Netscout Systems, Inc. v. Gartner, Inc.

334 Conn. 396

• Court: Supreme Court of Connecticut
• Decided: January 21, 2020
• Docket: SC20079
• Status: Published

Opinion

*********************************************** The “officially released” date that appears near the be- ginning of each opinion is the date the opinion will be pub- lished in the Connecticut Law Journal or the date it was released as a slip opinion. The operative date for the be- ginning of all time periods for filing postopinion motions and petitions for certification is the “officially released” date appearing in the opinion.

All opinions are subject to modification and technical correction prior to official publication in the Connecticut Reports and Connecticut Appellate Reports. In the event of discrepancies between the advance release version of an opinion and the latest version appearing in the Connecticut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports, the latest version is to be considered authoritative.

The syllabus and procedural history accompanying the opinion as it appears in the Connecticut Law Journal and bound volumes of official reports are copyrighted by the Secretary of the State, State of Connecticut, and may not be reproduced and distributed without the express written permission of the Commission on Official Legal Publica- tions, Judicial Branch, State of Connecticut. *********************************************** NETSCOUT SYSTEMS, INC. v. GARTNER, INC. (SC 20079) Robinson, C. J., and Palmer, McDonald, Mullins, Kahn and Ecker, Js.

Syllabus

The plaintiff, which develops and sells information technology products, sought to recover damages for the defendant’s alleged violation of the Connecticut Unfair Trade Practices Act (CUTPA) (§ 42-110a et seq.) and for defamation in connection with certain allegedly false statements that the defendant had published in a market research report. The defendant provides consulting services to vendors of information tech- nology and publishes market research reports in which it rates vendors, including those that purchase its services. The defendant claims in its marketing materials that the opinions expressed in its reports are objec- tive and impartial. The defendant published a report in which it rated vendors, including the plaintiff, in the network performance monitoring and diagnostics market, and, although the plaintiff previously had accepted the defendant’s invitation to participate in the report’s evalua- tion process, the plaintiff had declined to purchase any consulting ser- vices from the defendant. The report included a graphical ranking of vendors, which was presented in the form of a square divided into four quadrants, and vendors were plotted along the axes of the graph on the basis of the defendant’s application of certain weighted criteria, including the subjective evaluations of the vendors’ customers. On the basis of its placement in the upper left quadrant of the graph, the plaintiff was designated a ‘‘challenger,’’ whereas vendors placed in the upper right quadrant received the more desirable designation of ‘‘leader.’’ The report defined the term ‘‘challenger’’ in relevant part as a vendor that is currently struggling to deal with new technical demands and rising expectations, whereas a ‘‘leader’’ is defined in relevant part as having comprehensive portfolios and the ability to handle multiple application and technology types. The report also included three specific ‘‘cautions’’ about the plaintiff and its limitations in its market. The plaintiff alleged that the statements in the report were false and defamatory, and that the defendant had engaged in a pay to play scheme, pursuant to which it rated vendors in a biased manner, on the basis of the amount of consulting services each vendor purchased from the defendant. The defendant moved for summary judgment, claiming that, because the statements in the report constituted protected speech, the plaintiff’s claims were barred by the first amendment to the United States constitu- tion. Although the trial court concluded that the defendant’s designation of the plaintiff as a challenger rather than as a leader constituted nonac- tionable opinion, it also determined that certain statements contained in the definition of the term ‘‘challenger’’ and in the cautions specific to the defendant either were factual or implied, undisclosed facts. With respect to those statements, the court nevertheless concluded that the plaintiff’s claims were barred by the first amendment because the defen- dant was a limited purpose public figure, the defendant’s statements were on a matter of public concern, and the plaintiff failed to prove by clear and convincing evidence that the defendant had made the state- ments with actual malice. The court also found that there was no evi- dence that the defendant’s placement of the various vendors in the quadrants on the graph was correlated to the amount of consulting services the vendors had purchased from the defendant. Accordingly, the trial court granted the defendant’s motion for summary judgment on both the defamation count and the CUTPA count and rendered judgment thereon, from which the plaintiff appealed. Held that the trial court properly granted the defendant’s motion for summary judgment, this court having concluded that all of the defendant’s statements were nonactionable expressions of opinion, there was insufficient evidence to create a genuine issue of material fact regarding the truth of the defendant’s claims of objectivity and impartiality, and the plaintiff failed to present sufficient evidence to support its pay to play claim: a reason- able person could construe the defendant’s designation of the plaintiff as a challenger only to be an expression of opinion, as ratings of products and services are inherently subjective and the criteria used by the defen- dant in ranking the vendors, including the subjective evaluations of the vendors’ own customers, and the weight assigned to those criteria could not be proven true or false; moreover, contrary to the conclusion of the trial court, this court concluded that, in light of the context in which they were made, the statements contained in the definition of the term ‘‘challenger’’ and in the cautions about the plaintiff were neither factual nor implied, defamatory statements of fact, as the parties were operating in a sophisticated market and their customers understood that the rating of products and services is based on inherently subjective evaluations, the report expressly stated that it consisted of the defendant’s own opinions that should not be construed as statements of fact, and the language used in the definitions and cautions, which was highly technical and employed terms of art specific to the plaintiff’s specific market, was abstract, unquantifiable, and comparative, such that the statements were insusceptible of being verified; furthermore, the fact that the defen- dant had claimed in its marketing materials that the opinions presented in its research reports are objective and impartial, in the absence of evidence establishing that those claims of objectivity were false, did not transform the nonactionable statements of opinion contained in the report into express or implied, defamatory factual statements, as such claims of objectivity, like puffery, are insusceptible of being proven true or false and are unlikely to induce reliance in a reasonable person viewing such statements. Argued January 14, 2019—officially released January 21, 2020

Procedural History

Action to recover damages for, inter alia, violations of the Connecticut Unfair Trade Practices Act, and for other relief, brought to the Superior Court in the judicial district of Stamford-Norwalk, where the court, Lee, J., granted the defendant’s motion for summary judgment and rendered judgment thereon, from which the plain- tiff appealed. Affirmed. Jason D. Frank, pro hac vice, with whom were Emily E. Renshaw, pro hac vice, and, on the brief, James A. Budinetz, Michael T. Ryan, Elizabeth G. Hays, pro hac vice, and Michael D. Blanchard, for the appellant (plaintiff). Derek L. Shaffer, pro hac vice, with whom were Andrew M. Zeitlin and, on the brief, Diane C. Polletta, John J. DiMarco, Robert L. Wyld, Patrick M. Fahey, Michael D. Bonanno, pro hac vice, Kathleen M. Sulli- van, pro hac vice, and Robert L. Raskopf, pro hac vice, for the appellee (defendant). Michelle M. Seery and Eugene Volokh, pro hac vice, filed a brief for the Reporters Committee for Freedom of the Press as amicus curiae. Jennifer M. DelMonico and Proloy K. Das filed a brief for the Connecticut Business and Industry Associ- ation as amicus curiae. Opinion

ECKER, J. The plaintiff, NetScout Systems, Inc., is in the business of developing and selling information technology products that allow its customers to man- age, monitor, diagnose and service their computer net- works. The defendant, Gartner, Inc., publishes research reports in which it rates vendors, such as the plaintiff, that sell and service various forms of information tech- nology. The defendant also sells consulting services to some of the vendors that it rates. In 2014, the defendant issued a research report (2014 report), in which it ranked the plaintiff lower than some of its competitors and made critical comments about the plaintiff. There- after, the plaintiff brought this action alleging that the defendant had engaged in a ‘‘pay to play’’ scheme, in which it rewarded vendors that purchased consulting services from the defendant by giving them high ratings in its research reports. The plaintiff claimed that the alleged pay to play scheme constituted a false and deceptive business practice under the Connecticut Unfair Trade Practices Act (CUTPA), General Statutes § 42-110a et seq., and that the 2014 report contained false and defamatory statements about the plaintiff. The defendant, in response, raised a defense premised on the theory that its rankings and commentary were pro- tected speech under the first amendment to the United States constitution.1 The trial court agreed with the defendant. The court concluded that the defendant’s 2014 report was consti- tutionally protected speech, and the plaintiff, as a lim- ited purpose public figure, was required to present evi- dence that the defendant had acted with actual malice. The court found that the plaintiff had failed to do so and, accordingly, rendered summary judgment for the defendant with respect to both claims on that ground. The court also determined that the CUTPA claim failed because the plaintiff had not presented evidence to support the factual predicate for its pay to play allega- tion due to its own expert witness’ inability to conclude that the defendant’s ratings were correlated to the dollar volume of consulting services that the vendors had pur- chased from the defendant. The plaintiff appealed to the Appellate Court, and we transferred the appeal to this court pursuant to General Statutes § 51-199 (c) and Practice Book § 65-1. We affirm the trial court’s judgment on the alternative ground that all of the defendant’s statements regarding the plaintiff were nonactionable expressions of opinion. I The record, viewed in the light most favorable to the plaintiff, reveals the following relevant facts and procedural history. The plaintiff, a Delaware corpora- tion with its principal place of business in the town of Westford, Massachusetts, is a prominent provider of computer network performance monitoring products and services. Its customers include numerous busi- nesses around the world, including commercial banks, airlines, financial service providers, and telecommuni- cation service providers, as well as governmental agen- cies and five branches of the United States military. In 2014, the plaintiff had revenues of approximately $400 million. The defendant, a Delaware corporation with its prin- cipal place of business in the city of Stamford, describes itself as ‘‘ ‘the world’s leading information technology research and advisory company.’ ’’ Among the defen- dant’s research publications are its ‘‘Magic Quadrant’’ research reports, which are marketed to buyers of vari- ous information technology products to assist them in selecting a vendor. The centerpiece of each report is a graphic rating of vendors called ‘‘the Magic Quadrant,’’ presented in the form of a square divided into quadrants. The horizontal axis of the square depicts the vendors’ ‘‘Completeness of Vision,’’ and the vertical axis depicts their ‘‘Ability to Execute.’’ As illustrated by the graphic, vendors with high ratings for both completeness of vision and ability to execute are placed in the upper right quadrant and are designated as ‘‘Leaders’’; those with a high rating for ability to execute and a low rating for completeness of vision are placed in the upper left quadrant and are designated as ‘‘Challengers’’; those with a high rating for completeness of vision and a low rating for ability to execute are placed in the lower right quadrant and are designated as ‘‘Visionaries’’; and those with low ratings for both completeness of vision and ability to execute are placed in the lower left quad- rant and are designated as ‘‘Niche Players.’’ In addition to its research activities and associated publications, the defendant provides consulting ser- vices, which it calls ‘‘Strategic Advisory Services,’’ to some vendors of information technology products.2 The analysts who market and provide these consulting ser- vices also are part of the team that determines the placement of vendors on the Magic Quadrant graphic. An analyst’s job performance is evaluated in part based on the amount of revenue he or she generates, including revenue from the sale of consulting services. The plain- tiff’s pay to play allegations rest substantially on the claim that the defendant’s vendor ratings were influ- enced by the vendors’ willingness to use and pay for the defendant’s consulting services. In early 2013, Julie Dempster, an account executive with the defendant assigned to the plaintiff’s account, and Jonah Kowall, the defendant’s research vice presi- dent for information technology operations manage- ment, exchanged a number of e-mails regarding the plaintiff. In an e-mail to Dempster dated January 4, 2013, Kowall wrote, ‘‘I don’t understand why we don’t speak to other people at [the plaintiff], nor do I understand why [the plaintiff does not] attend shows, or do any [strategic advisory services]. With [the plaintiff] in lots of research and potentially a [Magic Quadrant report] in 2013 I’m not quite understanding the relationship at all. I normally go out of my way to make things happen, and that’s not how it should be. [The plaintiff] ha[s] potentially the worst [analyst relations/public relations] and poor[est] marketing out of all the vendors I deal with.’’ In a subsequent e-mail dated April 5, 2013, Kowall wrote that the plaintiff did not ‘‘work with us like [its] competitors do . . . and [it does not] engage us for [strategic advisory services], which I think could help [it] a lot more strategically.’’ In an e-mail to Kowall dated July 8, 2013, Dempster wrote that ‘‘[t]he [strategic advi- sory services] day is so key for us to gain exposure and further licensing and engagements at [the plaintiff]!’’ On July 29, 2013, the defendant publicly announced that it would be publishing a new Magic Quadrant report—that is, the 2014 report—for the network per- formance monitoring and diagnostics (NPMD) market. By e-mail dated September 2, 2013, the defendant invited the plaintiff to participate in the evaluation pro- cess for inclusion in the 2014 report. The defendant included in the e-mail a definition of the NPMD market, the criteria for inclusion in the report, the evaluation criteria, a research and process timeline, and a vendor survey. The plaintiff accepted the invitation and returned the completed survey to the defendant on October 1, 2013. On December 2, 2013, Kowall sent an e-mail to his coworkers containing a draft of the Magic Quadrant graphic to be included in the 2014 report, in which the plaintiff was placed in the leaders quadrant. On December 3, 2013, Rebecca Noriega, a senior analyst and public relations manager with the plaintiff, sent an e-mail to Dempster indicating that the plaintiff was not going to participate in the ‘‘strategic advisory services day’’ that Dempster had suggested. The next day, Kowall circulated another e-mail containing a revised draft of the Magic Quadrant graphic, in which the plaintiff was placed directly on the line between the leaders quadrant and the challengers quadrant. Kowall noted in the e-mail that he ‘‘still think[s] [that one of the other vendors placed in the challengers quadrant] and [the plaintiff] should technically be leaders here, or at least on the line (as [the plaintiff] is)’’ and observed that the plaintiff and the other vendor may be ranked ‘‘too low’’ in their ability to execute. On December 5, 2013, Kowall circu- lated yet another e-mail with a second revised draft of the Magic Quadrant graphic. Kowall indicated that he had ‘‘tweaked some of the weightings to give us better control.’’ As a result, the plaintiff was placed higher on the vertical ability to execute axis but farther to the left on the horizontal completeness of vision axis, thereby situating the plaintiff deeper into the challengers quad- rant. During the defendant’s internal peer review of the draft 2014 report, several reviewers questioned why the plaintiff was placed in the challengers quadrant instead of in the leaders quadrant. On January 9, 2014, the defendant provided the plain- tiff with a draft of the 2014 report, in which the plaintiff was placed in the challengers quadrant of the Magic Quadrant graphic. The draft 2014 report also contained narrative evaluations of the plaintiff’s ‘‘[s]trengths’’ and three ‘‘[c]autions’’ concerning the plaintiff. On January 14, 2014, Noriega and the plaintiff’s vice president of marketing, Steven Shalita, participated in a conference call with Dempster and two other employees of the defendant, Vivek Bhalla and Colin Fletcher. Shalita indi- cated that the plaintiff was ‘‘struggling with the assess- ment and some of the language’’ the draft 2014 report contained. Bhalla and Fletcher attempted to explain the reasons for the plaintiff’s placement and requested written feedback from the plaintiff. On January 17, 2014, Shalita sent an e-mail to Kowall, Fletcher, Bhalla and Michele Severance, another employee of the defendant, attaching a response to the draft 2014 report and explaining in detail why the plaintiff believed that it had been ranked too low on its completeness of vision. In an e-mail to Shalita dated January 20, 2014, Kowall indicated that the defendant needed ‘‘specific changes to the text [of the draft 2014 report] BEFORE the call today’’; (emphasis omitted); which would be the last opportunity to discuss the plaintiff’s positioning in the Magic Quadrant graphic and the report’s narrative pro- viding certain cautions concerning the plaintiff. The plaintiff did not suggest any specific changes to the draft report, and the conference call never took place. On January 22, 2014, the plaintiff’s president and chief executive officer, Anil Singhal, sent a letter to the defen- dant’s chief executive officer, Eugene A. Hall, in which he wrote that he did not find it ‘‘amusing that [the defendant] had the temerity to place [the plaintiff] in the ‘[c]hallengers’ quadrant . . . .’’ Singhal also indi- cated that, if the defendant published the 2014 report, he would ‘‘vigorously contest [that] action through whatever means available.’’ Over the next several days, Singhal had a number of telephone conversations with Nancy Erskine, the defendant’s ombudsman, in which he reiterated that the plaintiff believed that the chal- lengers ranking was unfair and discriminatory and that the plaintiff wanted to be designated as the leading vendor in the NPMD market or, in the alternative, removed entirely from the 2014 report. Singhal also indicated that, if the defendant refused to redesignate the plaintiff as a leader or to remove it from the report, the plaintiff would take legal action. On March 6, 2014, the defendant published the final 2014 report. The plaintiff was placed in the challengers quadrant. The report defined ‘‘[c]hallengers’’ as ‘‘those with high market reach and large deployments. Once leaders in the network performance monitoring market, they are currently struggling to deal with new technical demands and rising expectations. These established NPMD vendors bring a substantial installed base, but also architectures, feature sets and pricing structures that require modernization (often in progress) to better compete with those in the [l]eaders quadrant.’’ The report also provided the following three ‘‘[c]autions’’ regarding the plaintiff: (1) ‘‘[the plaintiff] has a limited ability to expand beyond its network management heritage, which would be the next logical step (for example, into [application performance monitoring] or [information technology] operations analytics)’’; (2) ‘‘[o]ffering only a hardware-based deployment model limits [the plaintiff’s] ability to address growing soft- ware and [software as a service] solution demand’’; and (3) ‘‘[the plaintiff] is perceived as a conservative stalwart in the NPMD space, and lacks the reach and mind share that many smaller competitors have.’’ In separate marketing publications directed at purchasers of the defendant’s research products, the defendant represented that the opinions expressed in its reports were ‘‘impartial,’’ ‘‘independent,’’ ‘‘objectiv[e],’’ and ‘‘unbiased.’’ The 2014 report was made available to 40,000 sub- scribers to the defendant’s research services, more than 4000 of whom viewed the report. Vendors purchased online reprints of the report that were viewed approxi- mately 18,700 times, as well as 2000 paper reprints. After the 2014 report was published, the plaintiff filed this lawsuit. The two count complaint alleged, in the first count, that the defendant had violated CUTPA by engaging in a pay to play scheme in which vendors purchase consulting services from the defendant in exchange for high ratings in the defendant’s Magic Quadrant report. In the second count, the plaintiff alleged that the defendant had published false and defamatory statements about the plaintiff in the 2014 report. The defendant moved for summary judgment, claim- ing that the first amendment barred both the defamation claim and the CUTPA claim. It contended that, because the 2014 report was on a matter of public concern and the plaintiff is a limited purpose public figure, the plaintiff was required to present evidence that the defendant had acted with actual malice, such that its statements about the plaintiff were ‘‘made with actual knowledge that [they were] false or with reckless disre- gard of whether [they were] false.’’ (Internal quotation marks omitted.) Gleason v. Smolinski, 319 Conn. 394, 431, 125 A.3d 920 (2015). The plaintiff’s failure to pro- duce any evidence of actual malice, the defendant claimed, required the court to render summary judg- ment in its favor. The defendant also argued that the statements in the 2014 report were not factual asser- tions but statements of opinion protected by the first amendment. In addition, the defendant contended that the CUTPA claim was barred because (1) there was no evidence that the defendant’s business model departed from standard business norms, (2) there was no evi- dence that the plaintiff’s placement in the Magic Quad- rant graphic was related to vendor payments to the defendant for consulting services, (3) professional mal- practice is nonactionable under CUTPA, and (4) there was no evidence that the plaintiff suffered an ascertain- able loss as a result of the 2014 report. The trial court rendered summary judgment in the defendant’s favor on both counts of the complaint. It concluded that the defendant’s designation of the plain- tiff as a challenger rather than a leader was nonaction- able opinion. Although the court determined that cer- tain statements in the defendant’s definition of a challenger and in the three cautions regarding the plain- tiff either were factual or implied undisclosed facts, it agreed with the defendant that the plaintiff was a limited purpose public figure and that the defendant’s state- ments were on a matter of public concern, which meant that the plaintiff, to overcome the defendant’s first amendment protections, was required to present clear and convincing evidence that the defendant had made its statements with actual malice. After concluding that the plaintiff had not done so, the court held that both the defamation claim and the CUTPA claim were barred by the first amendment. In addition, with respect to the CUTPA claim, the court concluded that the plaintiff had presented no credible evidence that it was damaged by the defendant’s statements or that the placement of vendors in the Magic Quadrant graphic was correlated to the amount of consulting services that the respective vendors had purchased from the defendant. This appeal followed.3 The plaintiff contends that the trial court erroneously concluded that the plaintiff was required to prove actual malice on the basis of its incor- rect determination that the plaintiff was a limited pur- pose public figure and that the defendant’s statements about the plaintiff in the 2014 report were on a matter of public concern. The plaintiff also contends that, even if the trial court correctly determined that the actual malice standard applies, the court incorrectly deter- mined that the plaintiff had not presented evidence sufficient to create a genuine issue of material fact as to that issue. Finally, the plaintiff contends with respect to the CUTPA claim that the trial court incorrectly deter- mined that the plaintiff had not presented evidence sufficient to create a genuine issue of material fact as to whether it had been damaged or whether the Magic Quadrant vendor rankings were correlated to the pur- chase of consulting services. The defendant disputes these claims of error. It also contends that the trial court’s decision may be affirmed on the alternative grounds that (1) the defendant’s state- ments about the plaintiff were in all respects statements of opinion, not fact, and (2) even if the statements were factual, the plaintiff presented no evidence that they were false. We agree with the defendant that its state- ments about the plaintiff were expressions of opinion and, therefore, cannot provide the basis for a defama- tion claim. With respect to the plaintiff’s claims resting on the alleged falsity of the defendant’s representations of independence and objectivity in rendering its opin- ions, we reject those claims as well on this record. Accordingly, we affirm the judgment of the trial court.4 II The standard of review on summary judgment is well established. ‘‘Summary judgment shall be rendered forthwith if the pleadings, affidavits and other proof submitted show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law. . . . The scope of our appellate review depends upon the proper characteriza- tion of the rulings made by the trial court. . . . When . . . the trial court draws conclusions of law, our review is plenary and we must decide whether its con- clusions are legally and logically correct and find sup- port in the facts that appear in the record. . . . ‘‘In deciding a motion for summary judgment, the trial court must view the evidence in the light most favorable to the nonmoving party. . . . The party seek- ing summary judgment has the burden of showing the absence of any genuine issue [of] material facts which, under applicable principles of substantive law, entitle him to a judgment as a matter of law . . . and the party opposing such a motion must provide an evidentiary foundation to demonstrate the existence of a genuine issue of material fact.’’ (Internal quotation marks omit- ted.) Meyers v. Livingston, Adler, Pulda, Meiklejohn & Kelly, P.C., 311 Conn. 282, 289–90, 87 A.3d 534 (2014). First amendment principles always must remain firmly in mind when a court considers whether legal liability may attach to harm allegedly attributable to a defendant’s speech,5 and both the parties and the trial court in the present case understandably have framed much of their respective analyses under the rubric of the first amendment. The operative principles of sub- stantive law governing this appeal, however, primarily involve the common law of defamation. We make refer- ence to constitutional doctrine when necessary, but, in our view, the issues presented here are largely resolved by the straightforward application of defamation law. ‘‘At common law, [t]o establish a prima facie case of defamation, the plaintiff must demonstrate that: (1) the defendant published a defamatory statement; (2) the defamatory statement identified the plaintiff to a third person; (3) the defamatory statement was published to a third person; and (4) the plaintiff’s reputation suffered injury as a result of the statement. . . . ‘‘A defamatory statement is defined as a communica- tion that tends to harm the reputation of another as to lower him in the estimation of the community or to deter third persons from associating or dealing with him . . . .’’ (Citation omitted; footnote omitted; inter- nal quotation marks omitted.) Gleason v. Smolinski, supra, 319 Conn. 430–31. But it is not enough that the statement inflicts reputational harm. ‘‘To be actionable, the statement in question must convey an objective fact, as generally, a defendant cannot be held liable for expressing a mere opinion. See Mr. Chow of New York v. Ste. Jour Azur S.A., 759 F.2d 219, [229–30] (2d Cir. 1985) (no liability where restaurant review conveyed author’s opinion rather than literal fact); Hotchner v. Castillo-Puche, 551 F.2d 910, 913 [2d Cir.] ([a] writer cannot be sued for simply expressing his opinion of another person, however unreasonable the opinion or vituperous the expressing of it may be) [cert. denied sub nom. Hotchner v. Doubleday & Co., 434 U.S. 834, 98 S. Ct. 120, 54 L. Ed. 2d 95 (1977)].’’ (Internal quotation marks omitted.) Daley v. Aetna Life & Casualty Co., 249 Conn. 766, 795–96, 734 A.2d 112 (1999); see also Milkovich v. Lorain Journal Co., 497 U.S. 1, 13, 110 S. Ct. 2695, 111 L. Ed. 2d 1 (1990) (noting that, at early common law, defamatory opinion was actionable, but, ‘‘due to concerns that unduly burdensome defamation laws could stifle valuable public debate, the privilege of fair comment was incorporated into the common law as an affirmative defense to an action for defamation’’ [internal quotation marks omitted]). ‘‘A statement can be defined as factual if it relates to an event or state of affairs that existed in the past or present and is capable of being known. . . . In a libel action, such statements of fact usually concern a person’s conduct or character. . . . An opinion, on the other hand, is a personal comment about another’s conduct, qualifica- tions or character that has some basis in fact.’’ (Cita- tions omitted; emphasis in original.) Goodrich v. Water- bury Republican-American, Inc., 188 Conn. 107, 111, 448 A.2d 1317 (1982). It should surprise no one that the distinction between actionable statements of fact and nonactionable state- ments of opinion is not always easily articulated or discerned. See id., 112 n.5 (‘‘the difficulty in distinguish- ing fact from opinion has been recognized by a number of writers; see, e.g., [H.] Titus, ‘Statement of Fact versus Statement of Opinion—A Spurious Dispute in Fair Com- ment,’ 15 Vand. L. Rev. 1203 [1962]; [D.] Noel, ‘Defama- tion of Public Officers and Candidates,’ 49 [Colum.] L. Rev. 875, 878 [1949]; [N]ote, ‘Fair Comment’, 62 Harv. L. Rev. 1207 [1949]’’); see also Biro v. Condé Nast, 883 F. Supp. 2d 441, 460 (S.D.N.Y. 2012) (‘‘the seemingly simple proposition that expressions of opinion are pro- tected belies an often complicated task of distinguishing between potentially actionable statements of fact and nonactionable expressions of opinion’’ [internal quota- tion marks omitted]). See generally J. Kirchmeier, Note, ‘‘The Illusion of the Fact-Opinion Distinction in Defama- tion Law,’’ 39 Case W. Res. L. Rev. 867 (1988–1989). The difficulty arises primarily because the expression of an opinion may, under certain circumstances, reason- ably be understood to imply the existence of an underly- ing basis in an unstated fact or set of facts. See Gleason v. Smolinski, supra, 319 Conn. 435 (citing Goodrich v. Waterbury Republican-American, Inc., supra, 188 Conn. 117–19, for proposition that statements of opin- ion may be actionable statements of implied fact); 3 Restatement (Second), Torts § 566, p. 170 (1977) (‘‘[a] defamatory communication may consist of a statement in the form of an opinion, but a statement of this nature is actionable only if it implies the allegation of undis- closed defamatory facts as the basis for the opinion’’); see also Milkovich v. Lorain Journal Co., supra, 497 U.S. 19 (under common law, privilege of fair comment ‘‘did not extend to a false statement of fact, whether it was expressly stated or implied from an expression of opinion’’ [internal quotation marks omitted]). Context is a vital consideration in any effort to distin- guish a nonactionable statement of opinion from an actionable statement of fact. As this court previously has recognized, ‘‘[t]his distinction between fact and opinion cannot be made in a vacuum . . . for although an opinion may appear to be in the form of a factual statement, it remains an opinion if it is clear from the context that the maker is not intending to assert another objective fact but only his personal comment on the facts which he has stated. . . . Thus, while this distinc- tion may be somewhat nebulous . . . [t]he important point is whether ordinary persons hearing or reading the matter complained of would be likely to understand it as an expression of the speaker’s or writer’s opinion, or as a statement of existing fact.’’ (Citation omitted; emphasis in original; internal quotation marks omitted.) Goodrich v. Waterbury Republican-American, Inc., supra, 188 Conn. 111–12. A central feature of the analy- sis undertaken by virtually every court called on to distinguish opinion from fact involves a careful exami- nation of the overall context in which the statement is made. Courts will examine a variety of factors as part of this contextualized analysis, and, although no uniform test exists, the relevant case law focuses on a common set of considerations. One prevalent approach, first articulated by the United States Court of Appeals for the Ninth Circuit, uses a three part test to determine whether a reasonable fact finder could conclude that an expression of opinion implies an actionable assertion of fact: ‘‘(1) whether the general tenor of the entire work negates the impression that the defendant was asserting an objective fact, (2) whether the defendant used figurative or hyperbolic language that negates that impression, and (3) whether the statement in question is susceptible of being proved true or false.’’ Partington v. Bugliosi, 56 F.3d 1147, 1153 (9th Cir. 1995); see also Piccone v. Bartels, 785 F.3d 766, 772 (1st Cir. 2015) (distinguishing statement of fact from expression of opinion requires ‘‘an examination of the totality of the circumstances in which the specific challenged state- ments were made, including the general tenor and con- text of the conversation and any cautionary terms used by the person publishing the statement’’); Mr. Chow of New York v. Ste. Jour Azur S.A., supra, 759 F.2d 226 (court considers [1] context in which statements were made and circumstances surrounding statements, [2] whether language was ‘‘precise’’ and ‘‘literal’’ or, instead, was ‘‘loose, figurative or hyperbolic,’’ [3] whether statements were ‘‘objectively capable of being proved true or false,’’ and [4] whether statements imply ‘‘undisclosed defamatory facts as the basis for the opin- ion’’); Ollman v. Evans, 750 F.2d 970, 979 (D.C. Cir. 1984) (in determining whether average person under- stands speech at issue to be factual or expression of opinion, court considers [1] ‘‘the common usage or meaning of the specific language of the challenged state- ment itself,’’ [2] ‘‘the statement’s verifiability—is the statement capable of being objectively characterized as true or false,’’ [3] ‘‘the full context of the statement— the entire article or column, for example,’’ and [4] ‘‘the broader context or setting in which the statement appears’’), cert. denied, 471 U.S. 1127, 105 S. Ct. 2662, 86 L. Ed. 2d 278 (1985); Belly Basics, Inc. v. Mothers Work, Inc., 95 F. Supp. 2d 144, 145 (S.D.N.Y. 2000) (‘‘New York law dictates a three factor test to distin- guish statements of fact from statements of opinion: [1] whether the challenged statements have a precise and readily understood meaning; [2] whether the state- ments are susceptible of being proved false; and [3] whether the context signals to the reader that the state- ments are more likely to be expressions of opinion rather than fact’’ [internal quotation marks omitted]); K Corp. v. Stewart, 247 Neb. 290, 296, 526 N.W.2d 429 (1995) (‘‘a court looks at the nature and full content of the communication and to the knowledge and under- standing of the audience to whom the publication was directed, considering the specificity of the statement, its verifiability, the literary context, and the broader setting in which the statement appears’’). Without strip- ping these formulations of their nuance, they can be summarized as requiring analysis of three basic, over- lapping considerations: (1) whether the circumstances in which the statement is made should cause the audi- ence to expect an evaluative or objective meaning; (2) whether the nature and tenor of the actual language used by the declarant suggests a statement of evaluative opinion or objective fact; and (3) whether the statement is subject to objective verification. In light of the context in which the present case arises, we also find helpful the extensive case law from other jurisdictions involving speech that rates or reviews products, services or businesses. Some of these cases involve defamation or other claims against defen- dants in the business of rating the quality of sophisti- cated financial instruments of one kind or another; oth- ers are brought against defendants that rate or review more garden variety consumer products and services such as restaurants, hotels, and the like. See footnotes 6 through 8 of this opinion. Within the broader analytic framework described in the preceding paragraph, courts in these ‘‘ratings’’ cases resolve the issue of whether a reasonable person could conclude that the rating or review implies a statement of objective fact by considering whether (1) the speaker has exercised discretion when weighing the underlying data,6 (2) the defendant’s rating system uses abstract terms, such as a number between one and ten or ‘‘fuzzy descriptive phrases like ‘superb,’ ‘good,’ and ‘strong caution’ ’’; Browne v. Avvo, Inc., 525 F. Supp. 2d 1249, 1252 (W.D. Wn. 2007);7 and (3) the publication containing the rat- ings defined the terms that it used. See Smith v. Humane Society of the United States, 519 S.W.3d 789, 800–801 (Mo. 2017) (defendant’s use of term ‘‘puppy mill’’ to describe plaintiff’s kennel was nonactionable when defendant did not define term). Courts generally have held that ‘‘claims for defama- tion based upon ratings or grades fail because [ratings or grades] cannot be objectively verified as true or false and thus, are opinion . . . .’’8 Castle Rock Remodeling, LLC v. Better Business Bureau of Greater St. Louis, Inc., 354 S.W.3d 234, 241 (Mo. App. 2011); see also id., n.3 (citing cases). ‘‘Liability for [defamation] may attach, however, when a negative characterization of a person is coupled with a clear but false implication that the author is privy to facts about the person that are unknown to the general reader. If an author represents that he has private, [firsthand] knowledge which sub- stantiates the opinions he expresses, the expression of opinion becomes as damaging as an assertion of fact.’’ (Internal quotation marks omitted.) Mr. Chow of New York v. Ste. Jour Azur S.A., supra, 759 F.2d 225; cf. Biospherics, Inc. v. Forbes, Inc., 151 F.3d 180, 185 (4th Cir. 1998) (opinion based on fully disclosed facts is not actionable); Partington v. Bugliosi, supra, 56 F.3d 1156–57 (same). The case law in this area also makes it clear that an opinion that is based on the opinions of others does not imply defamatory facts and, there- fore, is not actionable. See Seaton v. TripAdvisor, LLC, 728 F.3d 592, 599–601 (6th Cir. 2013) (when defendant’s inclusion of plaintiff’s hotel on ‘‘ ‘Dirtiest Hotels’ ’’ list was based on opinions of hotel customers, it did not imply false facts but constituted nonactionable opin- ion); ZL Technologies, Inc. v. Gartner, Inc., 709 F. Supp. 2d 789, 798 (N.D. Cal. 2010) (‘‘[t]he use of a rigorous mathematical model to generate a ranking . . . based upon [subjective evaluations by vendors and their cus- tomers] does not transform [the defendant’s] opinion into a statement of fact that can be proved or dis- proved’’), aff’d, 433 Fed. Appx. 547 (9th Cir.), cert. denied, 565 U.S. 963, 132 S. Ct. 455, 181 L. Ed. 2d 295 (2011). ‘‘[T]he determination of whether a statement is opin- ion or rhetorical hyperbole as opposed to a factual representation is a question of law for the court.’’ Mr. Chow of New York v. Ste. Jour Azur S.A., supra, 759 F.2d 224; see also Smith v. Humane Society of the United States, supra, 519 S.W.3d 798 (‘‘[w]hether an alleged statement is capable of being treated as an opin- ion or as an assertion of fact is a question of law’’ [internal quotation marks omitted]). ‘‘Where the court cannot reasonably characterize the allegedly libelous words as either fact or opinion because, for example, innuendo is present, this becomes an issue of fact for the jury, which would preclude a directed verdict. This is similar to the rule which requires the jury to decide whether an ambiguous assertion is reasonably capable of a defamatory meaning.’’9 Goodrich v. Waterbury Republican-American, Inc., supra, 188 Conn. 112 n.5. With these principles in mind, we turn to the plain- tiff’s argument that the defendant’s statements concern- ing the plaintiff in the 2014 report reasonably could be construed as defamatory statements of fact. The plaintiff’s first claim of error is that the trial court incor- rectly determined that the defendant’s placement of the plaintiff in the challengers quadrant of the Magic Quadrant graphic was an expression of nonactionable opinion. We agree with the trial court that this speech was not factual and did not imply defamatory facts. Vendor ratings of this nature, in our view, typically are inherently subjective, and no reasonable person would consider a vendor’s placement in the Magic Quadrant graphic to be anything other than the expression of the rater’s opinion. This very issue was addressed in ZL Technologies, Inc. v. Gartner, Inc., supra, 709 F. Supp. 2d 800, in which the United States District Court for the Northern District of California observed that the defendant determined the placement of vendors in a Magic Quadrant graphic by ‘‘looking to a number of factors, [and] applying differing weights based on its subjective assessment of a company’s ability to execute and completeness of vision.’’ The relative weight assigned to the criteria and the criteria themselves ‘‘are not facts that can be proved true or false but a reflection of a subjective valuation . . . .’’ Id. The court in ZL Technologies, Inc., also observed that the defendant made the placement determination in part by consider- ing the subjective evaluations of the vendors’ custom- ers, which were merely opinions once removed. Id., 797, 800. Likewise, in the present case, the defendant assigned the plaintiff a place in the Magic Quadrant graphic on the basis of the defendant’s subjective evalu- ation of a variety of factors that were, in turn, assigned relative importance or ‘‘weigh[t]’’ in accordance with the subjective preferences embedded in its evaluative process, and by considering the subjective evaluations of the vendors’ customers. The trial court correctly determined that the placement of the plaintiff in the challengers section of the Magic Quadrant graphic was nonactionable opinion. This brings us to our single area of disagreement with the trial court. Although the trial court found that the defendant’s designation of the plaintiff as a challenger was nonactionable, it agreed with the plaintiff that the generic description of the term ‘‘[c]hallengers,’’ as used in the 2014 report, and the three particularized ‘‘[c]au- tions’’ explaining the plaintiff’s placement as a chal- lenger contained actionable statements of fact. With respect to the generic description of the challengers category, the trial court agreed with the plaintiff’s con- tention that the defendant’s rating impliedly conveyed a factual assertion that the plaintiff, having been desig- nated by the defendant as a challenger rather than a leader, lacked ‘‘comprehensive portfolios and the ability to handle multiple application and technology types’’ and was ‘‘currently struggling to deal with new technical demands and rising expectations.’’10 We disagree. To begin with, we return to the starting point of our analysis, which requires consideration of the context in which the challenged statement is made. The 2014 report set forth views regarding the relative strengths and weaknesses of vendors operating within a particu- lar commercial market. Whether expressed using color- ful jargon, numerical or letter grades, stars, or the stan- dard terminology of ‘‘good, better, best,’’ such ratings appear virtually any place a potential customer might look—in magazines and newsletters, television adver- tisements, billboards, waiting rooms, websites, and every other conceivable physical or electronic surface. Reasonable viewers, and certainly those consumers operating in the sophisticated market involved here, understand that these ratings normally rest, at bottom, on inherently and irreducibly subjective evaluations of value, quality and performance. This assumption does not mean that the speaker is at liberty to make false statements of fact merely by labelling them ‘‘opinions,’’11 but it does lead us to believe that the audience ordinarily recognizes that the context bespeaks caution, in the sense that most ratings of goods and services reflect an expression of evaluative opinion rather than verifi- able fact. Indeed, consistent with the nature and context of its undertaking, the defendant expressly states in the 2014 report that ‘‘[t]his publication consists of the opinions of [the defendant’s] research organization and should not be construed as statements of fact.’’ Although this statement does not automatically immunize the speaker against claims of defamation; see footnote 11 of this opinion; we conclude that the abstract, unquantifiable, and comparative nature of the defendant’s descriptions of the challenger and leader categories, using terms such as ‘‘comprehensive portfolios,’’ ‘‘multiple applica- tion and technology types,’’ and ‘‘struggling to deal with new technical demands and rising expectations,’’ ren- ders the statements irreducibly vague and insusceptible of being proved true or false. The statements, moreover, were not specific to the plaintiff but applied generally to all vendors placed in the referenced quadrants. A reasonable reader would understand that all aspects of the challengers description did not necessarily apply equally to all vendors.12 We reach a similar conclusion with respect to the statements contained in the 2014 report’s cautions about the plaintiff, namely, that the plaintiff ‘‘has a limited ability to expand beyond its network manage- ment heritage, which would be the next logical step (for example, into [application performance monitoring] or [information technology] operations analytics),’’ that ‘‘[o]ffering only a hardware-based deployment model limits [the plaintiff’s] ability to address growing soft- ware and [software as a service] solution demand,’’ and that the plaintiff ‘‘is perceived as a conservative stalwart in the NPMD space, and lacks the reach and mind share that many smaller competitors have.’’ The terms ‘‘lim- ited’’ and ‘‘limits’’ used to describe the plaintiff’s abilities are subjective, vague, unquantifiable, relative and unverifiable, and, although ‘‘reach’’ and ‘‘mind share’’ may be terms of art that have particular meanings in the relevant marketing field,13 these qualities are subjective and unverifiable. We also consider it significant that the 2014 report explained that, like the defendant’s decision about where to place the plaintiff in the Magic Quadrant graphic, these additional statements were based in part on the subjective evaluations of the plaintiff’s custom- ers. Accordingly, we conclude that the trial court incor- rectly determined that these statements were factual or implied defamatory statements of fact. The plaintiff contends that, even if these statements were not actionable as defamatory statements of fact in and of themselves, the defendant’s speech was still actionable because the defendant falsely claimed that its opinions were objective and impartial. Again, we disagree. Demonstrably false claims of objectivity, under certain circumstances, conceivably might supply sufficiently suggestive innuendo to transform otherwise nonactionable statements into defamatory speech, but this is not such a case. A false claim of objectivity, without more, cannot be defamatory of the plaintiff because the statement refers to the speaker rather than the vendor. We detect no obvious or even implied corre- spondence between the putative ‘‘objectivity’’ of the speaker and the perception of the speaker’s statement as fact or opinion. ‘‘Objective’’ speakers may publish subjective opinions, whereas biased or self-interested speakers may publish statements of fact. In the defama- tion context, the speaker’s claim to objectivity, whether true or not, does not turn that speaker’s negative state- ment of opinion about the plaintiff into an actionable statement of fact. That the defendant claimed a measure of objectivity in its marketing materials did not convert its vaguely worded, comparative and evaluative state- ments into express or implied defamatory factual state- ments. See Seaton v. TripAdvisor, LLC, supra, 728 F.3d 600 (defendant’s claim that it was most trusted advisor and that it provided ‘‘ ‘the whole truth’ ’’ about rated hotels did not render defendant’s opinion about plain- tiff’s hotel actionable); ZL Technologies, Inc. v. Gar- tner, Inc., supra, 709 F. Supp. 2d 797–98 (defendant’s statements that its analysis was ‘‘fact-based and knowl- edge-centric, built on objectivity, and founded on a methodology it says ensures the ultimate objectivity’’ were ‘‘insufficient to transform the tenor of the rankings in the [Magic Quadrant] [r]eport from opinion to fact’’ [internal quotation marks omitted]).14 The concept of ‘‘puffery,’’ although typically applied outside the defamation context, usefully illuminates the basic point.15 A claim of objectivity, like claims of integ- rity, credibility, high ethical norms and the like, is often considered nonactionable puffery because it is unlikely to induce reliance and is insusceptible to being proved true or false. See Singh v. Cigna Corp., 918 F.3d 57, 63 (2d Cir. 2019) (‘‘general statements about reputation, integrity, and compliance with ethical norms are inac- tionable puffery, meaning that they are too general to cause a reasonable investor to rely upon them’’ [internal quotation marks omitted]); Intermountain Stroke Cen- ter, Inc. v. Intermountain Health Care, Inc., 638 Fed. Appx. 778, 788 (10th Cir. 2016) (advertising that speaks generically to caliber of defendant’s product is ‘‘classic puffery’’ and incapable of being proved true or false); Boca Raton Firefighters & Police Pension Fund v. Bahash, 506 Fed. Appx. 32, 37 (2d Cir. 2012) (‘‘[t]he statements alleged in the [plaintiff’s] complaint regard- ing [the] integrity and credibility [of the company whose stock was being sold] and the objectivity of [the defen- dant security rating provider’s] credit ratings are the type of mere puffery that we have previously held to be not actionable’’ because ‘‘no reasonable purchaser of [the stock] would view statements such as these as meaningfully altering the mix of available information about the company [whose stock was being sold]’’ [internal quotation marks omitted]); ECA, Local 134 IBEW Joint Pension Trust of Chicago v. JP Morgan Chase Co., 553 F.3d 187, 206 (2d Cir. 2009) (defendant’s statements regarding its good reputation for integrity were ‘‘no more than ‘puffery’ ’’); ECA, Local 134 IBEW Joint Pension Trust of Chicago v. JP Morgan Chase Co., supra, 206 (‘‘[the] [p]laintiffs conflate the importance of a . . . reputation for integrity with the materiality of [the defendant’s own] statements regarding its repu- tation,’’ and ‘‘[n]o investor would take [the defendant bank’s statement that it set the standard for best prac- tices] seriously . . . for the simple fact that almost every investment bank makes these statements’’). Putting aside our conceptual problems with the plain- tiff’s theory of defamation premised on the allegedly false claims of objectivity, we also agree with the trial court that the plaintiff failed to present any credible evidence that the Magic Quadrant rankings in the 2014 report were based on the amount of consulting services that each vendor purchased from the plaintiff. Specifi- cally, the court noted that the two vendors who spent the most on consulting services were ranked compara- bly to or lower than the plaintiff in the Magic Quadrant graphic—which the plaintiff’s expert had disregarded when he found a correlation between expenditures and rankings. Although the plaintiff’s expert attempted to salvage his pay to play correlation by reference to these two vendors’ higher rankings in other Magic Quadrant reports, and the plaintiff claims on appeal that, at the very least, this broader historical point created a genu- ine issue of material fact, we remain unpersuaded. The unavoidable reality is that, with respect to the publica- tion at issue, the plaintiff’s own expert was unable to provide an opinion supporting the plaintiff’s fundamen- tal theory of wrongdoing without entirely disregarding the data pertaining to the two highest paying vendors in the rankings.16 We agree with the trial court’s conclusion that ‘‘[s]uch manipulation of evidence does not create a genuine issue of material fact or support a question for the jury’’ as to whether a correlation between expen- ditures and rankings existed. The plaintiff also contends that, even if the trial court properly rejected its expert’s conclusions correlating expenditures and rankings, there was other evidence that would allow a jury to infer that the defendant’s claims of objectivity and impartiality were false. The plaintiff points in particular to evidence that the defen- dant had told the plaintiff that the defendant ‘‘need[ed] to fix things with [the plaintiff] considering the focus our research agenda has on [the plaintiff’s] market in 2013’’ and that ‘‘market leaders need to be represented properly.’’ Singhal, the plaintiff’s president and chief executive officer, testified at his deposition that one of the defendant’s analysts had told him that the plaintiff should spend more money on marketing, which Singhal interpreted as ‘‘code . . . for spending more money on [the defendant’s] services.’’ The evidence also estab- lished that the plaintiff was placed in the leaders quad- rant in early drafts of the 2014 report, but, after Kowall ‘‘tweaked’’ the underlying weightings, the plaintiff ended up in the challengers quadrant. The plaintiff addi- tionally points to evidence that the defendant repeat- edly referred to the plaintiff as a leader in various docu- ments, thereby effectively admitting that the plaintiff should have been placed in the leaders quadrant. We agree with the trial court’s findings that this evi- dence was insufficient to create a genuine issue of mate- rial fact regarding the truth of the defendant’s claims of objectivity and impartiality. With respect to the defendant’s references to the plaintiff as a leader, the trial court correctly observed that all but one of these references were made solely in connection with the plaintiff’s market share, which was consistent with its placement in the challengers quadrant. With respect to the earlier draft placing the plaintiff in the leaders quadrant, it is indisputable that the defendant’s analysts engaged in a lengthy process of repositioning various vendors based on a variety of criteria. Although Kowall at one point wrote that he believed that the plaintiff ‘‘technically’’ should be placed in the leaders quadrant, or that it at least should be placed on the line between the leaders quadrant and the challengers quadrant, and other analysts questioned the plaintiff’s ranking as a challenger, these exchanges took place at a time when the analysts were openly debating the position of ven- dors, including the plaintiff, in the graphic. The evidence showed that the positions of almost all of the vendors in the Magic Quadrant graphic worsened over time as the analysts ‘‘tweaked’’ the data, and the plaintiff’s posi- tion relative to the other vendors remained relatively constant. There is no suggestion in any of these commu- nications that the adjustments were made on the basis of the vendors’ expenditure levels. We likewise con- clude that the statement made by the defendant’s ana- lysts to the plaintiff that the defendant needed to ‘‘fix things’’ with the plaintiff and that the plaintiff should spend more on marketing reflect nothing more than a recommendation for curative action and cannot reason- ably support an actionable inference that the plaintiff’s ranking was dependent on pay to play expenditures. The availability of the defendant’s consulting services was no secret to anyone; the defendant publicly dis- closed the fact that some of the vendors that it ranked in its Magic Quadrant reports purchased consulting ser- vices from it, enabling readers to consider the defen- dant’s claims of objectivity and impartiality in their proper context. The plaintiff contends that any ambiguity in the fore- going evidence must, at the summary judgment stage, be construed in its favor because a reasonable fact finder could resolve the ambiguity and conclude that the defendant’s claims of objectivity and impartiality were false. The flaw in this argument is that it misappre- hends the particular nature of the ambiguity at issue. The plaintiff is correct that, generally speaking, sum- mary judgment will not be rendered if there is an ambi- guity in the evidence such that a reasonable person might decide a disputed issue of material fact in favor of either the plaintiff or the defendant. See, e.g., H.O.R.S.E. of Connecticut, Inc. v. Washington, 258 Conn. 553, 565, 783 A.2d 993 (2001) (ambiguous mean- ing of critical deposition testimony precluded summary judgment). It also is true that, in defamation cases, a jury issue can arise when an expression of opinion contains an ambiguity that reasonably can be under- stood to convey by implication an actionable factual assertion.17 See Goodrich v. Waterbury Republican- American, Inc., supra, 188 Conn. 112 n.5 (referring to ‘‘the rule which requires the jury to decide whether an ambiguous assertion is reasonably capable of a defama- tory meaning’’). But the ambiguity at work in the present case is of a different kind. When we say that certain statements in the 2014 report are properly characterized as vague and ambiguous, we mean that the statements could not be understood by a reasonable juror to imply a factual statement; their ambiguity does not invite the listener to infer a latent factual assertion but, rather, suggests an imprecise and irreducibly subjective mean- ing that cannot be understood to convey a statement of fact. Thus, the statements are expressions of opinion as a matter of law. See, e.g., Mr. Chow of New York v. Ste. Jour Azur S.A., supra, 759 F.2d 224 (whether statement is factual or expression of opinion is question of law). This conclusion does not preclude summary judgment in the present context but, instead, compels it. Accordingly, we conclude that the trial court correctly determined as a matter of law that no reasonable juror could find by a preponderance of the evidence18 that the defendant’s claims of objectivity and impartiality were defamatory.19 Any evidence presented by the plain- tiff that might arguably support an inference that the defendant had a conflict of interest would not render its statements actionable in a defamation action. See Abu Dhabi Commercial Bank v. Morgan Stanley & Co., 651 F. Supp. 2d 155, 179 (S.D.N.Y. 2009) (‘‘[t]he existence of conflicts of interest alone typically is not sufficient to establish that [the] defendants ‘knowingly’ made a false and misleading statement’’). Indeed, as we have explained, a reasonable purchaser of the defen- dant’s research reports would understand that compar- ative ratings of products and services cannot be strictly objective. See Browne v. Avvo, Inc., supra, 525 F. Supp. 2d 1253 n.1 (‘‘[c]omparisons and comparative ratings are often based as much on the biases of the reviewer as on the merits of the reviewed: they should, therefore, be relied upon with caution’’); id., 1253 (stating, with respect to defendant’s comparative ratings, that ‘‘that and $1.50 will get you a ride on Seattle’s new South Lake Union Streetcar’’ [internal quotation marks omitted]); Castle Rock Remodeling, LLC v. Better Business Bureau of Greater St. Louis, Inc., supra, 354 S.W.3d 241 (ratings and grades ‘‘cannot be objectively verified as true or false’’). Our conclusion that the trial court properly rendered summary judgment for the defendant on the defamation claim is also dispositive of the plaintiff’s claim that the trial court improperly rendered summary judgment for the defendant on the CUTPA claim. Because the defen- dant’s statements about the plaintiff in the 2014 report were nonactionable expressions of opinion, and because the plaintiff failed to present sufficient evi- dence to support its pay to play claim, we are compelled to conclude that the plaintiff has failed to establish a viable claim within the purview of CUTPA. In summary, we conclude that the trial court incor- rectly determined that the 2014 report contained certain statements about the plaintiff that a reasonable juror could find to be defamatory and either to be factual or to imply undisclosed defamatory facts. In our view, all of the statements that the defendant made about the plaintiff were expressions of nonactionable opinion. We further conclude that, at least in the absence of any credible evidence tending to establish that the defen- dant’s claims of objectivity and impartiality were prov- ably false and that it was engaged in a pay to play scheme, such speech cannot support either the plain- tiff’s defamation claim or its CUTPA claim. Accordingly, we affirm the trial court’s judgment on the alternative ground that all of the challenged statements by the defendant were nonactionable expressions of opinion. The judgment is affirmed. In this opinion the other justices concurred. 1 The first amendment to the United States constitution provides in rele- vant part: ‘‘Congress shall make no law . . . abridging the freedom of speech . . . .’’ This prohibition is made applicable to the states through the due process clause of the fourteenth amendment to the United States constitution. See, e.g., 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 489 n.1, 116 S. Ct. 1495, 134 L. Ed. 2d 711 (1996). 2 The precise nature of these consulting services is unclear from the record, but that has no bearing on our analysis or the disposition of this appeal. 3 We subsequently granted the application of the Reporters Committee for Freedom of the Press to file an amicus curiae brief in support of the defendant’s contention that its statements about the plaintiff were nonaction- able opinion and the application of the Connecticut Business and Industry Association to file an amicus curiae brief on the issue of whether this court should stop applying the so-called ‘‘cigarette rule’’ in CUTPA cases and, instead, apply the ‘‘substantial unjustified injury’’ test. 4 Because we affirm the trial court’s judgment on the basis of our conclu- sions that the defendant’s statements regarding the plaintiff were expres- sions of nonactionable opinion and that the plaintiff failed to present evi- dence of a pay to play scheme, we need not address the other claims raised by the plaintiff on appeal. 5 See Gertz v. Robert Welch, Inc., 418 U.S. 323, 325, 94 S. Ct. 2997, 41 L. Ed. 2d 789 (1974) (‘‘[t]his [c]ourt has struggled for nearly a decade [since New York Times Co. v. Sullivan, 376 U.S. 254, 84 S. Ct. 710, 11 L. Ed. 2d 686 (1964)] to define the proper accommodation between the law of defamation and the freedoms of speech and press protected by the [f]irst [a]mendment’’); Goodrich v. Waterbury Republican-American, Inc., 188 Conn. 107, 114–15, 448 A.2d 1317 (1982) (observing that common-law ‘‘privi- lege of ‘fair comment’ ’’ was ‘‘elevated to constitutional status’’ in New York Times Co. v. Sullivan, supra, 376 U.S. 254). The distinction between fact and opinion at the center of the present case figures prominently in first amendment law, as well. See, e.g., Milkovich v. Lorain Journal Co., 497 U.S. 1, 20, 110 S. Ct. 2695, 111 L. Ed. 2d 1 (1990) (‘‘a statement of opinion relating to matters of public concern which does not contain a provably false factual connotation will receive full constitutional protection’’); see also 3 Restatement (Second), Torts § 566, comment (c), pp. 172–73 (1977) (entitled ‘‘Effect of the Constitution’’). Under Milkovich, the characterization of the subject matter at issue as a matter of ‘‘public’’ or ‘‘private’’ concern may affect the first amendment analysis. See Milkovich v. Lorain Journal Co., supra, 19 (‘‘a statement on matters of public concern must be provable as false before there can be liability’’ under first amendment [emphasis added]). Milkovich appears to have left open the issue of whether a statement of opinion that involves a matter of private concern could be actionable under state law consistent with the first amendment. See id., 20. There is no need to address the issue here because the plaintiff’s claims fail. 6 See Compuware Corp. v. Moody’s Investors Services, Inc., 499 F.3d 520, 529 (6th Cir. 2007) (credit rating based on ‘‘subjective and discretionary weighing of complex factors’’ was nonactionable opinion); Aviation Charter, Inc. v. Aviation Research Group/US, 416 F.3d 864, 871 (8th Cir. 2005) (‘‘sub- jective interpretation of multiple objective data points’’ constitutes opinion); ZL Technologies, Inc. v. Gartner, Inc., 709 F. Supp. 2d 789, 798 (N.D. Cal. 2010) (when defendant weighed importance of certain facts differently based on relative value that it assigned to different criteria, rating was nonaction- able opinion), aff’d, 433 Fed. Appx. 547 (9th Cir.), cert. denied, 565 U.S. 963, 132 S. Ct. 455, 181 L. Ed. 2d 295 (2011); Browne v. Avvo, Inc., supra, 525 F. Supp. 2d 1252 (when underlying data were weighted based on defendant’s subjective opinions regarding relative importance of various attributes, com- parative rating of plaintiff attorney was nonactionable opinion); Castle Rock Remodeling, LLC v. Better Business Bureau of Greater St. Louis, Inc., 354 S.W.3d 234, 242–43 (Mo. App. 2011) (when defendant’s rating system weighted objective data, consumer rating was nonactionable opinion). 7 See also Aviation Charter, Inc. v. Aviation Research Group/US, 416 F.3d 864, 867, 870–71 (8th Cir. 2005) (when defendant assigned ratings of ‘‘Does Not Qualify,’’ ‘‘Silver,’’ ‘‘Gold’’ and ‘‘Platinum,’’ rating of plaintiff as ‘‘Does Not Qualify’’ was not sufficiently factual to be susceptible of being proved true or false); Jefferson County School District No. R-1 v. Moody’s Investor’s Services, Inc., 175 F.3d 848, 856 (10th Cir. 1999) (defendant’s use of vague phrases such as ‘‘negative outlook’’ and ‘‘ongoing financial pressures’’ to describe plaintiff’s financial condition constituted nonaction- able opinion [internal quotation marks omitted]); Castle Rock Remodeling, LLC v. Better Business Bureau of Greater St. Louis, Inc., 354 S.W.3d 234, 242–43 (Mo. App. 2011) (defendant’s ‘‘C’’ rating of plaintiff was not ‘‘suffi- ciently factual to be susceptible of being proved true or false’’ [internal quotation marks omitted]); K Corp. v. Stewart, supra, 247 Neb. 296–97 (defendant’s use of subjective terms such as ‘‘minimum standards,’’ ‘‘unac- ceptable level of deterioration,’’ ‘‘acceptable quality,’’ and ‘‘first class condi- tion’’ constituted expression of opinion). 8 See, e.g., Seaton v. TripAdvisor, LLC, 728 F.3d 592, 600–601 (6th Cir. 2013) (defendant’s placement of plaintiff’s hotel on ‘‘ ‘Dirtiest Hotels’ ’’ list constituted nonactionable opinion); ZL Technologies, Inc. v. Gartner Group, Inc., 433 Fed. Appx. 547, 548 (9th Cir.) (defendant’s placement of plaintiff in Magic Quadrant graphic constituted nonactionable opinion), cert. denied, 565 U.S. 963, 132 S. Ct. 455, 181 L. Ed. 2d 295 (2011); Compuware Corp. v. Moody’s Investors Services, Inc., 499 F.3d 520, 529 (6th Cir. 2007) (defen- dant’s credit rating of plaintiff constituted nonactionable opinion); Aviation Charter, Inc. v. Aviation Research Group/US, 416 F.3d 864, 870–71 (8th Cir. 2005) (defendant’s safety rating of plaintiff air charter services provider constituted nonactionable opinion); Jefferson County School District No. R-1 v. Moody’s Investor’s Services, Inc., 175 F.3d 848, 856 (10th Cir. 1999) (defendant’s credit rating of plaintiff school district constituted nonaction- able opinion); Mr. Chow of New York v. Ste. Jour Azur S.A., supra, 759 F.2d 229 (with exception of statement that plaintiff restaurant served Peking duck in one dish instead of traditional three dishes, which was susceptible of being proved true or false, defendant’s review of plaintiff’s restaurant constituted nonactionable opinion); Browne v. Avvo, Inc., supra, 525 F. Supp. 2d 1251 (defendant’s comparative rating of attorneys constituted non- actionable opinion); Smith v. Humane Society of the United States, supra, 519 S.W.3d 800–801 (defendant’s inclusion of plaintiff’s kennel in report listing worst puppy mills in state constituted nonactionable opinion); Castle Rock Remodeling, LLC v. Better Business Bureau of Greater St. Louis, Inc., 354 S.W.3d 234, 242–43 (Mo. App. 2011) (defendant’s rating of plaintiff’s business constituted nonactionable opinion). 9 See Burns v. Telegram Publishing Co., 89 Conn. 549, 552, 94 A. 917 (1915) (‘‘[i]t is only when the court can say that the publication is not reasonably capable of any defamatory sense, that the court can rule, as [a] matter of law, that the publication is not libelous, or withdraw the case from the jury, or order a verdict for the defendant’’ [internal quotation marks omitted]); 3 Restatement (Second), supra, § 614, p. 311 (‘‘[t]he court determines . . . whether a communication is capable of bearing a [defama- tory] meaning, and . . . [t]he jury determines whether a communication, capable of a defamatory meaning, was so understood by its recipient’’); 3 Restatement (Second), supra, § 614, comment (d), p. 312 (‘‘if, in the opinion of the court, the question [of whether a statement is capable of bearing a defamatory meaning] is one on which reasonable men might differ, it is for the jury to determine which of the two permissible views they will take’’). 10 The trial court did not expressly address the former statement, which was included in the defendant’s definition of ‘‘[l]eader’’ in the 2014 report, but concluded that a reasonable juror could find that the latter statement was factual in nature, or at least that it implied undisclosed defamatory facts. 11 See Milkovich v. Lorain Journal Co., supra, 497 U.S. 18 (observing that statement couched as opinion will be actionable if it implies assertion of objective fact); Gleason v. Smolinski, supra, 319 Conn. 435 n.34 (‘‘the ‘mere recitation of prefatory phrases such as ‘‘in my opinion’’ or ‘‘I think’’ will not render innocent an otherwise defamatory statement’ ’’). 12 The trial court also concluded that an assertion of fact is conveyed in the defendant’s statement that challengers had ‘‘architectures, feature sets and pricing structures that require modernization (often in progress) to better compete with those in the [l]eaders quadrant.’’ The plaintiff does not address or defend this particular finding on appeal. In any event, we would conclude that this statement is an expression of opinion for the same reasons the other statements in the challengers and leaders descriptions are expres- sions of opinion. 13 See, e.g., T. Marabella, Note, ‘‘Elemental Copyright: The Complexity of Ideas and the Alchemy of Mind-Share,’’ 90 B.U. L. Rev, 2149, 2161 (2010) (‘‘[mind share] is a term used in the marketing and advertising world, defined as the likelihood that a consumer will think of a particular brand when a type of product is mentioned’’). 14 Section 566 of the Restatement (Second) of Torts makes the same point with respect to the closely related concept of the speaker’s honesty or sincerity when expressing an opinion. See 3 Restatement (Second), supra, § 566, comment (c), p. 175 (‘‘Should it be a significant issue whether the expression of opinion was the actual opinion of the defendant? Though an asserted factual statement that it is the actual opinion of the defendant may be held to be implied, its truth or falsity would apparently have no effect on the defamatory character of the communication.’’). But see Goodrich v. Waterbury Republican-American, Inc., supra, 188 Conn. 123 (‘‘[a] comment is fair when [inter alia] it . . . is an honest expression of the writer’s real opinion or belief’’ [emphasis in original; internal quotation marks omitted]). The court in Goodrich quoted the decision of the New York Court of Appeals in Briarcliff Lodge Hotel, Inc. v. Citizen-Sentinel Publishers, Inc., 260 N.Y. 106, 118–19, 183 N.E.193 (1932), which cited no authority to support the statement. Goodrich v. Waterbury Republican-American, Inc., supra, 123–24. It has been held in cases involving claims of financial fraud that a false claim of objectivity may render a statement actionable. See, e.g., In re International Business Machines Corporate Securities Litigation, 163 F.3d 102, 109 (2d Cir. 1998) (‘‘an opinion may still be actionable if the speaker does not genuinely and reasonably believe it or if it is without a basis in fact’’); Abu Dhabi Commercial Bank v. Morgan Stanley & Co., 651 F. Supp. 2d 155, 176 (S.D.N.Y. 2009) (same); State v. Moody’s Corp., Superior Court, judicial district of Hartford, Docket No. X04-HHD-CV-XX-XXXXXXX-S (May 10, 2012) (54 Conn. L. Rptr. 116, 121) (when defendant claimed that it had employed ‘‘specific policies and procedures . . . to protect its indepen- dence and objectivity,’’ statement was actionable under theory of misrepre- sentation because it was susceptible to being proved false). In our view, there would seem to be a significant difference between the business of rating securities, which is predictive and widely perceived as employing standardized econometric methodologies requiring a high degree of mathe- matical and technical skill, and the business of rating products and services, which involves commentary on an existing state of affairs and is generally understood to be inherently subjective in nature. In light of this difference, as well as the differences between the elements of a fraud claim and a defamation claim; see F. Harper & M. McNeely, ‘‘A Syntheses of the Law of Misrepresentation,’’ 22 Minn. L. Rev. 939, 947 (1938) (‘‘[I]n misrepresenta- tion, the defendant has misled the plaintiff himself, whereas in defamation he has misled a third person to the plaintiff’s loss. Thus, in the case of misrepresentation, the plaintiff complains because he himself has been induced to enter into some commercial transaction by his reliance upon the misstatements which the defendant has made. In defamation, however, he complains because some third person has believed or is presumed to have believed the defendant’s misstatements about him, and has been induced or is presumed to have been induced to act in reliance thereon to the plaintiff’s loss.’’ [Footnote omitted.]); it is far from clear to us that the same rule should apply in the context of a defamation claim brought in the ratings context by the subject of the purportedly defamatory opinion. 15 ‘‘Puffery is an exaggeration or overstatement expressed in broad, vague, and commendatory language.’’ Castrol, Inc. v. Pennzoil Co., 987 F.2d 939, 945 (3d Cir. 1993); see also Newcal Industries, Inc. v. IKON Office Solution, 513 F.3d 1038, 1053 (9th Cir. 2008) (‘‘[u]ltimately, the difference between a statement of fact and mere puffery rests in the specificity or generality of the claim’’), cert. denied, 557 U.S. 903, 129 S. Ct. 2788, 174 L. Ed. 2d 290 (2009). 16 The plaintiff contends that the trial court prematurely concluded that there was no correlation between expenditures and placement in the Magic Quadrant graphic because the trial court barred the plaintiff ‘‘from taking discovery related to [the defendant’s] relationship with, payments from, and treatment of companies other than [the plaintiff] . . . .’’ The plaintiff, how- ever, has neither challenged the correctness of the court’s discovery ruling on appeal nor explained what additional information it would require to determine whether there was a correlation between expenditures and place- ment in the Magic Quadrant graphic. Indeed, the plaintiff acknowledges that it had information concerning the total amounts that other vendors had paid the defendant for consulting services and the placement of the two vendors who paid most in other Magic Quadrant reports for markets other than NPMD. 17 For example, this court found that an ‘‘ambiguous’’ statement appearing in a newspaper article that the plaintiff had charged the city of Bridgeport for 8000 gallons of gasoline, but, when measured, the volume had ‘‘shrunk [1000] gallons,’’ was actionable because it was ‘‘susceptible of a libelous meaning [i.e., overcharging], and capable of being injurious to the plaintiff. It might have been understood by one reading it as implying that the plaintiff was guilty of some dishonest action in connection with the loss of the gasoline. So understood, the article was libelous. It being capable of such a meaning, it was proper for the plaintiff in his complaint, by way of innuendo, to allege that it conveyed such a meaning to its readers.’’ (Internal quotation marks omitted.) Burns v. Telegram Publishing Co., 89 Conn. 549, 551–52, 94 A. 917 (1915). 18 The defendant argues that the plaintiff is required to prove falsity by clear and convincing evidence because the plaintiff is a limited purpose public figure and that the 2014 report constituted speech on a matter of public concern. See Gleason v. Smolinski, supra, 319 Conn. 432. We need not address this issue because we have concluded that the plaintiff’s claims fail even under the preponderance standard. 19 This conclusion is reached in the context of a defamation claim, and the plaintiff is a vendor alleging harm arising from statements made as part of the defendant’s rating and review of the plaintiff’s information technology products and services. Different claims in a different context brought by a differently situated plaintiff may fare differently. Compare Abu Dhabi Commercial Bank v. Morgan Stanley & Co., 651 F. Supp. 2d 155, 176, 179 (S.D.N.Y. 2009) (when defendant rating agencies, which held themselves out as being unbiased and independent from companies that they rated, knew that ‘‘the ratings process was flawed . . . that the portfolio was not a safe, stable investment, and . . . that [they] could not issue an objective rating because of the effect it would have on their compensation, it may be plausibly inferred that [they] knew they were disseminating false and misleading ratings’’); see also footnote 14 of this opinion.