‘Avvo’ Raises Important Questions About Opinion-Fact Dichotomy in False Advertising Cases
ALM Media
Milton Springut
Avvo is a popular website that profiles and rates attorneys and is used by many attorneys to promote their practices. Avvo was recently acquired by Internet Brands, which owns several other legal websites, including Nolo, Lawyers.com, and Martindale-Hubble. But Avvo’s business model has been challenged repeatedly by attorneys unhappy with its profile and rating system.
In December, the Southern District of New York dismissed a false advertising claim, on the pleadings, brought by New York attorney Kevin Davis under the Lanham Act and New York General Business Law §349, Davis v. Avvo, 2018 U.S. Dist. LEXIS 213789 (S.D.N.Y. 2018), holding that Avvo’s ratings are non-actionable “opinion.” (The court also held that Avvo’s awarding certain attorneys the status of “Pro” was mere non-actionable puffery. This article does not discuss that part of the opinion.) But a review of Davis’ complaint indicates that the matter may not be so simple.
Avvo’s Site and Davis’ Claims
Avvo’s online platform lists many attorneys, whether or not they ask to be listed. Attorneys can “claim” their listing for free and provide additional information about themselves and their professional expertise. Attorneys can also pay for advertising to boost their visibility to consumers looking for lawyers in particular categories (e.g., personal injury, divorce, etc.) and make it easier to get in touch (paying attorneys may also earn the label “Pro” on their listing).
Avvo also provides a numerical rating, from 1 to 10, for each attorney that has claimed their profile, based on information it obtains from public sources as well as information that attorneys provide. It does not publicly disclose the algorithm it uses so that attorneys cannot “game the system,” but Avvo’s site does state that by adding information to their profiles, lawyers can boost their ratings.
Davis’ class action complaint alleged, among other things, that Avvo’s 1 to 10 rating is skewed in that paying attorneys receive more favorable ratings. The complaint does not disclose any additional detail in support of this allegation.
The Court’s Ruling
The district court ruled that Avvo’s numerical rating is a pure expression of opinion, protected by the First Amendment, and hence not actionable under either the Lanham Act nor New York law. It noted that Avvo chooses which criteria will be used by its mathematical model, determines how to weigh them, and that results in a number (1 to 10) that reflects Avvo’s subjective evaluation of the attorney’s experience and ability. “A reasonable consumer would view an Avvo rating as just that—the defendant’s evaluation.” An Avvo rating, the court reasoned, is pure opinion, since it cannot be proven true or false. So, it dismissed Davis’ false advertising claims based on those ratings.
Potential Issues With the Ruling
In the author’s view, the district court’s dismissal is subject to criticism in that it focuses myopically on Avvo’s 1 to 10 rating in isolation, and ignores statements made by Avvo about its rating system and the process by which it arrives at the ratings.
Just because Avvo claims its rating system is an opinion is not enough. There are several avenues for the court to find actionable false advertising by Avvo.
First, courts recognize that “the line between fact and opinion is not always a clear one,” ONY v. Cornerstone Therapeutics, 720 F.3d 490, 496 (2d Cir. 2013), and that opinions can imply knowledge of facts. The leading Supreme Court case on the issue, Milkovich v. Lorain Journal Co., 497 U.S. 1, 110 S. Ct. 2695 (1990), dealt with a newspaper article that allegedly implied that the plaintiff had perjured himself during a trial. The court rejected an “opinion” privilege for defamation claims, because opinion may often imply fact. It colorfully noted that the statement “in my opinion John Jones is a liar,” implies a knowledge of facts which lead to that conclusion, and that can be as defamatory as the statement “Jones is a liar.” 497 U.S. at 18-19.
That principle arguably applies to Avvo’s ratings. A rating on Avvo could imply some basis in fact, especially given Avvo’s claim to use of a “mathematical model” to arrive at its ratings. Query whether individual consumers, understanding that ratings are based on a mathematical formula or not, don’t understand them as facts used to judge the lawyers on the site.
Second, misrepresentations about opinions have been held actionable as false advertising.
Here’s what Avvo’s website says about its numerical rating system:
• How the Avvo Rating is calculated. The rating is calculated using a mathematical model, and all lawyers are evaluated on the same set of standards.
• The Avvo Rating cannot be bought. At Avvo, all lawyers are treated equally. We don’t play favorites, and attorneys can’t pay us to improve their ratings.
Davis’ complaint cites these express Avvo claims and alleges that in fact they are false—that paying attorneys received more favorable treatment. These strike us as factual claims that could be shown to be true or false.
In ONY, the Second Circuit dealt with a false advertising claim concerning a scientific paper about medical products used to treat respiratory diseases in premature infants. The Second Circuit held that such scientific findings constitute non-actionable opinion. It further held that the touting of the scientific paper in promotional literature by a manufacturer was also not actionable. But it emphasized that the manufacturer’s promotional literature had not misrepresented or distorted the scientific paper’s findings. Had that been alleged, the Second Circuit opined, it would face a harder case.
Picking up on that dictum, the Southern District of New York in Mimedx Group v. Osiris Therapeutics, 2017 U.S. Dist. LEXIS 114105 (S.D.N.Y. 2017), allowed such a claim to proceed. There it was alleged that promotional literature by a manufacturer touting its tissue-graft products (used in treatment of wounds) and their superiority over those of a competitor had misrepresented the results of a scientific study about the products. Such misrepresentation about the study was actionable, even if the underlying study was non-actionable opinion under ONY. The Fifth Circuit reached a similar result in Eastman Chemical Company v. Plastipure, 775 F.3d 239 (5th Cir. 2014). Thus, misrepresentations about an opinion (especially a scientific opinion) may well be actionable, and likewise Avvo’s alleged misrepresentations about its rating system could be actionable.
Further support for this thesis is implied by a long-standing FTC rule about product endorsements. Although product endorsements are generally understood to be opinion or puffery, the FTC requires that where there are some connections between the product endorser and the product that would affect the credibility of the endorsement (e.g., a paid endorsement), that such connections be prominently disclosed in the advertising. 16 C.F.R. §255.5. So the fact that something is “opinion” does not necessarily insulate it from legal action.
As quoted above, Avvo makes specific, express claims about its rating system, claiming that “all lawyers are treated equally” and that “attorneys can’t pay us to improve their ratings.” These claims about Avvo’s rating system and the process it uses to arrive at its ratings could be falsified, even if the weight Avvo’s algorithm assigns to the various factors that go into the rating and the rating itself are “opinion.”
A review of complaints against Avvo on Internet review sites indicates some possible ways Avvo could manipulate its rating system to favor paying lawyers, even if its algorithm is being applied similarly to paying and non-paying attorneys.
First, Avvo’s site allows other attorneys to “endorse” a particular attorney, and the site avows that a larger number of endorsements will increase the attorney’s ratings. Many internet posters complain that this allows attorneys to game the system by having their friends, colleagues, and even complete strangers “endorse” one another to ramp up an attorney’s ratings. It is not clear whether Avvo polices this activity, but it is not difficult to see how such policing could be manipulated to favor paying attorneys in the process.
Second, Avvo’s site allows for clients to review attorneys and their services. Many attorneys have bitterly complained that negative reviews are often removed from their competitors’ profiles for no apparent reason. (Davis’ complaint mentions this issue but does not tie it to his claims.) Avvo’s site states that it has to review and approve client reviews before they are posted, and there is a dispute process that attorneys can initiate (which results in an automatic temporary takedown until the dispute is resolved.) Again, this dispute process could easily be abused to favor paying attorneys.
Third, Avvo’s site states that attorney ratings can go down over time because as positive information ages, it becomes less important in Avvo’s algorithm. Some attorneys have complained that when their ratings have gone down, that was closely followed by a solicitation from Avvo to become a paid member, and they suspect that there is an implicit connection between the rating and the solicitation.
What’s left as a question is whether Avvo uses its algorithm either directly or by application to favor paying attorneys, even though it represents that ratings “cannot be bought.” Davis’ complaint alleged factual misrepresentations about the process that Avvo uses to arrive at is ratings—that Avvo’s express claims are in fact false. Such should be actionable.
Davis did not appeal the dismissal, so the Second Circuit will not have an opportunity in this case to grapple with the opinion/fact dichotomy in false advertising cases. Future cases, however, may re-examine whether the distinction is subtler than the decision indicates.
