness facilities and unknown to respondents);

New Standard Publ'g. Co., Inc.,

47 F.T.C. 1350, 1366 (1951) (some of the testimonials and letters recommending encyclopedia allegedly were not genuine).

The use of fake celebrity endorsements is widespread. A 2018 Better Business Bureau in-depth investigative study found many celebrity endorsements are fake.

165

According to one news report, Ellen DeGeneres and Sandra Bullock both sued 100 anonymous defendants who fraudulently used their names in promoting an anti-aging serum and weight-loss products, and dozens of other celebrities' names have been misappropriated in similar fashion.

166

165

See

Better Business Bureau, Subscription Traps and Deceptive Free Trials Scam Millions with Misleading Ads and Fake Celebrity Endorsements (Dec. 12, 2018),

https://www.bbb.org/article/investigations/18929-subscription-traps-and-deceptive-free-trials-scam-millions-with-misleading-ads-and-fake-celebrity-endorsements.

166

See

Randy Hutchinson, Opinion, Endorsements by stars such as Ellen DeGeneres and Sandra Bullock might be fake, The Tennessean, Jan. 8, 2020,

https://www.tennessean.com/story/opinion/2020/01/08/celebrity-endorsement-of-products-could-be/2834860001/.

The FTC has challenged numerous allegedly false claims that specific celebrities endorsed certain products, services, or businesses.

See, e.g.,

Complaint at 22-23, 27-28, 38-39,

FTC

v.

Effen Ads, LLC,

No. 2:19-cv-00945-RJS (D. Utah Nov. 26, 2019); Complaint at 15, 19-20, 30-31,

Global Cmty. Innovations LLC,

No. 5:19-CV-00788 (N.D. Ohio Apr. 10, 2019); Complaint at 5, 18-20, 22-23, 36,

FTC

v.

Tarr, Inc.,

No. 3:17-cv-02024-LAB-KSC (S.D. Cal. Oct. 3, 2017); Complaint at 12-13,

FTC

v.

Tachht, Inc.,

No. 8:16-cv-01397-JDW-AEP (M.D. Fla. June 1, 2016); Complaint at 13-15, 18,

FTC

v.

Sales Slash, LLC,

No. CV15-03107 (C.D. Cal. Apr. 27, 2015); Complaint at 2, 4-5,

Norm Thompson Outfitters, Inc.,

No. C-4495 (Sept. 29, 2014); Complaint at 15-17,

FTC

v.

Central Coast Nutraceuticals, Inc.,

No. 10 C 4931 (N.D. Ill. Aug. 5, 2010);

The Raymond Lee Org., Inc.,

92 F.T.C. 489 (1978) (use of the names, photographs, and words of public officials, including members of Congress, allegedly misled consumers that the officials recommended or endorsed the business). Most recently, FTC staff published a blog post to warn consumers about scammers using fake Shark Tank celebrity testimonials and endorsements.

167

167

See

Karen Hobbs, Did your favorite Shark Tank celebrity really endorse THAT? Probably not, Fed. Trade Comm'n Consumer Blog (Feb. 17, 2023),

https://consumer.ftc.gov/consumer-alerts/2023/02/did-your-favorite-shark-tank-celebrity-really-endorse-probably-not.

Consumer reviews and testimonials that are not entirely fabricated can still misrepresent the experiences of the purported reviewers and testimonialists, and such misrepresentations are prevalent. This conclusion is reflected in NAIMA's comment, which asserted testimonials by those misrepresenting their experiences with products are plentiful.

The Commission has challenged many advertisements that allegedly misrepresented endorsers' experiences.

See, e.g., FTC

v.

Cardiff,

2020 U.S. Dist. LEXIS 210930, at *15-16, 48 (testimonialists had already lost weight without using the product); Complaint at 14, 18,

FTC

v.

A.S. Resch., LLC (Synovia),

No. 1:19-cv-3423 (testimonialists had allegedly used a prior product formulation that contained substantially different ingredients); Complaint at 22, 25,

NextGen Nutritionals, LLC,

No. 8:17-cv-2807-T-36AEP (M.D. Fla. Jan. 9, 2018) (testimonials in ads allegedly did not represent the actual experiences of customers); Complaint at 22-24, 27,

FTC

v.

Russell T. Dalbey,

No. 1:11-cv-01396-CMA-KLM (D. Colo. May 26, 2011) (testimonials allegedly misrepresented earnings from brokering promissory notes using defendants' system);

FTC

v.

Data Med. Capital, Inc.,

No. SA CV 99-1266 AHS (EEx), 2010 U.S. Dist. LEXIS 3344, *27 (C.D. Cal. Jan. 15, 2010) (testimonial for one defendant recycled as a fictitious testimonial for a different defendant); Complaint at 17,

FTC

v.

Advanced Patch Techs., Inc.,

No. 104-CV-0670 (N.D. Ga. Mar. 10, 2004) (allegedly testimonialists attributed their weight loss to simply wearing the Pound A Patch but were also provided supervised exercise sessions three times per week);

Esrim Ve Sheva Holding Corp.,

132 F.T.C. 736, 740 (2001) (testimonial from respondent allegedly did not represent his actual findings and experience with the product);

Computer Bus. Servs., Inc.,

123 F.T.C. 75, 78-79 (1997) (testimonials by purchasers of home-based business ventures allegedly did not reflect their actual experiences);

Twin Star Prods., Inc.,

113 F.T.C. 847, 853-54 (1990) (endorsement allegedly did not reflect the honest opinions, findings, beliefs, or experience of the endorser);

National Sys. Corp.,

93 F.T.C. 58, 63-65 (1979) (some testimonials were allegedly untrue);

Federated Sanitary Corp.,

85 F.T.C. 130, 133 (1975) (alleging testimonials represented to be from salesmen, franchisees, or other distributors of respondents' products were not made by such individuals, and a substantial number of purported testimonialists had never dealt with the respondents);

Natpac, Inc.,

79 F.T.C. 454, 459 (1971) (testimonial letters were allegedly prepared by respondents and signed before the purported authors had received the products and had time to evaluate them);

P. Lorillard Co.,

46 F.T.C. 735, 740 (1950) (alleging testimonials did not present or reflect the actual personal experiences, knowledge, or beliefs of the signers; some testimonialists did not smoke Old Gold cigarettes or any cigarettes; many testimonials were prewritten by respondent's representatives; and many were known by the respondent to be false);

R.J. Reynolds Tobacco Co.,

46 F.T.C. 706, 731-32 (1950) (endorsements communicated that endorsers exclusively smoked Camel cigarettes when they did not smoke cigarettes, did not smoke Camels exclusively, or could not tell the difference between Camels and other cigarettes).

Accordingly, based on the foregoing evidence, the Commission concludes fake consumer reviews and testimonials, as well as reviews and testimonials that otherwise misrepresent the experiences of the reviewers and testimonialists, are prevalent.

B. Consumer Review or Testimonial Reuse or Repurposing

One type of review deception known as “review hijacking” or “review reuse fraud” appears to primarily or solely affect online marketplaces with third-party sellers, such as Amazon.com. Vendors and third-party sellers on Amazon's platform can make their own modifications to product pages, or request Amazon's assistance to do so, using features referred to as “product merging” and “product variation.” Products that are substantially similar and that differ only in narrow, specific ways—such as color, size, or quantity—but that do not alter the core essence of the item, such as a shirt that comes in multiple colors and different sizes, may share a variation relationship. Products in a variation relationship share the same product detail page. Each product will appear as an alternative on the product detail page, and, when a shopper selects a different product in the variation relationship, the content of the product detail page, such as the pictured product, may change. The variation relationship enables buyers to compare and choose among product attributes from a single product detail

page, thereby facilitating customer choice and ease of shopping.

Some vendors and sellers abuse these features by repurposing a listing page for a product that has positive reviews (

e.g.,

a shower caddy or a jar of honey) and using it to sell a completely unrelated product (

e.g.,

a phone charger or a neck brace), thus inflating the star rating for the latter—and going unnoticed unless consumers read the individual reviews closely. By repurposing the page, the review hijacker is implicitly misrepresenting the repurposed reviews are for the second product and the product has more ratings and reviews than it does. The review hijacker may also be misrepresenting that the second product has a higher average star rating or that it has earned “Best Seller” or “Amazon's Choice” badges. These claims are unquestionably deceptive and of no redeeming value to legitimate marketers.

This problem has persisted since at least 2018 and is prevalent as reflected in reporting by Consumer Reports, The Verge, Buzzfeed News, and others.

168

The reporting provides many examples of review hijacking found on Amazon.com across multiple product categories. The author of the Consumer Reports article stated that experts believe it is an “acute problem” and some legitimate Amazon.com sellers are overwhelmed with fighting it. The Verge article calls it a “common tactic” and quotes a former Amazon employee as saying “the problem is way bigger than people realize.”

169

168

See, e.g.,

Sara Spary, “Top-rated Amazon headphones boosted by `fake reviews' for toys, mugs and umbrellas,” WHICH?, Apr. 7, 2022,

https://www.which.co.uk/news/2022/04/top-rated-amazon-headphones-boosted-by-fake-reviews-for-toys-mugs-and-umbrellas/;

Timothy B. Lee, “Amazon still hasn't fixed its problem with bait-and-switch reviews,” ARS Technica, Dec. 20, 2020,

https://arstechnica.com/tech-policy/2020/12/amazon-still-hasnt-fixed-its-problem-with-bait-and-switch-reviews/;

Jon Keegan, “Is This Amazon Review Bullshit?,” The Markup, July 21, 2020,

https://themarkup.org/ask-the-markup/2020/07/21/how-to-spot-fake-amazon-product-reviews;

Josh Dzieza, “Even Amazon's own products are getting hijacked by imposter sellers,” The Verge, Aug. 29, 2019,

https://www.theverge.com/2019/8/29/20837359/amazon-basics-fake-sellers-imposters-third-party-marketplace;

Jake Swearingen, “Hijacked Reviews on Amazon Can Trick Shoppers,” Consumer Rep., Aug. 26, 2019,

https://www.consumerreports.org/customer-reviews-ratings/hijacked-reviews-on-amazon-can-trick-shoppers/;

Nicole Nguyen, “Here's Another Kind Of Review Fraud Happening On Amazon,” Buzzfeed News, May 29, 2018,

https://www.buzzfeednews.com/article/nicolenguyen/amazon-review-reuse-fraud.

169

See

Dzieza, “Even Amazon's own products are getting hijacked by imposter sellers,”

supra

note 168.

The Commission recently brought its first case involving this type of review deception, suing a large vendor that boosted its newly launched products on Amazon.com. The vendor allegedly had Amazon establish variation relationships between the newer products and successful, established products that had more ratings and reviews, high average ratings, or “Amazon's Choice” or “#1 Best Seller” badges.

See

Complaint at 1-6,

The Bountiful Co.,

No. C-4791 (Apr. 10, 2023).

Accordingly, based on the foregoing evidence, the Commission concludes that the unfair or deceptive reuse or repurposing of consumer reviews is prevalent.

C. Buying Positive or Negative Consumer Reviews

It is common for sellers or their agents to give incentives in exchange for reviews with the incentives conditioned on the sentiment of the reviews. In the review markets discussed in the comments and described above, prospective reviewers are offered free merchandise or money in exchange for 5-star reviews. Social media groups for procuring misleading reviews are prevalent. The UC Davis researchers found 242,000 products for which Amazon sellers solicited incentivized five-star Amazon.com reviews.

170

In addition, the comment from the small business employee said that a competitor of the company for which she worked is providing incentives for 5-star reviews.

171

In another academic study, UCLA researchers analyzed these review markets and resulting reviews on Amazon.com and found the market for fake reviews is large and the practice of buying and selling reviews is widespread.

172

170

Oak & Shafiq Cmt. at 3-4.

171

Provencal Cmt. at 1.

172

He et al., “The Market for Fake Reviews,”

supra

note 158.

The Commission has brought cases in which a marketer allegedly provided an incentive for a review or endorsement that was required to be positive.

See, e.g.,

Complaint at 14, 19-20,

FTC

v.

A.S. Resch., LLC (Synovia),

No. 1:19-cv-3423 (allegedly offered consumer endorsers free product in exchange for “especially positive and inspiring” reviews); Complaint at 5-6, 8,

Urthbox, Inc.,

No. C-4676 (Apr. 3, 2019) (allegedly provided compensation for the posting of positive reviews on the BBB's website and other third-party websites); Complaint at 2-3,

AmeriFreight, Inc.,

No. C-4518 (Feb. 27, 2015) (allegedly past customers were regularly encouraged to submit reviews of respondent's services in order to be eligible for a $100 “Best Monthly Review Award” given to “the review with the most captivating subject line and best content” and told that they should “be creative and try to make your review stand out for viewers to read!”).

Such conduct has also been challenged in private actions.

See, e.g., Marksman Sec. Corp.

v.

P.G. Sec.,

2021 U.S. Dist. LEXIS 196580, at *4 (denying plaintiff's motion for default and granting in part its motion for summary judgment in which it was undisputed that defendants paid for positive Google reviews).

Regulators in Australia, Denmark, and the United Kingdom, as well as the OECD, have issued guidelines or business guidance indicating companies should not provide incentives for giving positive reviews.

173

While it may lend some additional support to the conclusion that the acts or practices are prevalent, the Commission is not concluding that the conduct is prevalent on the basis that other countries have taken actions.

173

Australian Competition & Consumer Commission,

Online product and service reviews, https://www.accc.gov.au/business/advertising-and-promotions/online-product-and-service-reviews;

Danish Consumer Ombudsman,

Guidelines on publication of user reviews

(2015),

https://www.forbrugerombudsmanden.dk/media/49717/guidelines.pdf;

United Kingdom Competition and Markets Authority,

Online reviews and endorsements

(2015),

https://www.gov.uk/government/consultations/online-reviews-and-endorsements;

OECD,

Good Practice Guide on Online Consumer Ratings and Reviews

(2019),

https://www.oecd-ilibrary.org/science-and-technology/good-practice-guide-on-online-consumer-ratings-and-reviews_0f9362cf-en.

Accordingly, based on the foregoing evidence, the Commission concludes the giving of incentives for reviews conditioned on the sentiment of the reviews is prevalent.

D. Insider Consumer Reviews and Testimonials

It is quite common for a company's owners, officers, managers, executives, employees, agents, or their relatives, to write consumer reviews or testimonials of its products or services.

According to Trustpilot, in 2021, more than 8,000 reviews for U.S. businesses were written by their owners, officers, or employees, or their family members.

174

In addition, an individual commenter complained of having relied upon misleading reviews written by a business's employees or their spouses before selecting an auto repair shop.

175

174

Trustpilot Cmt. at 7.

175

Monday Cmt. at 1.

The Commission has challenged numerous instances of deceptive

reviews allegedly written by company insiders.

See, e.g.,

Complaint at 21, 26-27,

United States

v.

Vision Path, Inc.,

No. 1:22-cv-00176 (D.D.C. Jan. 25, 2022) (allegedly an executive of the company wrote a review on a third-party site and the co-CEO posted a public response thanking the reviewer); Complaint at 2-4,

Sunday Riley Modern Skincare, LLC,

No. C-4729 (Nov. 6, 2020) (company owner and managers allegedly asked company employees to write product reviews on third-party retailer's website);

Creaxion Corp.,

167 F.T.C. 71, 78-79 (2019) (company allegedly conducted program that reimbursed individuals, including the CEO and other company employees, for purchasing its product and posting online reviews); Complaint at 5-6, 8-9,

Mikey & Momo, Inc.,

No. C-4655 (May 3, 2018) (Amazon.com reviews allegedly written by company officer and her relatives); Complaint at 10, 12,

FTC

v.

Aura Labs, Inc.,

No. 8:16-cv-02147 (C.D. Cal. Dec. 12, 2016) (app store review and website testimonials allegedly written by CEO or relatives of Chairman); Complaint at 25-27, 32-33,

FTC

v.

Universal City Nissan, Inc.,

No. 2:16-cv-07329 (C.D. Cal. Sept. 29, 2016) (customer reviews on third-party websites allegedly written by managers); Complaint at 10,

United States

v.

Spokeo, Inc.,

No. 2:12-cv-05001-MMM-SH (C.D. Cal. June 7, 2012) (allegedly defendant directed its employees to draft and post comments endorsing its products on news and technology websites; and comments were reviewed and edited by managers and then posted using account names provided by defendant);

Reverb Commc'ns, Inc.,

150 F.T.C. 782, 783-84 (2010) (owner of public relations agency, her managers, and employees allegedly wrote iTunes store reviews for clients' games).

At least one State Attorney General has challenged alleged insider reviews.

See

Complaint at 15,

State

v.

US Air Ducts & Sky Builders, Inc.,

No. 19-2-24757-6-SEA (Wash. Sup. Ct. Kings Cnty., Sept. 20, 2019) (Google reviews allegedly written by employees, relatives of employees, and the business owner).

The Commission has also challenged testimonials allegedly written by insiders in numerous instances.

See, e.g.,

Complaint at 15, 19-20,

FTC

v.

Health Ctr., Inc.,

No. 2:20-cv-00547 (D. Nev. Mar. 19, 2020) (defendants allegedly used testimonials from their employees that purported to be from ordinary consumers); Complaint at 14, 19,

FTC

v.

A.S. Resch., LLC (Synovia),

No. 1:19-cv-3423 (D. Colo. 2019) (ads allegedly included a testimonial by a 50 percent owner and officer); Complaint at 21, 25-26,

FTC

v.

NutriMost LLC,

No. 2:17-cv-00509-NBF (W.D. Pa. Apr. 20, 2017) (testimonials in ads were allegedly from licensees or franchisees, their relatives, or their employees);

Deutsch LA, Inc.,

159 F.T.C. 1163, 1168-69 (2015) (public relations firm allegedly asked employees to tweet about client's product); Complaint at 19, 21,

FTC

v.

Genesis Today, Inc.,

No. 1:15-cv-00062 (W.D. Tex. Jan. 26, 2015) (allegedly defendants' promotional materials linked to video testimonials by purported users of their weight-loss products that were provided by their employees); Complaint at 17,

FTC

v.

Advanced Patch Techs., Inc.,

No. 104-CV-0670 (N.D. Ga. Mar. 10, 2004) (shopping mall segment of infomercial with testimonials from “real people” allegedly included at least one employee of the defendants or their agents);

Brake Guard Prods., Inc.,

125 F.T.C. 138, 191 (1998) (published testimonial was allegedly from a dealer/distributor of the product);

Gisela Flick,

116 F.T.C. 1108, 113-14 (1993) (alleged infomercial endorsement by company's Athletic Director);

Cliffdale Assocs., Inc.,

103 F.T.C. 110, 144-45, 172 (1984) (testimonials were allegedly by business associates or relatives).

Accordingly, based on the foregoing evidence, the Commission concludes the use of consumer reviews and testimonials written by company insiders—that is, consumer reviews and testimonials written by a company's owners, officers, managers, executives, employees, agents, or their relatives—is prevalent.

E. Company-Controlled Review Websites or Entities

Numerous businesses have set up purportedly independent websites, organizations, or entities that review or endorse their own products.

In numerous cases, the Commission has challenged sellers who allegedly misrepresented that the websites they controlled provided independent opinions of products.

See, e.g.,

Complaint at 2, 8-9,

Son Le,

No. C-4619 (May 31, 2020) (respondents allegedly operated purportedly independent websites that reviewed their own trampolines);

FTC

v.

Roca Labs, Inc.,

345 F. Supp. 3d 1375, 1389-90 (M.D. Fla. 2018) (defendants operated a purportedly independent, objective website that endorsed defendants' products); Complaint at 21-25, 28,

FTC

v.

NourishLife, LLC,

No. 1:15-cv-00093 (N.D. Ill. Jan. 7, 2015) (defendants allegedly operated a purportedly independent, scientific research website that endorsed a supplement sold only by defendants).

The Commission has also challenged sellers who allegedly created purportedly independent organizations or entities that supposedly reviewed or approved their products or services.

See, e.g.,

Complaint at 3-5,

Bollman Hat Co.,

No. C-4643 (Jan. 23, 2018) (respondents allegedly created a U.S.-origin seal misrepresenting that an independent organization endorsed their products as made in the United States); Complaint at 18-20, 26,

NextGen Nutritionals, LLC,

No. 8:17-cv-2807-T-36AEP (M.D. Fla. Jan. 9, 2018) (alleged misrepresentation that sites displaying the Certified Ethical Site Seal were verified by an independent, third-party program); Complaint at 2-4,

Moonlight Slumber, LLC,

No. C-4634 (Sept. 28, 2017) (respondent represented its baby mattresses had been certified by Green Safety Shield and failed to disclose the shield was its own designation); Complaint at 4-6,

Benjamin Moore & Co., Inc.,

No. C-4646 (July 11, 2017) (respondent allegedly used a “Green Promise” seal of its own creation to misrepresent that paints had been endorsed or certified by independent third party); Complaint at 2-4,

ICP Constr. Inc.,

No. 4648 (July 11, 2017) (same); Complaint at 2-3,

Ecobaby Organics, Inc.,

No. C-4416 (July 25, 2013) (manufacturer allegedly misrepresented that seal-providing association was an independent, third-party certifier when it created and controlled that association);

Nonprofit Mgmt. LLC,

151 F.T.C. 144, 148-49 (2011) (respondents allegedly misrepresented their seal program was endorsed by two independent associations when respondents owned and operated them); Complaint at 34, 37,

FTC

v.

A. Glenn Braswell,

No. 2:03-cv-03700-DT-PJW (C.D. Cal. May 27, 2003) (defendants allegedly established the Council on Natural Nutrition and then misrepresented it was an independent organization of experts who had endorsed defendants' products);

Nat'l Media Corp.,

116 F.T.C. 549, 559-60 (1993) (respondents allegedly claimed the National Association of Advertising Producers was an existing, independent organization that evaluates commercials for their integrity and excellence);

Revco, D.S., Inc.,

67 F.T.C. 1158, 1163, 1208-18, 1250-51 (1965) (respondents allegedly created and controlled Consumer Protective Institute and gave their products its seal of approval).

Accordingly, based on the foregoing evidence, the Commission concludes the practice of marketers setting up

purportedly independent websites, organizations, or entities to review or endorse their own products is prevalent.

F. Review Suppression

The ANPR addressed two types of review suppression. One type involves a seller's website representing that the consumer reviews displayed represent most or all of the reviews submitted when, in fact, reviews are being suppressed based upon their negativity. Trustpilot commented that it was aware of the suppression of negative reviews on retailer or business websites.

176

176

Trustpilot Cmt. at 8.

In a recent case, the Commission alleged a retailer suppressed hundreds of thousands of 1-, 2-, and 3-star reviews submitted to its website.

See

Complaint at 1-2,

Fashion Nova, LLC,

No. C-4759 (Mar. 18, 2022). Staff also publicly addressed this issue in a 2020 closing letter to Yotpo, the company that provided review management services to Fashion Nova and numerous other merchants.

177

FTC staff's investigation of Yotpo revealed more than 4,500 Yotpo merchant clients were only automatically publishing 4- or 5-star reviews. Of the 1-star reviews submitted to merchants not automatically publishing 1-star reviews, just 21 percent were published; and of the 2-star reviews submitted to merchants not automatically publishing 2-star reviews, just 31 percent were published. After FTC staff began investigating Yotpo, it implemented clear and prominent guidance to its clients on their need to promptly post reviews, including negative reviews, and began to automatically post negative reviews that have not been promptly reviewed and acted upon by its clients.

178

177

See https://www.ftc.gov/system/files/documents/closing_letters/nid/202_3039_yotpo_closing_letter.pdf.

178

See id.

Foreign consumer protection entities have brought several actions involving companies that prevented the publication of negative reviews. An online health-care booking service in Australia, which published patient reviews, admitted it did not publish approximately 17,000 reviews and edited another 3,000 reviews either to remove negative aspects or to embellish positive aspects.

179

An Australian court found a home building company held back bad reviews from its review websites to give a more favorable impression of its services.

180

A New Zealand holiday home rental website pleaded guilty and was fined for removing negative comments about rental properties and its maintenance and management of them, and not publishing any reviews that gave a rating below 3.5 stars.

181

The United Kingdom's Competition and Markets Authority secured an undertaking from an online knitwear retailer that did not publish all genuine, relevant, and lawful reviews submitted by its customers.

182

The problem is sufficiently prevalent that an EU Directive prohibits “publishing only positive reviews and deleting the negative ones.”

183

These foreign actions lend additional support to the conclusion that the conduct is prevalent, but the Commission is not determining prevalence based upon such actions.

179

See

“HealthEngine to pay $2.9 million for misleading reviews and patient referrals,” Australian Competition & Consumer Commission, Aug. 20, 2020,

https://www.accc.gov.au/media-release/healthengine-to-pay-29-million-for-misleading-reviews-and-patient-referrals.

180

See

“Aveling Homes ordered to pay penalties of $380,000 for misleading review websites,” Australian Competition & Consumer Commission, Nov. 30, 2017,

https://www.accc.gov.au/media-release/aveling-homes-ordered-to-pay-penalties-of-380000-for-misleading-review-websites.

181

See

“Bachcare fined for removing negative comments in online reviews,” RNZ, Dec. 20, 2019,

https://www.rnz.co.nz/news/national/405929/bachcare-fined-for-removing-negative-comments-in-online-reviews.

182

See

“Retailer hosting reviews on its website: improvement of practices,” Competition and Markets Authority, Aug. 11, 2016,

https://www.gov.uk/cma-cases/retailer-hosting-reviews-on-its-website-improvement-of-practices.

183

See

Directive (EU) 2019/2161 of the European Parliament and of the Council, Nov. 27, 2019,

https://eur-lex.europa.eu/eli/dir/2019/2161/oj.

The other type of review suppression addressed in the ANPR is suppression by unjustified legal threat or physical threat. The comments in response to the ANPR support a determination that such review suppression is prevalent. Yelp said it “constantly confronts” the use of “abusive and questionable or unjustified legal threats” to suppress reviews.

184

Trustpilot has seen cases, mostly outside of the United States, where businesses have threatened consumers if they do not delete negative reviews.

185

A comment from Ubiquitous Advising described a company in its local area that is constantly threatening and bullying reviewers in order to suppress bad reviews.

186

The Transparency Company said every year thousands of lawyers are hired to intimidate the authors of negative reviews.

187

184

Yelp Cmt. at 11.

185

Trustpilot Cmt. at 9.

186

Ubiquitous Advising Cmt. at 1.

187

Transparency Company Cmt. at 16.

In a case against Roca Labs, Inc., the Commission successfully challenged as unfair the defendants' threats to enforce—and their actual enforcement of—non-disparagement clauses in form contracts that were intended to suppress customers' negative reviews.

188

A subsequent FTC case against World Patent Marketing challenged alleged review suppression through physical intimidation as unfair.

189

According to the Commission's complaint, the defendants in that matter “cultivate[d] a threatening atmosphere through emails to would-be complainants.”

190

For example, they distributed, through an email to all of their then-existing customers, a blog post discussing an incident that purportedly occurred in their offices: A consumer that allegedly wanted to speak with them about an invention idea was stopped, detained, and expelled by their “intimidating security team, all ex-Israeli Special Ops and trained in Krav Maga, one of the most deadly of the martial arts.”

191

The post continued: “The World Patent Marketing Security Team are the kind of guys who are trained to knockout first and ask questions later.”

188

See FTC

v.

Roca Labs, Inc.,

345 F. Supp. 3d at 1393-96.

189

See

Complaint at 8-10, 12,

FTC

v.

World Patent Mktg., Inc.,

No. 1:17-cv-20848-DPG (S.D. Fla. Mar. 6, 2017).

190

Id.

at 9.

191

Id.

State Attorneys General have also challenged the alleged use of unjustified legal threats in attempts to have consumers remove negative reviews.

See, e.g.,

Complaint at 12-14,

Maine

v.

Liberty Bell Moving & Storage, Inc.,

2:2022cv00204 (D. Me. July 8, 2022); Complaint at 4,

Washington

v.

Alderwood Surgical Ctr., LLC.

A State Attorney General challenged the alleged suppression of negative reviews through intimidation, albeit not physical intimidation, and false accusations. Complaint at 5-7 and Appendices A-C,

State

v.

Mechs. Heating & Air Conditioning, LLC

(alleged publishing or threatening to publish the names, home addresses, telephone numbers, email addresses, and photographs of consumers who wrote negative reviews, together with accusations that the consumers engaged in illegal or unethical activities or otherwise maligning their character).

Accordingly, based on the foregoing evidence, the Commission concludes the types of review suppression discussed above are prevalent.

G. Indicators of Social Media Influence

In order to sell or market themselves or their products or services, some individuals and businesses misrepresent their social media influence by buying fake followers, fake subscribers, fake

views, and other similar inauthentic indicators of such influence.

The Commission addressed the alleged sale and use of such fake and inauthentic indicators in complaint against Devumi, LLC.

192

In that matter, the Commission alleged that, by selling and distributing these indicators to users of various social media platforms, the defendants enabled the purchasers to “exaggerate and misrepresent their social media influence,” thereby providing the means and instrumentalities for the purchasers to engage in deception.

193

For example, the defendants allegedly sold fake Twitter followers to actors, athletes, musicians, writers, and other individuals who wanted to increase their appeal as influencers and to motivational speakers, law firm partners, investment professionals, experts, and other individuals who wanted to boost their credibility to potential clients for their services.

194

192

See

Complaint at 5,

FTC

v.

Devumi, LLC,

No. 9:19-cv-81419-RKA (S.D. Fla. Oct. 18, 2019).

193

Id.

at 5.

194

Id.

at 3-4.

At least one State Attorney General has brought a case challenging the alleged misuse of fake indicators of social media influence for commercial purposes. In December 2022, the Washington State Attorney General filed suit against a plastic surgery provider accused of buying tens of thousands of fake “followers” on Instagram and thousands of fake “likes” on Instagram and other social media to create a false appearance of popularity in its advertising to consumers.

195

195

Complaint at 4,

Washington

v.

Alderwood Surgical Ctr., LLC.

Platforms have also sued the sellers of fake indicators of social media influence. In April 2019, Facebook, Inc., and Instagram LLC sued the operators of websites offering fake engagement services.

196

The operators allegedly used a network of computers or “bots” and Instagram accounts to provide fake “likes,” “views,” and “followers” to their customers' Instagram accounts.

197

196

Complaint at 1, 5-9,

Facebook, Inc.

v.

Arend Nollen,

No. 3:19-cv-02262 (N.D. Cal. Apr. 25, 2019).

197

Id.

at 8-9.

Both the FTC and an independent organization have analyzed bots, inauthentic social media accounts, and fake followers. In July 2020, the Commission issued a report to Congress, titled “Social Media Bots and Deceptive Advertising.”

198

The report stated social media companies have reported removing or disabling billions of inauthentic accounts, the online advertising industry has taken steps to curb bot and influencer fraud, and the computing community is designing sophisticated social bot detection methods, but nonetheless, use of social media bots remains a serious issue.

199

A 2023 report by the NATO Strategic Communications Centre of Excellence, which analyzed the market for inauthentic social media accounts and fake followers for several years, found it is as fast and cheap to buy them now as several years ago, and the platforms' ability to detect and remove them is declining overall.

200

Citing TikTok's own reports that it had removed 1.4 billion fake followers in the second quarter of 2022, the researchers stated the total number of fake followers on that platform during that period was likely much higher, given their experiments found only five percent of all purchased fake engagement was identified and removed in a four-week period. Further, in a 2022 study, researchers found that fake views of YouTube videos are widespread, and that the platform does not correct them quickly.

201

198

See

Social Media Bots and Deceptive Advertising: Federal Trade Commission Report to Congress,

https://www.ftc.gov/system/files/documents/reports/social-media-bots-advertising-ftc-report-congress/socialmediabotsreport.pdf.

199

Id.

at 5.

200

See

NATO Strategic Communications Centre of Excellence, “Social Media Manipulation 2022/2023: Assessing the Ability of Social Media Companies to Combat Platform Manipulation,” January 2023,

https://stratcomcoe.org/publications/social-media-manipulation-20222023-assessing-the-ability-of-social-media-companies-to-combat-platform-manipulation/272. See also

Johan Lindquist and Esther Weltevrede, “Negotiating Authenticity in the Market for Fake Followers on Social Media,” Social Science Research Council, Oct. 5, 2021 (describing ability of manipulation services to evade platform detection),

https://items.ssrc.org/beyond-disinformation/negotiating-authenticity-in-the-market-for-fake-followers-on-social-media/;

Joseph Cox, “All of My TikTok Followers Are Fake,” Vice Motherboard, Aug. 13, 2020 (describing the speed and ease of buying fake followers in bulk),

https://www.vice.com/en/article/z3e8na/get-buy-tiktok-followers-likes-views-cheap-easy.

201

See

Maria Castaldo et al., “Doing data science with platforms crumbs: an investigation into fakes views on YouTube,” Sep. 28, 2022,

https://doi.org/10.48550/arXiv.2210.01096.

Accordingly, based on the foregoing evidence, the Commission concludes the sale and misuse of fake indicators of social media influence for commercial purposes is prevalent.

IV. Reasons for the Proposed Rule on the Use of Consumer Reviews and Testimonials

The Commission believes the proposed Rule will substantially improve its ability to combat certain specified, clearly unfair or deceptive acts or practices involving consumer reviews or testimonials. Although such practices are already unlawful under Section 5 of the FTC Act, which prohibits unfair or deceptive acts or practices, the proposed Rule may increase deterrence against these practices in the first instance and will allow the Commission to seek civil penalties against the violators and more readily obtain monetary redress for their victims. As discussed below, the proposed Rule would accomplish these goals without significantly burdening honest businesses and provide benefits to consumers and honest competitors.

The Commission's objective in commencing this rulemaking is to deter certain clearly unfair or deceptive acts or practices involving consumer reviews or testimonials, and expand the remedies available to it in instances where such practices are uncovered. A recent U.S. Supreme Court decision,

202

which overturned 40 years of precedent from the U.S. Circuit Courts of Appeal uniformly holding that the Commission could take action under Section 13(b) of the FTC Act, 15 U.S.C. 53(b), to return money unlawfully taken from consumers through unfair or deceptive acts or practices, has made it significantly more difficult for the Commission to return money to injured consumers.

203

Without Section 13(b) as it had historically been understood, the only method the Commission has to return money unlawfully taken from consumers is Section 19 of the FTC Act, 15 U.S.C. 57b, which provides two paths for consumer redress. The longer path, under Section 19(a)(2), requires the Commission to first issue a final cease-and-desist order—including any resulting appeal. Then, to recover money for consumers, the Commission must prove separately in Federal court that the violator engaged in fraudulent or dishonest conduct.

204

The shorter path to monetary relief is under Section 19(a)(1), which allows the Commission to recover redress directly through a Federal court action and is available only when the Commission alleges violation of a rule.

205

None of the Commission's cases challenging deceptive consumer reviews or testimonials has involved other misconduct for which the Commission

sought civil penalties under any of the rules it enforces.

202

See AMG Cap. Mgmt., LLC

v.

FTC,

141 S. Ct. 1341, 1352 (2021).

203

See

ANPR, 87 FR at 67425 & n.1 (discussing

AMG Cap. Mgmt.

).

204

See

15 U.S.C. 57b(a)(2) (“If the Commission satisfies the court that the act or practice to which the cease-and-desist order relates is one which a reasonable man would have known under the circumstances was dishonest or fraudulent, the court may grant relief.”).

205

Compare

15 U.S.C. 57b(a)(1) (rule violations),

with id.

57b(a)(2) (Section 5 violations).

In addition, the longer path to redress under Section 19(a)(2) provides relief only to redress consumer injury, which may be difficult to quantify in certain circumstances. By contrast, with a rule violation, the shorter path to redress under Section 19(a)(1) also gives the Commission the ability to obtain civil penalties, which punish the wrongdoer, provide general and specific deterrence, and do not require quantifiable proof of consumer injury.

Outlawing egregious review and testimonial practices by rule expands the Commission's enforcement toolkit and allows it to deliver on its mission by stopping and deterring harmful conduct and, in some cases, making American consumers whole when they have been wronged. Because fake reviews and the other unfair or deceptive review and testimonial practices described here are so prevalent and so harmful, the unlocking of additional remedies through this rulemaking, particularly the ability to seek civil penalties against violators and obtain redress for consumers or others injured by the conduct, will allow the Commission to more effectively police harmful review and testimonial practices that plague consumers and honest businesses.

V. Overview and Scope of Proposed Rule on the Use of Consumer Reviews and Testimonials

A. Key Definitions

Proposed § 465.1 would provide definitions for 12 terms as they appear in proposed 16 CFR part 465, including, among others, definitions for the terms “consumer reviews,” “consumer testimonials,” and “celebrity testimonials.”

The term “consumer review” is defined in proposed § 465.1(d) as a consumer's evaluation, or a purported consumer's evaluation, of a product, service, or business that is submitted by the consumer, or purported consumer, and that is published to a website or platform dedicated in whole or in part to receiving and displaying such reviews. The definition states that consumer reviews include consumer ratings, regardless of whether they include any text or narrative.

The definition includes “purported consumers” so that it covers reviews by authors who do not exist. It does not include all consumer evaluations of products or services, such as a blog post or other social media post evaluating a product; it is limited to those submitted to a website or platform or portion thereof dedicated to such reviews. Such websites and platforms would include, among other things, third-party review platforms and advertiser and retailer websites that collect and display consumer reviews. A consumer review submitted and published to one website that is republished on a second website is still a consumer review as republished. A consumer review is not necessarily advertising.

The term “consumer testimonial” is defined in proposed § 465.1(e) as an advertising or promotional message that consumers are likely to believe reflects the opinions, beliefs, or experiences of a consumer who has purchased, used, or otherwise had experience with a product, service, or business.

206

Proposed § 465.1(b) provides a corresponding definition of the term “celebrity testimonial.” It defines the term “celebrity testimonial” as an advertising or promotional message that consumers are likely to believe reflects the opinions, beliefs, or experiences of a well-known person who purchased, used, or otherwise had experience with a product, service, or business.

206

The definition of a consumer testimonial is based upon the definition of an “endorsement” in the Commission's Guides Concerning the Use of Endorsements and Testimonials in Advertising.

See

Guides Concerning the Use of Endorsements and Testimonials in Advertising, 16 CFR 255.0(b).

B. Fake or False Consumer Reviews, Consumer Testimonials, or Celebrity Testimonials

Proposed § 465.2 would prohibit certain types of deceptive conduct involving “consumer reviews,” “consumer testimonials,” and “celebrity testimonials.”

Proposed § 465.2(a) would prohibit a business from writing, creating, or selling a consumer review, consumer testimonial, or celebrity testimonial that: (a) is by someone who does not exist; (b) is by someone who did not use or otherwise have experience with the product, service, or business that is the subject of the review or testimonial; or (c) materially misrepresents the reviewer's or testimonialist's experience with the product, service, or business.

Proposed § 465.2(b) would render it a deceptive act or practice for a business to purchase consumer reviews, or disseminate or cause the dissemination of consumer or celebrity testimonials, about the business or one of its products or services, if the business knew or should have known that the review or testimonial: (a) was by someone who does not exist, (b) is by someone who did not use or otherwise have experience with the product, service, or business, or (c) materially misrepresents the reviewer's or testimonialist's experience with the product, service, or business. In accordance with proposed § 465.1(h), “purchase a consumer review” means to provide something of value, such as money, goods, or another review, in exchange for a consumer review.

Proposed § 465.2(c) would make it a deceptive act or practice for a business to procure consumer reviews about the business or one of its products or services for posting on a third-party platform or website, if the business knew or should have known that the review: (a) was by someone who does not exist, (b) is by someone who did not use or otherwise have experience with the product, service, or business, or (c) materially misrepresents the reviewer's experience with the product, service, or business.

Google's comment said a proposed rulemaking should not apply to review platforms.

207

Proposed § 465.2 accounts for this concern. The provision does not apply to businesses, like third-party review platforms, that disseminate consumer reviews that are not of their products, services, or businesses. Neither does it apply to any reviews that a platform simply publishes and that it did not purchase.

207

Google Cmt. at 9.

NRF opposed requiring the manual review of every consumer review and poster's profile.

208

Proposed § 465.2 accounts for this concern by not imposing any obligation on those publishing consumer reviews to manually review consumer reviews or poster profiles and by not applying to reviews that a platform simply publishes.

208

NRF Cmt. at 6.

Trustpilot asserted any rule should consider a consumer review to be legitimate if the consumer had experience with the business, even if no purchase was made, and NRF's comment opposed requiring retailers to restrict consumer reviews to verified purchasers.

209

In light of these concerns, proposed § 465.2 does not limit legitimate reviews to reviews by purchasers or verified purchasers. It requires only that the reviewer had experience with the product, service, or business.

209

Trustpilot Cmt. at 3-4; NRF Cmt. at 6.

NRF also recommended any rule provision addressing fake reviews be limited to review brokers and not apply to the parties purchasing the reviews.

210

It said that buyers of fake reviews should not be covered by a rule because the Commission can already bring a

“formal enforcement action seeking monetary damages,” and it was not opposed to Commission action against such purchasers.

211

The Commission believes a rule should indeed apply to those who knowingly purchase fake reviews given that they are no less culpable for deceiving consumers than the brokers. The Commission's ability to seek monetary relief without a rule applies to both brokers and buyers, and it does not obviate the need for a rule because, as discussed above, seeking such relief is much more difficult without a rule.

210

NRF Cmt. at 1, 7.

211

Id.

at 6.

C. Consumer Review Repurposing

Proposed § 465.3 would prohibit a business from using or repurposing, or causing the use or repurposing of, a consumer review written or created for one product so it appears to have been written or created for a substantially different product. This could consist of combining substantially different products so that they share consumer reviews or changing a product page so it features a different product but retains the reviews of the prior product, or copying reviews of other products from other sites. The term “substantially different product” is defined in proposed § 465.1(j), which establishes that the term refers to a product that differs from another product in one or more material attributes other than color, size, count, or flavor. Although differences in flavor are likely to be material to some consumers in some instances, the question can be highly fact specific. For this reason, combining reviews for a product that has multiple flavors would not be a rule violation, though it could still be a deceptive practice under the FTC Act.

D. Buying Positive or Negative Consumer Reviews

Proposed § 465.4 would prohibit a business from offering compensation or other incentives in exchange for, or conditioned on, the writing or creation of consumer reviews expressing a particular sentiment, whether positive or negative, regarding the product, service, or business that is the subject of the review.

ANA's comment asserted any proposed rulemaking should not address “review gating” or the “

mere

solicitation of positive reviews.”

212

Review gating occurs when a business asks past purchasers to provide feedback on a product and then invites only those who provide positive feedback to post online reviews on one or more websites. Review gating and the mere solicitation of positive reviews are not covered by the proposed Rule. Although the Commission believes review gating can be deceptive,

213

whether any given instance of review-gating is deceptive can be highly fact specific.

212

ANA Cmt. at 7.

213

See

Endorsement Guides, 16 CFR 255.2(e)(11) (review gating “may be an unfair or deceptive practice if it results in the posted reviews being substantially more positive than if the marketer had not engaged in the practice”) and 16 CFR 255.2(d).

Family First Life commented that the FTC should not promulgate a rule “that sweeps in and penalizes

any

review just because the reviewer was offered an incentive to write it . . . without otherwise dictating what the review says.”

214

The proposed Rule does not address incentivized reviews except for those required to express a particular sentiment, but the Commission notes that other uses of incentivized reviews can be deceptive and violate the FTC Act.

215

The deceptiveness of undisclosed incentivized reviews is highly fact specific.

214

Family First Life Cmt. at 12-13.

215

See, e.g.,

Endorsement Guides, 16 CFR 255.5(b)(6)(ii) (any resulting review that fails to clearly and conspicuously disclose the incentives provided to that reviewer is likely deceptive).

E. Insider Consumer Reviews and Consumer Testimonials

Proposed § 465.5 addresses company insider consumer reviews and consumer testimonials in three different ways. Proposed § 465.5(a) applies to insider reviews and testimonials; proposed § 465.5(b) applies to insider testimonials; and proposed § 465.5(c) applies to insider reviews.

Proposed § 465.5(a) would prohibit an officer or manager of a business from writing or creating a consumer review or consumer testimonial about the business or its products or services if the consumer review or consumer testimonial does not have a clear and conspicuous disclosure of the officer's or manager's relationship to the business. Proposed § 465.1(c) defines “clear and conspicuous” to mean that a required disclosure is easily noticeable (

i.e.,

difficult to miss) and easily understandable by ordinary consumers, including in all of the ways listed in the definition. This is the same definition the Commission proposed in its Negative Option Rule Notice of Proposed Rulemaking.

216

In accordance with proposed § 465.1(g), “officers” are defined to include a business's owners, executives, and managing members.

216

See

Fed. Trade Comm'n, NPRM: Negative Option Rule, 88 FR 24716, 24734 (Apr. 24, 2023),

https://www.govinfo.gov/content/pkg/FR-2023-04-24/pdf/2023-07035.pdf.

Proposed § 465.5(b) applies to consumer testimonials in advertisements disseminated by or on behalf of a business. It would prohibit a business from disseminating or causing the dissemination of certain consumer testimonials about the business or its products or services if the consumer testimonial is written by the business's officers, managers, employees, or agents, or any of their relatives without clear and conspicuous disclosures of those relationships. This provision would apply only when the business knew or should have known of the testimonialist's relationship.

Proposed § 465.5(c) applies to solicitations of employee and other insider reviews. It would prohibit under some circumstances an officer's or manager's solicitation of consumer reviews from employees, agents, or relatives that results in reviews that don't clearly and conspicuously disclose the reviewer's relationship. The provision is limited to situations when the person soliciting the review knew or should have known of the prospective reviewer's relationship and: (a) failed to instruct the prospective reviewer to disclose clearly and conspicuously that relationship, (b) knew or should have known the review appeared without such a disclosure and failed to take remedial steps or, (c) encouraged the prospective reviewer not to make such a disclosure.

NADA recommended any proposed rule provision addressing businesses writing, soliciting, or publishing reviews by their employees or family members clarify that a violation “only arises when the business, and not another entity, affirmatively writes, solicits, and publishes reviews that fail to provide clear and conspicuous disclosures of those relationships.”

217

Proposed § 465.5(c) would apply to reviews by employees or family members. Proposed § 465.5(c) is limited to solicitation by an officer or manager, and only when the solicitor failed to advise a disclosure, knew or should have known that a review appeared without such a disclosure and failed to take remedial steps, or encouraged the prospective reviewer not to make such a disclosure. The business would not be liable under the proposed provision for an unsolicited review, for a review about which the solicitor reasonably should not have known, or for a reviewer who refuses to make a disclosure. However, proposed § 465.5(c) reflects the Commission's

belief that businesses should be prohibited not only from publishing insider reviews themselves but also from causing their creation (

e.g.,

when an officer or manager of the business solicits employees to post reviews on third-party review websites and fails to instruct the employees to disclose their relationship to the business).

217

NADA Cmt. at 3.

NADA asked the FTC to define the term “relative.”

218

The Commission believes that the limitation to situations in which officers or managers know or should know that they are soliciting a relative for an endorsement or testimonial addresses the comment without the need for a definition.

218

Id.

Family First Life commented that when an independent contractor agent writes a review on a workplace-review platform such as Glassdoor, the reviewer's relationship to the company is obvious and assumed.

219

The Commission agrees, in reviews on such platforms, the relationship is readily apparent and, in effect, already disclosed. The Commission does not believe the proposed Rule needs to specifically address this scenario.

219

Family First Life Cmt. at 9-10.

F. Company-Controlled Review Websites or Entities

Proposed § 465.6 prohibits a business from representing that a website, organization, or entity is providing its independent reviews or opinions about a category of businesses, products, or services that includes the business or its products or services, when the business controls, owns, or operates that website, organization, or entity.

G. Review Suppression

Proposed § 465.7 addresses two types of review suppression. The first type, addressed in proposed § 465.7(a), would prohibit anyone from using an unjustified legal threat or a physical threat, intimidation, or false accusation to prevent the creation of a consumer review or cause the removal of all or part of a review. In accordance with proposed § 465.1(l), an “unjustified legal threat” is defined as a threat to initiate or file a baseless legal action, such as an action for defamation that challenges truthful speech or matters of opinion.

NADA recommended that any proposed rule not prohibit what it characterized as good faith online reputation management practices, such as a business: (a) reaching out to consumers who have posted negative reviews and attempting to improve their reviews by addressing their concerns (including sometimes giving customers something of value in satisfaction of their complaints), or (b) responding on a comment thread to each negative review, offering an explanation, making customers whole, and asking any successfully satisfied customers to update their previously negative review.

220

Neither proposed § 465.7(a) nor any other proposed Rule provision would prohibit such conduct (assuming that reviewers are not required to remove or change their reviews in order to be made whole).

220

NADA Cmt. at 3-4.

Proposed § 465.7(b) would prohibit a business from misrepresenting that the consumer reviews of one or more of its products or services displayed on its website or platform represent most or all the reviews submitted to the website or platform if reviews are being suppressed based upon their ratings or their negativity. As proposed, the provision makes clear that the non-publication of consumer reviews for certain enumerated reasons is not considered to be review suppression so long as the criteria for withholding reviews are applied to all reviews submitted without regard to the favorability of the review. The listed acceptable reasons for not publishing a review are: (a) that the review contains: (i) trade secrets or privileged or confidential commercial or financial information, (ii) libelous, harassing, abusive, obscene, vulgar, or sexually explicit content, (iii) the personal information or likeness of another person, (iv) content that is discriminatory with respect to race, gender, sexuality, ethnicity, or another protected class, or (v) content that is clearly false or misleading; (b) the seller reasonably believes it is fake; or (c) the review is wholly unrelated to the products or services offered by or available at the website or platform. These criteria are based upon those enumerated in the Consumer Review Fairness Act, 15 U.S.C. 45b(b)(2) and (3). Moreover, consumers would reasonably expect and often prefer that a business exclude reviews meeting these criteria, so the undisclosed exclusion of such reviews solely due to application of those criteria would be unlikely to mislead or be material to consumers.

NADA stated that businesses should be able to “remove reviews or comments that are off topic or include false statements, advertisements, inappropriate language, or confidential or personal identification information.”

221

As to reviews that are “off topic,” proposed § 465.7(b) would permit not publishing reviews that are “wholly unrelated to the products or services offered.” As to reviews that contain “false statements,” proposed § 465.7(b) would permit not publishing reviews that are “clearly false or misleading.” It is unclear what the comment meant by reviews that include “advertisements.” If NADA means that it is acceptable to delete a review that mentions a competitor, that is not an exception provided in proposed § 465.7(b).

222

With regard to reviews that contain “inappropriate language,” proposed § 465.7(b) would permit not publishing reviews containing “harassing, abusive, obscene, vulgar, or sexually explicit content” or “content that is discriminatory with respect to race, gender, sexuality, ethnicity, or another protected class.” As to reviews that contain “confidential or personal identification information,” proposed § 465.7(b) would allow a seller to not publish a review that contains “trade secrets or privileged or confidential commercial or financial information,” or the “personal information . . . of another person.” NADA also said that businesses should be able to “remove comments or review functions on their own websites or certain social media posts.”

223

The proposed Rule does not prohibit or address such conduct.

221

Id.

at 3.

222

Cf.

FTC Statement of Policy Regarding Comparative Advertising (1979),

https://www.ftc.gov/legal-library/browse/statement-policy-regarding-comparative-advertising

(“Commission policy in the area of comparative advertising encourages the naming of, or reference to competitors, but requires clarity, and, if necessary, disclosure to avoid deception of the consumer.”).

223

NADA Cmt. at 3. The NADA also posited that highlighting five-star reviews from satisfied customers on a dealer's websites is a legitimate practice that should not be prohibited under a possible rule.

Id.

at 4. The proposed Rule does not address such a practice. The Commission notes, however, that highlighting five-star reviews from satisfied customers on a dealer's websites or in its other advertising could be a deceptive practice depending on the facts.

H. Misuse of Fake Indicators of Social Media Influence

Proposed § 465.8 prohibits the misuse of indicators of social media influence. As defined by proposed § 465.1(f), the term “indicators of social media influence” refers to any metrics used by the public to make assessments of an individual's or entity's social media influence, such as followers, friends, connections, subscribers, views, plays, likes, reposts, and comments.

Proposed § 465.8(a) prohibits anyone from selling fake indicators of social media influence that can be used by persons or businesses to misrepresent their influence for a commercial purpose. Proposed § 465.8(b) prohibits

anyone from procuring fake indicators of social media influence to misrepresent their influence or importance for a commercial purpose.

I. Severability

Proposed § 465.9 is a severability provision. It provides that the provisions of the proposed Rule are separate and severable from one another. If any provision is stayed or determined to be invalid, the remaining provisions will continue in effect.

VI. The Rulemaking Process

The Commission can decide to finalize the proposed Rule if the rulemaking record, including the public comments in response to this NPRM, supports such a conclusion. The Commission may, either on its own initiative or in response to a commenter's express request, engage in additional processes, including those described in 16 CFR 1.12 and 1.13. If the Commission on its own initiative decides to conduct an informal hearing, or if a commenter files an express request for such a hearing, then a separate notice will issue under 16 CFR 1.12(a). Any person who would like to participate by providing an oral statement at any informal hearing must make an express request to do so in response to this NPRM. Based on the comment record and existing prohibitions against unfair or deceptive consumer reviews and testimonials under Section 5 of the FTC Act, the Commission does not here identify any disputed issues of material fact necessary to be resolved at an informal hearing. The Commission may still do so later, on its own initiative or in response to a commenter.

VII. Preliminary Regulatory Analysis

Under Section 22 of the FTC Act, the Commission, when it publishes any NPRM for a rule as defined in Section 22(a)(1), must include a “preliminary regulatory analysis.” 15 U.S.C. 57b-3(b)(1). The required contents of a preliminary regulatory analysis are (a) “a concise statement of the need for, and the objectives of, the proposed rule,” (b) “a description of any reasonable alternatives to the proposed rule which may accomplish the stated objective,” and (c) “a preliminary analysis of the projected benefits and any adverse economic effects and any other effects” for the proposed rule and each alternative, along with an analysis “of the effectiveness of the proposed rule and each alternative in meeting the stated objectives of the proposed rule.” 15 U.S.C. 57b-3(b)(1)(A)-(C). This NPRM already provided the concise statement of the need for, and the objectives of, the proposed Rule in Section IV above. It addresses the other requirements below.

A. Anticipated Costs and Benefits and Reasonable Alternatives

The Commission is proposing a rule to curb certain unfair or deceptive uses of reviews and testimonials. The proposed Rule contains several provisions to promote accuracy in consumer reviews (henceforth “reviews”) and, thus, will help the vast majority of American consumers who rely on such reviews to make better-informed purchase decisions. The proposed Rule prohibits: the creation, purchasing, procurement, or dissemination of fake or false reviews; repurposing of reviews for substantially different products; and buying of reviews in exchange for, or conditioned on, positive or negative sentiments. It also includes prohibitions on fake or false consumer or celebrity testimonials, insider reviews, misleading company-controlled review websites or entities, certain review suppression practices, and the misuse of indicators of fake social media influence.

The Commission believes that the benefits of proceeding with the rulemaking will significantly outweigh the costs, but it welcomes public comment and data (both qualitative and quantitative) on any benefits and costs to inform a final regulatory analysis.

In the preliminary analysis below, the NPRM describes the anticipated impacts of the proposed Rule. Where possible, it quantifies the benefits and costs. If a benefit or cost is quantified, it indicates the sources of the data relied upon. If an assumption is needed, the text makes clear which quantities are being assumed. The NPRM measures the benefits and costs of the proposed Rule against a baseline in which no rule regarding consumer reviews has been promulgated by the Commission. The Commission solicits comments from the public to improve these estimates before the promulgation of any final rule.

The estimates in this preliminary analysis attempt to include a broad set of economic actors, using data on the number of entities registered as businesses in the United States, data on retail sales, and data on U.S. consumers who shop online. The Commission invites submission of information pertaining to additional economic actors who would be affected by the proposed Rule. Conversely, the Commission solicits information on whether a more limited set of economic actors would yield improved estimates.

Quantifiable benefits stem from consumer welfare improvements and consumer time savings. With the proposed Rule, online reviews will be more accurate overall, leading consumers to purchase higher-quality products or products that are better-matched to their preferences. The proposed Rule will also lead to more trustworthy aggregate review ratings (

e.g.,

star ratings), leading some consumers to spend less time scrutinizing reviews to determine their validity. Quantifiable costs primarily reflect the resources spent by businesses to review the proposed Rule and to take any preemptive or remedial steps to comply with its provisions. Because the proposed Rule is an application of preexisting law under Section 5 of the FTC Act, the Commission expects these compliance costs to be minimal.

A period of 10 years is used in the baseline scenario because FTC rules are subject to review every 10 years. Quantifiable aggregate benefits and costs are summarized as the net present value over this 10-year period in Table 1.1. The discount rate reflects society's preference for receiving benefits earlier rather than later; a higher discount rate is associated with a greater preference for benefits in the present. The present value is obtained by multiplying each year's net benefit by the discount rate raised to the power of the number of years in the future the net benefit accrues.

Table 1.1—Present Value of Net Benefits, 2023-2033

[In billions]

Present value:

low-end

estimate

Present value:

high-end

estimate

Total Benefits:

3% Discount Rate

$59.31

$234.28

7% Discount Rate

50.16

200.26

Total One-Time Costs

0.83

0.00

Net Benefits

3% Discount Rate

58.48

234.28

7% Discount Rate

49.33

200.26

1. Estimated Benefits of the Proposed Rule

This section describes the beneficial impacts of the proposed Rule, provides preliminary quantitative estimates where possible, and describes benefits that are only assessed qualitatively. The quantifiable estimates reflect benefits stemming from the decrease in online review manipulation on third-party platforms or company websites, which covers most of the prohibitions contained in the proposed Rule. This analysis does not calculate benefits from the other aspects of the proposed Rule—prohibitions on fake or false celebrity testimonials; prohibitions on company-controlled entities that purportedly provide independent opinions; prohibitions on unjustified legal threats or physical threats, intimidation, or false accusations in an attempt to suppress negative consumer reviews; and prohibitions on the misuse of indicators of fake social media influence—because of the limited quantitative research in these areas. The Commission invites comment on research concerning these other aspects of the proposed Rule. The quantified benefits are presented by benefit category, rather than stemming from a specific provision in the proposed Rule, because the relevant provisions have the same end goal—that is, to improve the information available to consumers by reducing the level of review manipulation. Therefore, it is difficult to disentangle the benefits stemming from each provision.

Existing academic literature in economics, marketing, computer science, and other fields documents the importance of online reviews: the number of online reviews and aggregate ratings are extremely important for consumer purchase decisions. It is widely documented that the presence of online reviews improves consumer welfare via reductions in both search costs and the level of information asymmetry that exists prior to purchase.

224

224

See, e.g.,

Dina Mayzlin, “Promotional Chat on the Internet,” 25(2)

Mktg. Sci.,

155-63 (2006).

When making purchase decisions, consumers typically have incomplete information on product quality and attributes. Searching for additional information is costly. Consumers incur costs—including time and effort costs—to seek, evaluate, and integrate incoming information. Online platforms where past users share information about their experiences can significantly lower search costs.

Researchers have also demonstrated consumer reviews create value for consumers beyond a reduction in search costs. Consumers are better able to learn of a product's quality and attributes when there is free-flowing, non-manipulated commentary from past consumers. Consumer reviews lead to “better” decisions by increasing the level of information available prior to purchase and reducing uncertainty. By the same token, the academic literature also documents that manipulated or fake reviews lead to reductions in consumer welfare by leading consumers to buy low-quality products or otherwise make suboptimal purchase decisions.

A secondary benefit is deterrence of the specified review and testimonial practices. The proposed Rule is essentially the only means for imposing civil penalties in most cases involving such practices. Civil penalties are not generally available under the FTC Act for this conduct, unless parties are already subject to a relevant Commission order or have been served with a copy of a relevant Notice of Penalty Offenses. Also, as noted above, in many cases involving this conduct, calculating redress or other Section 19 relief may be difficult. Without civil penalties, bad actors have little fear of being penalized for using fraud and deception in connection with reviews and endorsements.

To obtain redress without alleging a rule violation, the Commission must successfully conclude an administrative proceeding including any appeal and file a follow-on Federal case under Section 19 to establish that the conduct is “one which a reasonable man would have known under the circumstances was dishonest or fraudulent.”

225

Although the Commission is likely to meet this standard in cases involving the conduct covered by the proposed Rule, it would take substantially more time and resources, and would significantly delay any redress to victims, compared to a case under the proposed Rule violation, which does not require multiple proceedings or a special knowledge requirement.

226

225

15 U.S.C. 57b(a)(2). Depending on the egregiousness of the misconduct and the harm it is causing, the Commission also may seek preliminary injunctive relief in Federal court. 15 U.S.C. 53(b).

226

See, e.g.,

Press Release, Fed. Trade Comm'n, Marketers of Ab Force Weight Loss Device Agree to Pay $7 Million for Consumer Redress (Jan. 14, 2009),

https://www.ftc.gov/news-events/news/press-releases/2009/01/marketers-ab-force-weight-loss-device-agree-pay-7-million-consumer-redress

(describing a 2009 settlement of a follow-on Section 19 action against Telebrands Corp. that was brought after litigation of a 2003 administrative complaint alleging violations of Section 5 concluded—in this case, the Section 19 action settled instead of being litigated to judgment, which would have taken more time).

Given the prevalence of unfair or deceptive conduct involving reviews and testimonials, the Commission will have no shortage of bad actors to investigate; it could invest the extra resources freed up by any final rule into more investigations and actions with respect to consumer reviews or testimonials. In sum, the potential consumer-redress benefits of the proposed Rule are significant: the Commission could put a stop to more inarguably unfair or deceptive consumer reviews or testimonials, return money to more victims, and obtain that redress more quickly.

a. Consumer Welfare Benefits From Better-Informed Purchase Decisions

The study containing the most direct estimate of welfare losses from review manipulation finds that the presence of fake reviews leads consumers to lose

$0.12 for every dollar spent in an experimental setting.

227

Due to limited quantitative estimates in the literature, the NPRM assumes this measure of welfare loss encompasses the various types of review manipulation covered by the proposed Rule. It also assumes the proposed Rule causes all fake or manipulated reviews to vanish. Thus, consumers will gain an estimated $0.12 for every dollar spent on goods whose online reviews included fake ones.

227

See

Akesson et al., “The Impact of Fake Reviews on Demand and Welfare,”

supra

note 158.

To estimate consumer welfare benefits from better-informed purchase decisions, the NPRM first estimates the total amount of sales for which consumers consult online reviews. U.S. e-commerce sales totaled $1.034 trillion in 2022.

228

The NPRM assumes all products sold online had some form of user-generated commentary (

e.g.,

on third-party review platforms, on discussion boards, on company websites, or on social media), and that this commentary factors into consumers' purchase decisions for these goods.

228

See

U.S. Census Bureau, “Quarterly E-Commerce Sales Report,” Feb. 17, 2023,

https://www.census.gov/retail/ecommerce.html.

Online reviews are also important for commerce not conducted online, including for revenues earned by the hospitality industry and by other services. Sales for businesses classified as “Food Services and Drinking Places” by the U.S. Census totaled $843.61 billion in 2021, which includes revenue from restaurants and bars.

229

The NPRM assumes consumers rely on reviews for only a portion of these sales. Some consumers—particularly those living in rural parts of the country and in smaller cities—may have a small set of familiar food and drink establishments available to them, making online reviews less influential to their decision to patronize a particular one. Moreover, prior research has found online reviews do not impact revenues of chain restaurants.

230

Accordingly, the NPRM assumes consumers rely on reviews for 25 percent of the total revenue generated in the food services and drinking places sector (25 percent of $843.61 billion, or $210.90 billion).

231

229

U.S. Census Bureau, “Service Annual Survey,” Nov. 22, 2022,

https://www.census.gov/programs-surveys/sas.html

(listing total revenue of $843,605,000,000 for NAICS Code 722 in 2021, the most recent year with data).

230

See

Michael Luca, “Reviews, Reputation, and Revenue: The Case of

Yelp.com.

” Harvard Bus. Sch. Working Paper 12-016 (2016).

231

25 percent is likely to be a reasonable estimate based on the difference in revenues for new restaurants and established restaurants. A study conducted by Toast, Inc. found that new restaurants earn approximately $112,000 in average revenue per year (

https://pos.toasttab.com/blog/on-the-line/average-restaurant-revenue

). This is approximately 25 percent of average revenue for restaurants overall ($486,000, according to the website Eat Pallet,

https://eatpallet.com/how-much-do-restaurants-make-in-a-day).

Online reviews are also important for sales in other service sectors. In 2021, total revenue was $247.25 billion for the accommodations sector (which includes hotels and vacation rentals), and total revenue was $56.85 billion for personal services (including beauty salons, barber shops, health clubs, and non-veterinary pet care), totaling $304.10 billion for both sectors.

232

About half of hotel revenue is generated by business travelers, who are likely to rely less on online reviews than leisure travelers do.

233

In addition, pre-paid hotel bookings and vacation rentals booked online are already accounted for in the e-commerce sales figure described above. Furthermore, some consumers may be loyal customers of local salons and other personal services, regardless of these businesses' online reputations. Because of these reasons, the NPRM assumes that a subset of accommodation and personal services revenues is affected by consumer reviews. Similar to the calculation for the food and drinking places industry, the NPRM assumes 25 percent of total accommodation and personal care services revenue is impacted by consumer reviews (25 percent of $304.10 billion, or $76.03 billion). The total estimated revenue for services impacted by consumer reviews is $286.93 billion (the sum of $210.90 billion and $76.03 billion). The Commission seeks comments to improve this estimate.

232

See

U.S. Census Bureau, “Service Annual Survey,”

supra

note 229 (listing total 2021 revenue of $247,246,000,000 for NAICS Code 721 and listing total 2021 revenue of $56,845,000,000 for NAICS Codes 812111 through 812199 and NAICS Code 81291).

233

See

Linchi Kwok, “Will business travel spending return to the pre-pandemic level soon?” Hospitality Net, Sept. 22, 2022,

https://www.hospitalitynet.org/opinion/4112075.html.

Combining the revenue estimates described above yields $1.321 trillion in estimated sales of goods or services for which consumers incorporate reviews into their decision-making. In this analysis, the NPRM does not incorporate revenues stemming from the physical sale of goods in retail stores where consumers read online reviews before purchasing items in person. The Commission invites commenters to submit information to quantify non-e-commerce retail sales that are impacted by reviews.

Quantitative estimates of the incidence of fake or false reviews vary by source.

234

Nevertheless, at least three prior studies examining the degree of review manipulation as a proportion of businesses or products (rather than as a proportion of

reviews

) contain similar findings: according to these studies, approximately 10 percent of products or businesses have some manipulated consumer reviews.

235

Thus, a basic approximation of total e-commerce sales involving some review manipulation is 10 percent of $1.034 trillion, or $103.4 billion. Similarly, a basic approximation of review-dependent service industry sales involving some review manipulation is 10 percent of $286.93 billion, or $28.69 billion. The Commission seeks submissions of additional research on the prevalence of review manipulation to improve this estimate.

234

These estimates range from the single digits to over 20 percent.

See

Tripadvisor, “2023 Review Transparency Report,”

https://www.tripadvisor.com/TransparencyReport2023

(finding that 4.4% of review submissions were fraudulent); Trustpilot, “Transparency Report 2022,”

https://cdn.trustpilot.net/trustsite-consumersite/trustpilot-transparency-report-2022.pdf

(stating that its software removed 5.8 percent of reviews due to being fake); Yelp, “2022 Yelp Trust & Safety Report,”

https://trust.yelp.com/trust-and-safety-report

(stating that 19 percent of submitted reviews were marked as “not recommended” by Yelp's software); Devesh Raval, “Do Gatekeepers Develop Worse Products? Evidence from Online Review Platforms,”

Working Paper,

Feb. 27, 2023,

https://deveshraval.github.io/reviews.pdf

(finding that the share of hidden (likely fake) Yelp reviews is as high as 47 percent).

235

See

Nan Hu et al., “Manipulation of online reviews: An analysis of ratings, readability, and sentiments,” 52(3)

Decision Support Systems

674-84 (Feb. 2012) (finding that 10.3 percent of books sold on Amazon had manipulated reviews); Luca and Zervas, “Fake It Till You Make It: Reputation, Competition, and Yelp Review Fraud,”

supra

note 158 (finding that 10 percent of Boston, MA restaurants had filtered 5-star reviews on Yelp) (Table 3, row 4); Raval, “Do Gatekeepers Develop Worse Products? Evidence from Online Review Platforms,”

supra

note 234 (finding that 9.7 percent of businesses with reviews or complaints with the Better Business Bureau are of low quality, where fake reviews inflate ratings) (Table III, column 3, row 1).

Importantly, online businesses that engage in review manipulation are likely to earn less revenue than other e-commerce companies. For example, prior research has found independent firms and sellers offering lower-quality products are more likely to engage in review manipulation.

236

Therefore, e-commerce sales affected by review manipulation are likely to be lower than the $103.4 billion in sales described above. A more conservative estimate of e-commerce sales involving review manipulation can be obtained by using

price differentials of review-manipulated products versus others. Because products with online review manipulation have price points approximately 19 percent of the average price of goods sold online (according to research using data from Amazon),

237

a more conservative estimate of review-manipulated products' revenue is 1.9 percent (19 percent × 10 percent) of all $1.034 trillion in e-commerce sales, or $19.65 billion. Because the Commission does not have data on the revenue or quantities sold of review-manipulated products, the NPRM assumes revenue is constant across price points and rely solely on the price differential to approximate revenue. The NPRM does not similarly adjust revenues for non-e-commerce firms (

e.g.,

restaurant and hotels) because there is less variation in prices in those industries. The Commission invites commenters to submit information to improve this estimate.

236

See

He et al., “The Market for Fake Reviews,”

supra

note 158; Dina Mayzlin et al., “Promotional Reviews: An Empirical Investigation of Online Review Manipulation,” 104(8) The Am. Econ. Rev. 2421-55 (2014).

237

See

Davide Proserpio et al., “How Fake Customer Reviews Do—and Don't—Work,” Harvard Bus. Rev., Nov. 24, 2020,

https://hbr.org/2020/11/how-fake-customer-reviews-do-and-dont-work.

The authors find that products sold on Amazon with manipulated reviews are typically in the $15 to $40 price range. The midpoint of this range ($27.50) represents 19 percent of the average product's price ($142.74, according to a study conducted by Semrush Inc.,

https://www.semrush.com/blog/amazon-pricing-study

).

The NPRM estimates annual welfare gains by applying the $0.12 estimate, described above, to the estimated amount of U.S. sales likely to have some manipulated consumer reviews, yielding an annual estimate of welfare gains in the range of $5.80 billion (12 percent of $48.34 billion, the sum of $19.65 billion and $28.69 billion) and $15.85 billion (12 percent of $132.09 billion, the sum of $103.4 billion and $28.69 billion). Assuming e-commerce sales increase linearly over the next ten years at the same rate as they did in the past year,

238

the present value of consumer welfare improvements from better-informed purchasing decisions is estimated to be between $50.16 and $199.40 billion as described in Table 2.1.

238

E-commerce sales increased by 7.7 percent from 2021 to 2022.

See

U.S. Census Bureau, “Quarterly E-Commerce Sales Report,”

supra

note 228. Using growth in the past year to predict future e-commerce sales results in a more conservative estimate than using a longer time frame. E-commerce sales experienced higher annual growth rates prior to 2021 (14 percent from 2018 to 2019, 43 percent from 2019 to 2020, and 14 percent from 2020 to 2021). The NPRM does not project revenues for non-e-commerce industries because the two most recent years of data are from 2021 and 2020; linear trends during these years are unique to the pandemic and are unlikely to be accurate for future years.

Table 2.1—Estimated Benefits From Consumer Welfare Improvements From Purchase Decisions

Percent of e-commerce revenue impacted by review manipulation

Total annual

welfare

improvements

from better-

informed

purchase

decisions

(in billions)

Total 10-year

(2023-2033)

welfare

improvement,

3% discount rate

(in billions)

Total 10-year

(2023-2033)

welfare

improvement,

7% discount rate

(in billions)

10

$15.85

$199.40

$170.43

1.9

5.80

59.31

50.16

b. Consumer Time Savings From Increased Reliability of Summary Ratings

The proposed Rule's prohibitions against deceptive and unfair consumer review practices would increase the reliability of consumer reviews. The NPRM assumes this improvement in the dependability of reviews will lead consumers to place more trust in aggregate measures (

e.g.,

aggregate star ratings), which many review settings use to summarize consumer reviews. This in turn will lead some consumers to spend less time scrutinizing individual reviews to detect red flags commonly found in manipulated reviews (

e.g.,

spelling and grammar mistakes, generic highly positive or negative statements, and lack of detail). Therefore, the proposed Rule is likely to result in some amount of time savings for consumers who consult online reviews before making purchases.

Approximately 80 percent of Americans are online shoppers.

239

Of those who shop online, 14 percent shop online more than once a week, 20 percent shop online once a week, 23 percent shop online once every two weeks, 25 percent shop online once a month, and the remainder do so every few months.

240

Different age groups of online shoppers spend various amounts of time reading reviews before making a purchase decision. On average, younger consumers spend more time reading reviews than older consumers.

241

This analysis does not incorporate time spent by consumers researching reviews of restaurants, hotels, and other goods and services that are not purchased online because of the limited amount of information available regarding consumers' total time spent on such activities. The Commission invites commenters to submit information related to the time consumers spend reading reviews for goods and services not purchased online.

239

See

Pew Research Center, “Online Shopping and E-Commerce,” Dec. 19, 2016,

https://www.pewresearch.org/internet/2016/12/19/online-shopping-and-e-commerce.

240

See

Int'l Post Corp., “Cross-Border E-Commerce Shopper Survey 2022,” Jan. 2023,

https://www.ipc.be/-/media/documents/public/publications/ipc-shoppers-survey/onlineshoppersurvey2022.pdf.

241

See

BrightLocal, “Local Consumer Review Survey 2019,” Dec. 11, 2019,

https://www.brightlocal.com/research/local-con

This text is long and has been trimmed here. Open the source document for the complete record.