consumer false advertising plaintiffs can't use surveys the way competitors can

Sinatro v. Barilla America, Inc., No. 22-cv-03460-ASK (N.D. Cal. Sept. 30, 2026)

Deception of a substantial number of reasonable consumers, even well less than a majority, can be the basis of Lanham Act liability. But consumer protection cases have difficulty handling this, and so they make up all-or-nothing tests. Here, that means nothing.

closeup on challenged claim

Sinatro sued Barilla for the usual California claims, alleging that the claim “Italy’s #1 Brand of Pasta” on its boxes deceived consumers into thinking it was made in Italy from Italian ingredients when, in fact, that pasta was made in the United States from non-Italian ingredients. (This claim has now been removed.) On each of the relevant boxes, the side label with nutrition facts also included the representation “Made in the U.S.A. with U.S.A. and imported ingredients” in white text directly beneath Barilla’s name and location, which are statements required by FDA regulation.

side disclosure

nutrition panel

In Sinatro’s survey, 57 percent indicated that the statement “does communicate” that “[t]he product’s ingredients are sourced from Italy,” 32.3 percent indicated that the statement “does not communicate this,” and 10.7 percent selected the “Don’t know / not sure” option. A separate conjoint survey found a 9.9% price premium associated with the claim.

plaintiffs' survey prompt

Barilla’s survey replicated Sinatro’s but included five—instead of just one—possible interpretations. In response, 84.0 percent of respondents answered affirmatively for “[t]he brand is the most popular pasta brand in Italy,” while 42.3 percent did so for “[t]he product’s ingredients are sourced from Italy.” Perhaps unsurprisingly, Barilla’s surveyor decided to remove certain “inattentive” or “confused” respondents from the sample, reducing the 42.3 percent “from Italy” figure to either 34.2 or 27.5 percent, depending on how you counted, whereas the 84.0 percent “most popular” figure remained nearly unchanged (dropping only to 83.5%).

Barilla survey prompt

I would note that these reduced figures are more than enough to win a Lanham Act false advertising claim. Also, the reduction came from counting misunderstandings of other claims, which is one way of declaring a bunch of actual consumers to be unreasonable. (The 34.2 percent figure comes from eliminating respondents who indicated that a “[n]on-GMO” label means that the product is gluten free, while 27.5 percent comes from eliminating these same respondents and also those who indicated that the Challenged Representation means that “[t]he brand is the oldest producer of pasta in Italy.”)

Did the survey even matter? Front-of-product deception is actionable under consumer protection law despite qualifiers on the side panel, but only if the front is “unambiguously” deceptive. This isn’t the Lanham Act concept: a front label is not “ambiguous” “simply because it is susceptible to two possible meanings” but rather “when reasonable consumers would necessarily require more information before reasonably concluding that the label is making a particular representation.” That is, reasonability is interposed as a separate, normative screen, regardless of what a substantial number of consumers actually think.

The court excluded plaintiff’s survey under Daubert. (Again, there’s plenty of Lanham Act precedent, but courts tend not to cite across regimes, which given the different meanings of “ambiguous” is probably overall a good thing, though a comprehensive harmonization would also be welcome.) The court found the consumer perception survey biased (and also criticized the conjoint analysis because it “impermissibly measures the undifferentiated impact of all interpretations of the Challenged Representation without isolating the impact attributable to only misleading interpretations”).

The survey asked only whether the package conveys that “[t]he product’s ingredients are sourced from Italy,” and provided only three possible responses: an affirmative answer, a negative answer, and “[d]on’t know / not sure.” This was biased because by only asking that one question, the survey steered respondents “to that very meaning and gives them no opportunity to indicate other meanings.” This seems inconsistent with most Lanham Act survey cases, which generally (1) allow questions about specific messages, at least after an initial “did the ad say anything about?” question, and (2) treat offering yes/no/not sure options as enough to avoid impermissible leading.

The court also held that the survey “fails to capture the possibility that respondents understand the Challenged Representation to have multiple meanings,” and thus couldn’t assist the trier of fact.  The don’t know/not sure option wasn’t enough because the survey design “bias[es] the survey by suggesting to respondents, at least implicitly, that they should believe there is at least some sort of relationship between the different items when the possibility might not even have occurred to the vast majority of consumers who see the items.”

Survey respondents who answered [the] question affirmatively might have never made an association with Italian ingredient sourcing had the survey not fed it to them. And respondents who normally would think the statement means something else entirely may have abandoned that opinion upon intuiting from the survey itself what they should take the Challenged Representation to mean. A “[d]on’t know /  not sure” option does not negate such effects, because the question by its design elicits them before respondents even go to choose an answer.

Query: Under this analysis, is any closed-ended question allowable?

Without the survey, “no reasonable juror could conclude on the remaining record evidence that ‘members of the public are likely to be deceived.’” Named plaintiffs’ testimony wasn’t enough because, “[a]lthough surveys and expert testimony regarding consumer expectations are not required, a few isolated examples of actual deception are insufficient.”

To the extent that the words “Italy’s #1 Brand of Pasta” make a claim about brand—namely, that Barilla is the foremost brand of pasta in Italy—undisputed record evidence showed that Barilla had leading market share, and the court also found this the most plausible meaning.

So, would consumers necessarily require more information before concluding that the pasta was Italian in origin? The court rejected the most “plaintiff-friendly” reading of 9th Circuit precedent, which would find ambiguous only those labels that “ma[k]e no representation at all” or, like “Nature Fusion,” “[are] so devoid of any concrete meaning that there [is] nothing from which any inference could be drawn or on which any reasonable belief could be based.” That would make ambiguity essentially a puffery defense. The court wasn’t willing to do that; to do so would mean finding the front label ambiguous “because it is at least amenable to the (undisputedly false) meaning that the Plaintiffs here advance—i.e., that the pasta is made in Italy with Italian-sourced ingredients.”

But using this standard “would effectively mean that any label susceptible to at least one concrete false meaning would be actionable.” [Note that this is in fact the rule in Lanham Act cases, where a substantial number of reasonable consumers would receive this message as shown by acceptable evidence of actual consumer reception. This is because Lanham Act plaintiffs are allowed to aggregate consumer behavior even if we accept substantial variation among consumers; probabilistic liability is ok for them, whereas consumer plaintiffs must rely on a presumption that all consumers are likely to be deceived, so courts set the standard higher (even though that’s not strictly logically entailed).] Anyway, to do so would be inconsistent with Trader Joe’s, which found that “100% New Zealand Manuka Honey” could support multiple meanings, including such false meanings as “that its contents were 100% derived from the Manuka flower.”

Thus, Barilla’s front label was ambiguous, and the survey, even if admissible, didn’t address ambiguity and even reinforced the conclusion that the brand popularity meaning was the predominant one.  “[A] reasonable consumer would necessarily require more information before concluding that the statement should instead be understood to mean something else entirely. Accordingly, Barilla has made no ‘unambiguously deceptive’ claim.” And the side label clarified US origin, making it not misleading as a matter of law.

The court further justified its holding as providing an easily administrable rule to lower courts and clear guidance to retailers. “[S]ellers may have confidence that so long as they do not make ‘unambiguously deceptive’ statements on their front labels, they may rely on their back and side labels to clear up misunderstandings.” And if consumers unreasonably persist in being fooled, so much the worse for them!

compounding pharmacy must face drugmaker's Lanham Act claims based on false 503B compliance statements

Pacira BioSciences, Inc. v. Nephron Sterile Compounding Center, LLC, No. 3:23-cv-05552-CMC, 2026 WL 2267750 (D.S.C. Aug. 6, 2026)

Pacira sells an FDA-approved pain management drug called EXPAREL. Nephron allegedly made various false and misleading statements about two of Nephron’s competing compounded drug products. Specifically, Nephron allegedly represented those products as compliant with Section 503B of the FDCA, and as safe, effective, and superior to EXPAREL. The court granted partial summary judgment to both sides on the resulting Lanham Act claim.

Section 503B defines an “outsourcing facility” as a facility that “is engaged in the compounding of sterile drugs”; “has elected to register as an outsourcing facility”; and “complies with all of the requirements of [Section 503B].” Outsourcing facilities may distribute compounded drugs without obtaining a patient-specific prescription. They are also exempt from the FDCA’s new-drug approval process and certain labeling and supply-chain requirements. But they have to satisfy 11 statutory “conditions.”

The condition central to this case restricts outsourcing facilities’ use of “bulk drug substances” in compounding. They can’t be used unless the government properly identifies a clinical need or a drug shortage. In 2016, the FDA said it needed time to evaluate nominations and wouldn’t act when a bulk drug substance appeared on a list of “Category 1” substances the FDA maintained on its website, which “may be eligible for inclusion on the [clinical need] list, were nominated with adequate supporting information for FDA to evaluate them, and [had] not been identified by FDA as presenting significant safety risks.”

Nephron compounded BKK, a compounded drug product consisting of bupivacaine, ketorolac, and ketamine; later, it switched to RKK, with ketorolac, ketamine, and ropivacaine. Nephron advertised and sold them as “503B products” and as lower-cost alternatives to Pacira’s EXPAREL. In August 2019, Pacira wrote to the FDA urging it to take action against Nephron over its “illegal marketing and promotion” of BKK. It sued in November 2023, alleging false advertising of both products in terms of 503B compliance, comparability to/substitutability for Exparel, and related safety/efficacy claims.

Laches barred claims based on BKK; Pacira knew about the alleged false advertising since at least August 2019 and failed to offer more than conclusory argument to rebut the presumption of prejudice arising from its more than four-year delay in filing suit (period borrowed from state law). Even without the presumption of prejudice, defendants noted that several key witnesses no longer work for Pacira; some witnesses often could not recall basic facts related to the litigation during their depositions, and two, “both of whom reside beyond the subpoena power of the court, indicated they do not intend to appear at trial.” Of the 28 Nephron employees identified in Pacira’s Rule 26(a) initial disclosures, 17 were now former employees, including 11 who “would have still been current employees had Pacira filed within three years of January 2020.” “The unavailability of witnesses, faded memories, and changes in personnel are all indicators of evidentiary prejudice.”

But RKK came later and was allegedly discovered later. Nephron identified no “facts otherwise indicating a lack of vigilance” on the part of Pacira.

The court first held that the RKK Section 503B claims weren’t statements of opinion. Though lay claims about law are generally opinion, sometimes a law is “so clear on its face that no good faith doubt concerning its interpretation [is] possible, even without an explicit statement from [a court or agency].” In that circumstance, the law’s meaning is “so clear as to be a fact for Lanham Act purposes,” and Section 503B is “clear on its face.” Determining whether an outsourcing facility’s claims of compliance with Section 503B’s “bulk drug substance” provision are true “requires no more than consulting two lists.”

Nephron argued that Section 503B permits an outsourcing facility to market and sell a drug (say, RKK) without FDA approval as long as the individual components of that drug appear on the drug shortage list. But that wasn’t true: the drug shortage exception allows outsourcing facilities to “compound using bulk drug substances” only if “the drug compounded from such bulk drug substance appears on the drug shortage list.” Section 503B also defines “compounding” as “the combining, admixing, mixing, diluting, pooling, reconstituting, or otherwise altering of a drug or bulk drug substance to create a drug.” That’s what making RKK is.

Nor did the FDCA preclude Pacira’s Lanham Act claim. The “logical building blocks” of Pom Wonderful apply with equal force to “drug marketing, medical device labeling, cosmetics branding, or any other kind of marking or representation which would fall under both the Lanham Act and the FDCA,” Thus, “courts adjudicating Lanham Act disputes outside of the food and beverage realm” have “almost uniformly” interpreted POM Wonderful “to cover products in other FDA-regulated industries,” including pharmaceuticals. The issue here was not an exception because it wouldn’t “require the expertise of the FDA to resolve,” would call on the court “to make an original determination on an issue committed to the FDA’s discretion,” or would “otherwise conflict with an affirmative policy judgment by the FDA.”

The court declined to consider alleged falsehoods not identified in the complaint: statements allegedly conveying that RKK was compounded “in compliance with Section 503B” or “in a 503B-compliant outsourcing facility,” along with the specific statements that RKK is “generic to or substitutable for EXPAREL,” provides “improved value” over EXPAREL, “reduces post-operative pain[,] complications, risk of readmission, length of stay, [and] patient morbidity and mortality,” “eliminates opioid related adverse drug effects,” and has “low reported incidents of nausea and vomiting.”

Were they false? Nephron’s product lists and pricing sheets were distributed to members of group purchasing organizations (GPOs), healthcare systems, and individual hospitals, enough to constitute commercial advertising/promotion. They listed RKK among Nephron’s “503B Products.” This was a false claim as matter of law: “By listing RKK under that heading, Nephron necessarily communicated to purchasers that RKK belonged to the class of products meeting the statute’s requirements; no other conclusion can be drawn from its designation as a ‘503B Product.’”

But statements that Nephron is “a leading manufacturer of 503B outsourcing products” weren’t literally false; appearing at the bottom of a Nephron “Opioid Free” product handout featuring BKK, RCK, and RKK “could be understood as conveying that Nephron manufactures ‘503B outsourcing products’ as part of its broader business, without necessarily implying that BKK, RCK, and RKK fall within that category.” Pacira’s survey didn’t test the “leading manufacturer” statement, so Nephron got summary judgment.

Nephron’s “503B outsourcing facility” logo appeared throughout Nephron’s materials, including the “Opioid Free” handout and the product label for RKK. Again, this was too ambiguous to be literally false. Here Pacira did have a survey, presenting respondents with a sample marketing email about BKK. Pacira’s expert concluded that a net 14.8% of respondents believed BKK was a Section 503B-compliant drug. “That percentage would ordinarily be sufficient to support a finding that an advertisement misled or tended to mislead consumers.”

The court denied Nephron’s motion to exclude. First, though the email shown to participants focused only on BKK, the deception rate calculated by Butler could be “extrapolated” to RKK; the difference went to weight rather than admissibility:

This is not a scenario where Pacira seeks to extend survey results to an entirely separate and untested statement (as with the “leading manufacturer” statement). Rather, the same “503B outsourcing facility” logo appeared on the product labels for both BKK and RKK, and it is not immediately clear consumers would perceive the logo differently depending on the product to which it was affixed. Under these circumstances, the weight to be given [the survey’s] findings, as applied to RKK, is for the jury to decide.

Other challenges also went to weight: Using closed-ended questions after open-ended questions has both critics and supporters and their limits could be the subject of cross-examination. The survey asked, “Which of the following, if any, are message(s) communicated by this email?” and offered the following response options:

1. BKK, an admixture of bulk drug substances from Nephron, is a 503B compliant drug

2. BKK, an admixture of bulk drug substances from Nephron, is opioid free

3. BKK, an admixture of bulk drug substances from Nephron, is packaged in syringe form

4. None of these

5. Don’t know / unsure

Respondents who selected the first option were asked a follow-up open-ended question to assess why they believed this.

Nephron argued that closed-ended questions should be avoided when testing consumer deception because they are “highly susceptible to demand effects and focalism.” But “there is nothing inherently wrong in using closed-ended questions to test consumer impressions.” They can be “suitable for assessing choices between well-identified options” and “may remind respondents of options that they would not otherwise consider or which simply do not come to mind as easily.” And open-ended questions have their own problems, such as decreasing reporting of actually-held views: “respondents answering open-ended questions may be less likely to report some information that they would reveal in response to a closed-ended question when that information seems self-evident or irrelevant” (quoting Shari Diamond). They also require subjective coding: “Because respondents are answering in their own words and may not provide precise answers, many answers are simply not clear enough to definitively evaluate or categorize reliably.” Also, the “Don’t know / unsure” answer choice is a recognized way “to screen out respondents who may not have an opinion on the issue under investigation.”

Nephron also argued that the survey expert made “excessive and biased” changes to the stimulus shown to the control group. The control group saw an email that (1) removed both instances of the “503B outsourcing facility” logo; (2) added a disclaimer stating, “Note: the admixture of bulk drug substances sold as BKK is not compounded in compliance with Section 503B”; (3) removed the statement “Nephron continues to strive to provide drugs that are critically short and meet all needs of the Country!”; and (4) modified the statement “This is something that will be huge in helping hospitals decrease opioid and Exparel use” by deleting the reference to EXPAREL. The disclaimer, Nephron argued, was “overly broad and inaccurate” because it “would likely lead participants to conclude that no part of BKK is 503B-compliant.”

It was logical to address only BKK in the disclaimer because that was the specific product being tested by the survey. “The court struggles to see why respondents needed to be informed of the compliance status of bupivacaine, ketorolac, and ketamine individually when they were being asked about the admixture BKK.” Anyway, this was for cross-examination.

Finally, Nephron argued that the survey was an improper “reading test” because the stimulus was visible while respondents answered questions. Again, experts disagree on this, so it was for cross-examination and competing expert testimony. Thus, there was a genuine issue of material fact on misleadingness for use of “503B outsourcing facility” logo on the “Opioid Free” product handout and RKK’s product label.

Statements that RKK (1) is “generic to or substitutable for EXPAREL” and (2) provides “improved value” over EXPAREL: Pacira didn’t show any use of the exact phrase. The closest it got was one email stating “Nephron produces BKK, RKK, & RCK as generic, clinical need medications as requested by physicians across the country.” That didn’t mention EXPAREL at all.  Summary judgment granted on (1).

The “improved value” statement was in the “Opioid Free” product handout. But it didn’t make an express or implicit comparison to EXPAREL, and it was puffery.

RKK safety and efficacy: these statements claimed that RKK (1) “reduces post-operative pain[,] complications, risk of readmission, length of stay, [and] patient morbidity and mortality”; (2) “eliminates opioid related adverse drug effects”; and (3) has “low reported incidents of nausea and vomiting.” These weren’t establishment claims [side note that courts used to be much more open to the argument that scientific/health claims are establishment claims by necessary implication, which seems right to me], and Pacira showed only that they weren’t substantiated.

Of the remaining statements, there was no genuine dispute that “503B Product” was material, given that GPO customers “required Nephron to warrant that the compounded drugs it sold complied with Section 503B.” “Parties do not typically bargain for guarantees on matters they consider unimportant. That Nephron’s contracts with GPOs included an express warranty of Section 503B compliance is compelling evidence that such compliance mattered to purchasers of RKK.”

But the materiality of the “503B outsourcing facility” logo was still in dispute.

There was also a factual issue of injury based on evidence that certain customers who had previously purchased EXPAREL (1) began buying RKK when it came on the market, (2) simultaneously reduced their purchases of EXPAREL, and (3) were members of GPOs that had received Nephron pricing sheets listing RKK as a “503B Product.” So these issues, and damages and/or disgorgement, remained for trial.

IPSC Breakout Session 3 Trademark Confusion & Consumer Perception

Trademark Confusion as a Matter of Law, Andrew Michaels

Is infringement a proposition about the world, or about the law? Fact: an empirical predictive question about the world. Law: a normative judgment about whether there should be legal responsibility—is the confusion likely enough that we should find infringement? 2d/Minority view: Q of law based on underlying facts. 9th and majority view: Q of fact reviewed for clear error. 9th said that LOC decisions have “limited precedential value” b/c they stand on their own facts, reducing the need for de novo review. But that might be a reason to treat it as a Q of law to get more consistency and predictability. Issue of law would make it easier to decide on SJ/without trial, compared to claims that SJ should usually be avoided.

Judges and juries are thought to be good at different things. Jury: community; judge: compare with other cases/predictability allowing businesses to order their affairs more easily. Easier to decide on SJ: might help prevent bullying of parodists, other users.

Appellate v. trial court: underlying factors of intent, actual confusion might be better assessed by trial court; balancing/weighing of factors and legal comparison might be more suited for appellate court. Some factors may be more factual: evidence of actual confusion; intent (witness testimony/credibility). Similarity of marks should be legal because the jury has to ask “compared to what?” whereas the court can look at other cases. Same with products. Strength of mark conceptually should be Q of law; commercial strength is a matter of fact for witness testimony.

Lisa Ramsey: Matal v. Tam—constitutional issue exists, and LOC is a speech protective doctrine, implicating Bose. JDI even says that MTD can be ok because of contextual considerations. If it can be resolved on a MTD, it can be a Q of law.

RT: Bose v. Consumers Union on de novo review when the facts have constitutional significance. On the “compared to what” for similarity of marks/similarity of products? the theory is “similar enough that consumers are likely to confuse them.”

James Dabney: time was that likely confusion would be enough for an injunction, not damages or disgorgement; now things are different.

Google v. Oracle—is this legal or factual? Similar issue of mixed question of law & fact.

Q: right to jury trial?

A: could ask them questions about the factors; could ask for an advisory jury verdict, which they do a lot for patent obviousness/did with GvO. Multifactor=often an issue of law.

McKenna: LOC factors were made up; makes it feel more fact bound b/c courts think they have to walk through the factors even when they are ill-fitting. The legal standard is supposed to be: substantial number of reasonable consumers. Look at negligence where courts are more willing to grant SJ because they are more willing to consider what reasonableness is.

Factors and Fictions: The Empirical Collapse of the Likelihood-of-Confusion Test Across the Federal Courts, Thomas Reichert

Every circuit makes the same 4 commitments: (1) the test is flexible; bright lines misfire; (2) no factor is dispositive; (3) the set is open so you can bring in other considerations/add factors; (4) provides structure and allows appellate review. But: How often is this true?

Used an LLM to read every confusion opinion 1970-2025, temperature set low to inject less randomness, and ask whether the court considered a factor and how strongly it favored/disfavored confusion. The model is not trying to judge factor weight itself, just trying to identify what the court said about how the factor weighed. Around 11,000 opinions analyzed. Courts analyze 6.35 factors/case; only 40% consider them all.

Hand audited 1002 codings; 97% agreement on weight and 100% on direction.

The key factors: similarity of marks and proximity of goods/services. If both favor confusion, predict confusion; if both don’t, predict lack of confusion. Can predict 93.5% of every federal TM case. Consistent across circuits and time, though less in 8th circuit where the case count is small, and there’s a dip in the 90s (his hypothesis is domain names). The other factors operate as “structured overrides.” Defeaters are where both factors favor confusion but the court finds none. A lot of cases: no actual confusion, high buyer sophistication, good faith adoption, and weak mark strength. Substitutes: a predicate factor was weak/divided, but confusion found anyway: strong/famous mark; bad faith intent.

The test is already hierarchical. Courts should say so. Appellate practice wrongly rewards factor by factor mark. Could right size discovery/do less initial discovery. Tell juries the machinery: model jury instructions in 9th Circuit already tell juries what weighs more and we could do more.

Could apply the same questions to © fair use; sentencing.

Betsy Rosenblatt: has been done for © fair use—you may not want to reinvent the wheel. But one interesting thing about © fair use that might or might not match w/TM infringement is that how one comes out on transformativeness tends to predict how the case comes out. It doesn’t mean that transformativeness is the whole game; but it influences how the other factors work rather than rendering them unimportant. In general we may want those other factors to be doing more work than they’re doing, not less. You may have identified a problem rather than a solution. Should juries perhaps pay more attention to sophistication? Right now they don’t have a good definition.

Should parody be a special case b/c the factors work differently?

Q: Fed Cir has criticized TTAB for relying too much on similarity, so that result is pretty funny (the Fed Cir was most likely to rely on the 2).

McKenna: it’s not that the other factors just come out—the question of how much similarity there is b/t marks and goods are not found in nature. It’s not a © comparison. All the other information is just influencing the judgment about similarity. It’s context for which you understand levels of similarity, informed by all the other information. So sequencing discovery would be difficult b/c you’d be ruling out the contextual information you need to make judgments about similarity of marks and similarity of products.

A: we’re measuring the opinions, not the reasoning process.

McKenna: sure, but your prescriptions make assumptions about how the reasoning works.

A: Crowding in the market can definitely change similarity assessments.

Q: if courts were honest and said it’s a 2 element test, with a determination made through a bunch of subfactors, would that work better?

A: that’s the next paper. You can create a flowchart of how to do the analysis with substitutes/defeaters. You could do a burden shift! [Burden of production I assume, not burden of proof.]

Ramsey: dilution doesn’t consider relatedness of goods—does that matter?

A: didn’t look for any correlation w/dilution.

Ramsey: some courts say strength increases likely confusion, but academics and parody cases say that strength can decrease likely confusion b/c people know what the real thing looks like.

A: strength moves w/the verdict generally, but can substitute for proximity if the mark is very strong.

Ramsey: should separate out commercial & conceptual strength & see what happens.

Q: the other factors were originally not relevant to competing goods situations; practitioners got into the habit of applying Polaroid/etc. in all circumstances when it wasn’t needed in the direct competition cases. That would support the empirical observation that competition and similarity, the two pre-Polaroid metrics, were actually always the most important.  

21st Century Trademark Surveys, Rebecca Tushnet (with Chris Sprigman & Stephan Tontrup)

A statutory interpretation component: what do the terms in the statute like affiliation and connection actually mean? Weird that we don’t have much of an answer after 80 years, isn’t it? So we believe the definitions we are using are grounded in the proper legal meaning of affiliation et cetera.

The empirical part: we currently don’t tell survey respondents (or jurors) what “affiliation” etc. means and we also don’t have any good reason to think that they know what it means for legal purposes, which doesn’t include references—if you think of Sprite when you see Poppi Lemon-Lime, there can’t be deception about affiliation or connection because you really did think that, but a layperson could say “yes, there’s a connection”—the survey may not even be revealing mistakes of law, as Sotomayor et al have discussed with parodies, but mistakes about the meaning of the words used in the survey! So let’s try to fix that with a training module as in genericness surveys and see what happens. Including allowing a response “this is about the trademark/trademark owner.”

And implications for jury instructions: survey respondents and jurors are in the same position.

Larger questions: there are lots of areas where we want to know how some audience perceives communication: 1A compelled speech/will you be associated w/the statement; labor law: how employees perceive employer speech—we don’t ask the workers! True threats—hypothetical reasonable person.  But only in the Lanham Act do we actually use surveys! [Probabalistic—less than half can still be a large number of people for economically, socially, or politically significant messages. Why is probabilistic thinking persuasive in TM & not other areas? Plaintiffs’ bar? Courts willing to credit that “substantial numbers” matter even if not majority b/c they can imagine the harm to the substantial consumer mass/the consumer mass may not be imagined to share any other minoritarian identities (or may be imagined to need special protection—cite Ann Bartow on gender)? Possible lesser importance of public interest lets courts defer to surveys in TM and rely on policy preferences in 1A—though that lack of interest in reality on the ground is not necessarily good for 1A jurisprudence.

Rosenblatt: affiliation and approval are easy to get wrong; even experts get them wrong. Pattern jury instruction?

Q: we’re in the post literate era: disconnect b/t regular people and lawyers. If literacy rates are going down, we need to define terms for them.

Ramsey: courts focusing on text of statute: approval language concerns me. We don’t want people to be confused about permission. Don’t ask compound questions. “goes along with” is a bad definition too.

RT: approval and permission aren’t the same thing but this is where mistake of law comes in. There is an issue with repeating questions too—that’s more likely to get a “yes” somewhere in there.

Rothman: Working on project w/Joel Steckel—one of the things we worked w/was mini survey about meaning of these terms and people were using lay definitions. How should they be defined?

An Axe to Grind? The Legal History and Trademark Challenges of Guitars, Mark Blankenship

When does a guitar shape identify a type of guitar v. manufacturer? Sears Roebuck catalog is the precursor to Amazon and Temu, making gear affordable to players who didn’t live near a music store. Different claims over time—Japanese “knockoffs” that eventually resolved into new body shapes as well as some generic ones. Other issues: German court allowed © claim in guitar body shapes; separability would be an issue in the US unless the guitar also includes features like shark fins.

Under Armour defeats Lanham Act claim, must continue to fight tortious interference

Multiple Energy Technol., LLC v. Under Armour, Inc., 2026 WL 2016679, 2:20-CV-664 (W.D. Pa. Jul. 13, 2026)

Bioceramic powder is “a substance that is integrated into clothing, and, when worn, allegedly improves sleep and muscle recovery.” Plaintiff MET made a type of bioceramic powder called Redwave. It entered into “essentially an exclusive supply agreement” with Under Armour, which makes athletic apparel. But then Under Armour changed course and partnered with another producer, Hologenix, whose bioceramic powder was called Celliant. MET alleged that Under Armour falsely marketed its Celliant apparel as essentially being “FDA approved” when it wasn’t, encouraged some of its manufacturing partners not to deal with MET, and used confidential information from MET. Tortious interference is the only claim that survives; I’ll only discuss that and false advertising under the Lanham Act.  

After ending its relationship with MET, Under Armour encouraged Tom Brady’s company TB12 and two other textile manufacturing partners to use Celliant, not MET’s Redwave. Under Armour falsely promoted Celliant as “FDA approved” or “FDA designated” to TB12. Likewise, “Under Armour promoted on its website that the FDA determined that Celliant increases localized circulation, leading to faster recovery,” and issued a press release with similar language.

blurry image from opinion, sorry

The alleged statements to manufacturing partners weren’t actionable because there was no evidence that Under Armour made false statements about its FDA designation to two partners, “much less that these false statements affected the companies’ decisions to end their relationships with MET.” Instead, Under Armour required the use of Celliant (and not Redwave) as a condition of its potential partnerships; it was that pressure that mattered. Under Armour did make the FDA claims about Celliant to TB12, but there they didn’t qualify as “commercial advertising or promotion.”

Using the Gordon & Breach standard, the statements to TB12 were neither for the purpose of influencing consumers to buy the defendants goods or services nor “disseminated sufficiently to the relevant purchasing public to constitute ‘advertising’ or ‘promotion’ within that industry.” It was undisputed that TB12 wasn’t an Under Armour consumer, but instead a branding partner working with Under Armour to sell products, and preventing TB12 from partnering with MET “didn’t otherwise affect consumers’ decisions to buy Under Armour products.” Nor were the statements directed to the relevant purchasing public or sufficiently disseminated, since the relevant consumer market consisted of individuals interested in purchasing articles with bioceramics, not branding partners, so statements to TB12 weren’t “ ‘directed at the purchasing public.’ ”

[Why couldn’t manufacturers also be relevant consumers where the product is always a component of another product?] Even if TB12 counted, the market wasn’t made up of a “’relatively limited’ ” number of potential purchasers, so statements to TB12 alone would not be “ ‘disseminated sufficiently.’ ”

The website statements, by contrast, were actionable commercial advertising or promotion, but MET didn’t show that the press release had made a false FDA claim.

In a standing-limiting reading of Lexmark, the court found that MET lacked statutory standing.

First, MET couldn’t show a link between Under Armour’s website statements and lost royalties from Redwave-branded products. The court excluded MET’s expert report, without which there wasn’t enough evidence of proximate causation. Other economic and reputational harm was “simply too remote to establish a proximate causal link to the statements on the website.” There was no direct competition, reputational disparagement, or a “1:1 relationship” between the plaintiff’s sales lost due to the false advertising and the defendant’s gains. The most directly injured victims would be any other bioceramic apparel manufacturers, and then existing suppliers to Under Armour’s direct competitors, not MET.

MET also failed to provide enough evidence of materiality to get to a jury. “[N]othing in the record suggests that any specific consumers saw the false advertising on Under Armour’s website and changed their purchasing decisions because of it,” and MET’s survey was excluded (see below). Nor were the parties direct competitors, which can save some materiality arguments.

An antitrust claim failed because it was an antitrust claim. MET also failed to create a triable issue on its trade secret claim/breach of a non-disclosure agreement claim. But there was a triable issue on tortious interference with an existing contract with one manufacturer and with prospective business expectancies with two others. Although Under Armour’s discussions with them weren’t “commercial advertising or promotion,” but rather “business partners talking about what supplier to use” that didn’t make them necessarily justified or privileged.

Multiple Energy Technol., LLC v. Under Armour, Inc., 2026 WL 2015216, No. 2:20-CV-664 (W.D. Pa. Jul. 13, 2026)

The court excludes two of MET’s experts, one of whom focused on the antitrust claim, which I will ignore.

For the Lanham Act claim, MET’s proffered expert Dr. Maronick offered a supplemental survey and report.

His original survey and report concluded that “a statistically higher percentage of respondents seeing the Under Armour webpage claiming that products ‘Powered by Celliant have been determined by the FDA to increase localized circulation, leading to faster recovery’ believe th[at] Under Armour products will provide the increased circulation and faster recovery benefits claimed than do respondents who saw language that ‘the FDA had not made a determination whether products made with Celliant increase circulation and lead to faster recovery.’ ” It also concluded, “Under Armour’s claims about increased circulation and faster recovery are Important or Very Important in consumers’ decision to buy Under Armour products with either Celliant or Redwave fabric.”

In response to Under Armour’s deposition, Dr. Maronick “redid the survey,” creating the supplemental survey and report at issue. Specifically, (1) he matched the images and the font sizes and lengths of the “FDA determined/FDA has not determined” language shown to respondents across the test and control groups; and (2) rotated the positive and negative answers to both groups’ question on the “importance of the claims in consumers’ purchase decision.” The test cell showed Under Armour’s actual landing pages, and the control group was presented with landing pages that instead contained the statement, “The FDA has not made a determination about whether products powered by Celliant increase localized circulation leading to faster recovery.”

In his supplemental report, Dr. Maronick found that the “design differences” between the original and supplemental surveys “had no effect on consumers’ perceptions of Under Armour’s performance claims for its ‘Celliant-Powered’ products” and “played no role in assessing consumers’ reactions to claims made by Under Armour.”

The court initially found that the first survey’s flaws merited exclusion. It now evaluated the supplemental survey, and still found it too flawed to admit because it (a) used a fundamentally inadequate control; (b) materially distorted marketplace conditions; and (c) didn’t mitigate non-response bias or identify a target population.

The control was “problematic” because it “artificially [drew respondents’ attention] to the fine-print language at issue[,]” making it “impossible to determine [the cause of] any disparity between the respondents’ reactions to them.” First, the control survey focused its respondents on the fine-print text addressing the FDA claim at issue by framing that text in a white box, while no box surrounded the FDA-related text in the test group’s same question. Likewise, the control survey underlined the direction “[n]ote the statement about products ‘Powered by Celliant’ at the bottom of the statement,” “increasing the likelihood respondents would read the fine-print statement explaining that the FDA had not yet determined whether Celliant improved athlete recovery.” In the test cell, the same direction wasn’t underlined. The court found these differences to be “significant” and likely to skew the results.

one without a box, one with

Additionally, the test survey used improper negative language— stating that “[t]he FDA has not made a determination about whether products powered by Celliant increase localized circulation leading to faster recovery.” The Court previously found such language flawed because negative language creates different language than positive language. (I’m not sure why that is distorting—if the truth is that the FDA hasn’t evaluated the statement, how else can you say that? If negative language matters to consumers, then that in itself seems like evidence of materiality.) Alone this would go only to weight, but the variations amplified the negative language’s skewing effects, making the control improper.

The survey also “inappropriately distorted market conditions by flagging the variable about which the survey intended to ask respondents—the FDA claim—without having properly screened out respondents who otherwise would have ‘paid no attention’ to the fine-print claim.” [This is downstream of a larger debate about whether attention or comprehension, once perceived, is more important—from a regulatory perspective I favor comprehension, once perceived, because I don’t think there’s any social benefit in deceptive statements that might be overlooked, but reasonable minds can certainly differ.]

The survey “funneled” respondents by asking questions about magnified pictures of the fine-print language, and screened out respondents “[w]ere [not] able to see the images clearly.” But there was no filtering of respondents who, while physically able to see the webpages, didn’t notice or read the fine-print text, thus materially distorting marketplace conditions.

Finally, the supplemental survey failed to account for non-response bias and identify a target population.

breast pump rules: Think Green's trade dress claims against Medela proceed to trial

Think Green Ltd. v. Medela AG, 2025 WL 1826137, No. 21 CV 5445 (N.D. Ill. Jul. 2, 2025)

Think Green sued Medela for infringing its trade dress in its breast pump, as well as false advertising claims that were quickly dismissed on summary judgment because TG and Medela aren’t the only players in the market and the advertising didn’t disparage TG, meaning that TG couldn’t show harm. The trade dress claims, though they seem extremely weak to me (both in whether there would be confusion and whether the trade dress is nonfunctional), survive for a jury.

Think Green and Medela both sell manual, one-piece silicone breast pumps. Think Green’s pump is on the left and Medela’s pump is on the right within each image: 

Think Green claims trade dress rights in the “three-part shield-collector-base configuration, wherein the collector is bulbous yet fits underneath the outer diameter of the shield and contains an ornamental rib in its top part, while the base flares out from the bottom of the collector bulb to meet the surface upon which the whole unit rests.” Medela previously won summary judgment on Think Green’s design patent claim.

Much of the opinion focuses on the expert reports around secondary meaning and confusion. Think Green’s expert witness Harper opined that (1) among the relevant universe of consumers, there is a likelihood of confusion that Medela’s pump is sponsored or approved by Think Green due to the allegedly infringing trade dress; and (2) the relevant universe of consumers would likely ascribe secondary meaning to Think Green’s trade dress. Her surveys were not among the “rare” ones where fundamental flaws render them “completely unhelpful to the trier of fact and therefore inadmissible.”

Harper’s secondary meaning online survey presented half of the respondents with photos of Think Green’s trade dress, pictured immediately below, and asked those respondents whether they associated the pump with one particular company or brand, and if so, which brand and why. 

The control group saw a different pump and received the same questions. 

Harper’s point-of-sale confusion survey showed the test group an image of Medela’s pump, captured from the website buybuybaby.com, which included the “MEDELA” brand name; the product’s name, the Medela® Silicone Breast Milk Collector; its price; and an image of the pump that included a lanyard and yellow stopper. 

The control group was presented with a similar image of a different, “wearable” breast pump: 

The survey ultimately asked respondents whether or not “the manufacturer or brand of the breast milk pump/collector [they] just reviewed … is sponsored or approved by another manufacturer or brand.” If a respondent answered yes, they were asked who and why.

Medela’s criticism of the control was that it was wildly different from the trade dresses at issue. But, first, a secondary meaning survey might not even need a control. As for the confusion survey, it contained additional probes (why?), which the court thought helped (though the psychological literature suggests that people are really bad at answering that question), and anyway a bad control isn’t necessarily enough to kick out a survey. Choosing a control is really hard, so a bad control should bear on weight rather than admissibility.

Medela objected that the survey also didn’t reflect real-world conditions of online purchase, including images of the pump’s packaging, additional images of Medela’s “signature butter yellow,” a bullet point list of the pump’s features, and the ability to click on the pump’s image to expand it. Again, that went to weight; these were not core components of the shopping experience.

There was also a post-sale confusion survey. [Here one real issue is that unlike point of sale confusion, post-sale confusion’s theory requires that the post-sale confusion negatively affect subsequent consumers—either by causing them to think poorly about the plaintiff or by making the plaintiff’s product seem too easily available to the hoi polloi. There’s no evidence of this in the survey, and if you leave out that part of the theory you remove one of its few constraints.] Harper’s survey first screened for respondents who purchased a pump in the last year or who would consider purchasing one in the next year then showed them four images of Medela’s pump, collected from Medela’s online retail product pages, less Medela’s name, design elements, and measurements. The control group saw the same control pump. 

control images

test images

The survey asked respondents what company, companies, or brand(s) they believed put out the pictured pump, assuming they had an opinion in that regard; why they had that belief; and whether they believed the pump was affiliated with, or sponsored or approved by, any other company or brand. Although there were more similar noninfringing pumps out there, the control pump was also made of silicone, clear/white, and curved; whether the control was good was for cross-examination and the jury.

Plus, Medela objected that most people won’t encounter other people’s pumps in the wild, just like they mostly won’t encounter other people’s underwear, making post-sale confusion unlikely. The court was offended on behalf of those who pump in public, wash out/pour bottles in public, etc.

The survey also sufficiently isolated the claimed trade dress, even though it included the “distinctive ridge around the breast shield” visible in the test stimulus, which Think Green disclaimed as its trade dress. Again, this went to weight rather than admissibility.  

Think Green’s objections to Medela’s expert Cohen were equally unavailing. There’s a first time for every expert, so the fact that this was her first trade dress survey didn’t require exclusion. Medela’s survey showed Medela’s pump “as it had been displayed on the Amazon platform.” With the stimulus picture still on the screen, respondents were asked open-ended questions such as, “if you have an opinion, what company or organization makes or puts out this breast milk collector” and whether the respondent “believe[d] that this breast milk collector is affiliated with or sponsored by any other company.” The court didn’t buy that keeping the stimulus on screen created untoward demand effects. The court agreed that, because “[a] consumer would be staring directly at the product, together with its packaging and description, at the moment he/she made the decision to purchase (e.g., clicked ‘Add to Cart’),” the methodology underlying Cohen’s survey sufficiently mimicked marketplace conditions.

With that out of the way, there was enough evidence of secondary meaning to avoid summary judgment on that ground. Along with the survey, there was at least a minimal showing of Think Green’s volume and history of sales, even with no contextualization of those sales or of its advertising expenditures. As for intentional copying, “[c]opying is only evidence of secondary meaning if the defendant’s intent in copying is to confuse consumers and pass off his product as the plaintiff’s,” and whether that happened was for the jury.

And for confusion, “[i]t is a rare trade dress case where the ‘evidence is so one-sided that there can be no doubt about how the question [of whether likelihood of confusion exists] should be answered.’” [This is either trivial (a contested case generally means that both parties believe that they have a shot) or wrong (TM cases aren’t actually all that special in terms of evidence, much as we like to think TM is especially complicated). This claim contributes to making TM doctrine worse in practice by making courts reluctant to grant summary judgment—usually for the defendant—which has systematic effects on who gets to do what.]

Emails among Medela employees were not convincing evidence of passing off. “In at least some of the quoted emails, the employees discuss their desire to avoid copying Think Green’s pump and—crucially—never evince a desire to pass their pump off as Think Green’s,” the court concluded, citing phrases like “I do not believe this is a winning strategy,” “[W]e need to be more innovative,” and discussions of potential advantages of Medela’s pump over Think Green’s. “Although other email excerpts appear to support Think Green’s narrative, in asking the Court to interpret this entire body of emails as unequivocal proof of Medela’s intent to pass off Think Green’s trade dress as its own, Think Green is asking the Court to ‘choose between competing inferences.’”

On functionality, once again a lower court relies on the existence of alternative designs to deny summary judgment. The reason this is supposedly consistent with Traffix is that, if utility is shown by other means (in practice the only possibilities are patent claims, advertising that touts utilitarian advantages, or maybe effects on quality/cost), then alternative designs need not be considered, even though that’s not what Traffix says. Thus, the court says “[a] design does not need to be the only, or the best, way to do things to be considered functional: It only needs to represent one of many solutions to a problem,” then immediately proceeds to disregard that statement.

The court found that an international patent application involving the pump was not dispositive. Although the Federal Circuit held that “an applied-for utility patent that never issued has evidentiary significance for the statements and claims made in the patent application concerning the utilitarian advantages, just as an issued patent has evidentiary significance,” the court didn’t think that rules for the TTAB were relevant to “a court’s Lanham Act analysis.” [??? They are interpreting the same statutory language? Why would you ever want the PTO and judicial analyses of functionality to differ?] “A never-approved patent application … could include any number of useless features that would never have passed muster with the Patent and Trademark Office”; only approved or expired utility patents show functionality. [This seems wrong, including as a matter of applicant estoppel; it would also make evaluating the effect of foreign patents into a mini-trial on the relationship between foreign and US patent law.]

Anyway, even considering the application, it wouldn’t persuade the court that functionality was present. The court thought that there was no evidence that the “central advance” application matched the “essential feature” of the claimed trade dress. “And although the absence of a utility patent does not preclude the functionality of a design, the lack of a presumption of functionality does make it considerably less likely that functionality can be determined on summary judgment.”

Medela’s expert Faltum opines that the shape of Think Green’s design was ideally suited for the human hand, while other shapes are “awkward to hold, difficult to manipulate, inferior in performance,” and less effective at creating a vacuum seal: 

The court was skeptical; 3D pumps shouldn’t be likened to 2D shapes, and this analysis disregarded “the prominent breast shield topping the pumps.” Likewise, a jury should evaluate the claim that the bulb design was more effective, based on a study Faltum conducted that tested how much water various pumps could draw through a hose from a beaker when the bulb was compressed and released. The court agreed with Think Green’s argument that the study merely “demonstrate[d] that there is a statistically significant difference in the efficiency with which liquid is collected by one-piece silicone breast milk collectors as a consequence of their shape.” But it didn’t evaluate other potential utilitarian features of a pump (such as adherence or comfort), “whether this alleged difference in water-drawing efficiency translates into a more effective vacuum ‘essential for these collectors’ remains a factual question.” [I really don’t get this logic—something doesn’t have to be comprehensively better to be functional; tradeoffs are a thing, and each solution within the solution space should not be monopolizable. If you’d told me “this pump will be more efficient but be less comfortable,” I probably would have gone for speed. Perhaps there’s more to the story, but the explanation doesn’t persuade me.]

And Think Green sufficiently countered Medela’s claim that the suction base is ideal for stability “by pointing to several other similar pumps that use different types of non-suction bases to provide stability.” [Alternative designs, in one of the factors that’s supposed to determine whether alternative designs even need to be considered!]

Advertising didn’t unequivocally claim advantages to the design, only of the breast shield and suction base. (E.g., “Available in two larger capacities of 100ml and 150ml, and with a new suction base that sticks to flat surfaces to prevent accidental spills, it is the easiest and most simple way to express breast milk.”) That just meant that having a suction base is good, nothing about its size or shape or overall design. [This seems particularly anti-Traffix -- that suction base looks like the size and shape of a suction base you'd make if you wanted a suction base; a circle uses less material than anything else of the same size, and surely they shouldn't be forced to use a different shape or cover up the suction base.] “In short, none of these advertisements urge consumers to purchase the pump on the strength of any particular utilitarian quality of the suction base’s design.” [It would be interesting to see what consumers actually thought was being claimed.]

different breast sizes

suction base claim

suction base claim

The court pointed to Think Green’s evidence of “numerous successful, highly rated alternative silicone one-piece breast pumps that are designed differently than, but function the same as, Think Green’s pump.” “Where Think Green has presented evidence that other non-infringing designs are available, a triable issue of fact exists on functionality.” [So much for Traffix.]

ambiguity in consumer class actions v. the Lanham Act: convergence or divergence?

Slaten v. Christian Dior Perfumes, LLC, 2025 WL 1840026, No. 23-cv-00409-JSC (N.D. Cal. Jul. 3, 2025)

The concept of ambiguity is now on a path to become as entrenched in consumer protection cases as in Lanham Act cases. My thinking on this is still evolving, but right now I’m inclined to say that an “ambiguous” claim for purposes of Lanham Act claims means what an “ambiguous” claim for purposes of consumer protection litigation does, even though the assessment process is formally different. That is, a statement challenged under the Lanham Act is ambiguous if there are multiple plausible interpretations, some of which are not false. As the Ninth Circuit has clarified, a statement challenged under consumer protection law can be found plausibly deceptive if reasonable consumers could think it has a false factual meaning, even if other reasonable consumers would think it was ambiguous and required more information to interpret/had a different non-false meaning. That latter group can be expected to consult the back of the label for clarification, if present. But the first group of reasonable consumers has no reason to inquire further and therefore can be deceived; they are functionally equivalent to the deceived group in a Lanham Act false implication case.

The Lanham Act cases use the modifier “substantial” to describe the relevant subset (reasonable consumers who are deceived), but accepts much less than half as a sufficient percentage. Consumer protection cases are right now generally stricter, requiring bigger percentages where there are surveys (as there often are these days), so there may still be doctrinal divergence. I think that divergence, to the extent it exists, is likely unjustified—it is hard to see why competitors should have an easier path to remedies than the directly deceived consumers—but it is early days for both the “ambiguity” concept and the new prominence of surveys in such cases. 

I suspect that courts are thinking that “half or more” is better for class action treatment, but formally it really isn’t. That is, the common question in a consumer protection class action in the key states is “is this ad deceptive?” and the answer to that should be “yes” if it is likely to deceive a substantial number of reasonable consumers. Then, NY and California (etc.) presumptions about deception kick in to allow the class to proceed. An ad that deceives 49% of consumers—or 30%—about a material fact is actually pretty bad! [Caveat: our concept of deception should incorporate a “compared to what?” inquiry. If it’s impossible to provide the information in a non-deceptive way, but the information is also truthful and useful to some people, then we have to balance those considerations; if it’s not useful/the ambiguous meaning is just puffery, then we don’t have to worry so much.]

A related question is the role of the jury, at least in a Lanham Act case: If courts applied similar analysis in such cases, they'd ask whether a reasonable jury could find that a claim was literally false with respect to a substantial number of reasonable consumers, such that those consumers would feel no need to inquire further. That's not how Lanham Act courts tend to treat the issue; it would probably counsel against determining "ambiguity" as a matter of law. 

Anyway, this case involves a remand on claims over alleged deceptive labeling/advertising of SPF in cosmetics as lasting for 24 hours. The court initially interpreted McGinity v. Procter & Gamble Co., 69 F.4th 1093 (9th Cir. 2023), to mean that if a front label is ambiguous in that it “could mean any number of things,” some of which would not be deceptive, a court must look to the product’s back label to determine whether a reasonable consumer would be deceived. Upon review of the back label, the court dismissed plaintiff’s claims.

The court of appeals eventually remanded based on Whiteside v. Kinberly Clark Corp., 108 F.4th 771 (9th Cir. 2024). Whiteside held that “[a] front label is not ambiguous in a California false-advertising case merely because it is susceptible to more than one reasonable interpretation.” On a 12(b)(6) motion, a label “may have two possible meanings, so long as the plaintiff has plausibly alleged that a reasonable consumer would view the label as having one unambiguous (and deceptive) meaning.” That is:

a front label is not ambiguous simply because it is susceptible to two possible meanings; a front label is ambiguous when reasonable consumers would necessarily require more information before reasonably concluding that the label is making a particular representation. Only in these circumstances can the back label be considered at the dismissal stage.

Whiteside specifically rejected this court’s earlier “more than one possible meaning” standard for ambiguity. Bryan v. Del Monte Foods, Inc., 2024 WL 4866952 (9th Cir. Nov. 22, 2024), did not change matters. Bryan considered whether a front label describing a fruit cup using the phrase “fruit natural” falsely led consumers to believe all ingredients in the cups were natural. The court explained the word “naturals” was “a noun, not a descriptive adjective,” and so the label suggested “the phrase is just the name of the product.” Further, the front label context indicated “although the fruit itself is natural, the syrup may not be,” and customer surveys were insufficient because they “asked people what they thought ‘natural’ should mean on the label of a product, not what they thought it actually did mean as used on these labels.”

But here, reasonable consumers could conclude from the front label alone that defendant was advertising 24 hours of sunscreen protection.

Southern discomfort: class certified over malt beverage dressed like Southern Comfort whiskey

Andrews v. Sazerac Co., 2025 WL 1808797, No. 23-cv-1060 (AS) (S.D.N.Y. Jul. 1, 2025)

Plaintiffs alleged that Sazerac deceived consumers by selling a malt beverage that looks like Southern Comfort whiskey but in fact contains only “whiskey flavor.” The court certified a class of “[a]ll persons who purchased the Southern Comfort Malt Products in the State of New York at any time during the period February 8, 2020, to the date of judgment” with one named plaintiff.

The malt beverage comes in three sizes: 50ml, 100ml, and 355ml. The 50ml bottle is cylindrical, while the two larger sizes are relatively flat. But each has “colors, themes, fonts, symbols[,] and spacing” identical to Southern Comfort whiskey bottles. Each bottle has a statement of composition, which until April 2023 described the drink as a “malt beverage with natural whiskey flavors, caramel color and oak extract.” Inclusion of the “whiskey flavors” and “oak extract” language allegedly contributed to this misleading impression.

Addressing only the parts that interest me:  

Sazerac argued that there was no classwide proof that the bottles’ labeling was materially misleading. Although plaintiff’s survey found that 62.9% of consumers believed that the malt-beverage mini bottles contained whiskey, Sazerac argued that it was fatally flawed, and anyway only applied to the 50ml bottles.  The 50ml bottle is cylindrical, while the larger bottles have “relatively flat front[s],” and the statement of composition, which says that the drink contains “malt beverage,” appears in larger font on the bigger bottles. But the court didn’t find these differences to be material:

That the larger bottles are flat, instead of round, might be material if their shape would tend to indicate to reasonable consumers that the bottles contain malt beverage, not whiskey. But Sazerac doesn’t say that its whiskey is only sold in round bottles, so it’s not clear why the bottle shape makes a difference here. Sazerac’s observation that the statement of composition appears in larger font on the larger bottles seems similarly irrelevant. On the one hand, “malt beverage” is in larger font. But so too were the allegedly misleading “whiskey flavor” and “oak extract” phrases, at least until April 2023. Regardless, the Court sees no reason why the impression created by a specific combination of elements on a small bottle would vary “significantly” from the impression created by those same elements on a larger bottle.

Anyway, misleadingness was a merits question.

Plaintiffs offered a choice-based conjoint survey to measure their claimed damages, which estimated a 8.8% price premium from false beliefs that there was non-malt liquor in the beverage. Sazerac argued that this study failed to (1) show that the price premium is attributable to the beverage’s misleading packaging, as opposed to flavor and convenience, or (2) consider supply-side factors. The court disagreed. Plaintiffs’ theory was that the overall packaging contributed perceived value, and the survey tested that theory. Even if the study asked respondents to assume that the products were all available in the same store (and thus didn’t control for convenience), that was a matter of ultimate persuasiveness, not a matter of whether it tested the plaintiffs’ theory. “If [the] model missed the mark, then it did so in one fell swoop for the entire class.” Likewise, an alleged failure to measure supply-side factors can be accounted for “when (1) the prices used in the surveys underlying the analyses reflect the actual market prices that prevailed during the class period; and (2) the quantities used (or assumed) in the statistical calculations reflect the actual quantities of products sold during the class period,” as the survey here did. After all, “[a] conjoint survey that asks respondents whether they would rather pay x for a product labeled ‘100% Fruit Juice’ or y for a similar product labeled ‘50% Fruit Juice’ ... would account for supply-side factors if both x and y reflect the prices for which juice companies actually sell similarly labeled products in the marketplace.” Sazerac argued that it would refuse to sell its malt beverage at the lower price of a generic competitor. But again, that was not relevant to whether the survey was good enough for class certification. “Moreover, it would be improper to give Sazerac the benefit of the doubt—and to take its CEO’s self-interested statements as controlling—at this stage.”

 

Scotts loses trade dress claim over green & gold for Miracle-Gro

Scotts Co. v. Procter & Gamble Co., 2025 WL 1779167, No. 2:24-cv-4199 (S.D. Ohio Jun. 27, 2025)

A different Scotts trade dress claim than the one I blogged last year. While it’s hard to get rid of trademark claims on a motion to dismiss, a preliminary injunction may be a different matter—as it is here, where the court does a thorough job with an expansive trade dress claim (which frankly should have John Deere’s lawyers taking notice, given its own reliance on green and yellow). This might be a good case to give students, given its accessibility.

Scotts makes the Miracle-Gro line of plant food and lawn and garden products, some of which are depicted below: 

P&G recently introduced a new non-selective herbicide—a weed killer—called “Spruce”:

Scotts is the market leader in the lawn and garden business. What is its Miracle-Go trade dress? It has an incontestable registration that “consists of a rectangular shaped box in the colors green and yellow” for “plant food.” When the appropriate colors are transposed onto the lined image in the registration, the mark looks something like this:

But Scotts claimed more, alleging a trade dress comprising:

(1) A green and yellow color combination;

(2) With each color presented as a separate horizontal band and the top color taking up a smaller ratio than the bottom color;

(3) With the two bands sharing a common border that runs horizontally along the package;

(4) With a straight line dividing the two colored bands; and

(5) A circular horizontally centered graphic element.

The court referred to the “rectangular shaped box” combination as the Registration, to avoid confusion with this broader trade dress claim—broader because it lacked a shape restriction and applied to more than plant food. Nonetheless, Scotts has never used the Miracle-Gro Trade Dress with any herbicide, nor did it plan to. Also, the broader trade dress did have some greater specificity—specifically, the “circular horizontally centered graphic element.” As implemented, this element was the Miracle-Gro logo or wordmark, which consists of “white text overlaid on (and extending beyond the horizontal border of) a black circle with some additional graphic sheen.”

Products using this broader trade dress have been on the market from 15 to 70 years, depending on the product. Plant food was the original, sold for over 70 years, and was most closely associated with the registration.

Since 2014, Scotts has sold around 104 million units of this product for approximately $650 million. During the same period, Scotts sold roughly one billion units of Miracle-Gro, generating approximately $5.6 billion in revenue, although just under one-third came from so-called “specialty products” or “flavors,” “which come in quite different packaging (although sometimes with at least some of the design elements from the Trade Dress).” 

Ninety percent of sales occur at brick-and-mortar stores, including “do-it-yourself ... home centers” like “Lowe’s, Home Depot, Menards”; large chain retailers like Target, Walmart, and Meijer; and hardware, garden, and club stores. The cost ranges from $6 to $20 depending on the product and configuration.

Spruce became available to consumers mid-November 2024. It costs between $12.99 to $39.99 depending on the configuration. Spruce is carried in brick-and-mortar retailers such as Home Depot, Lowe’s, Walmart, Target, Ace, and True Value, as well as online. P&G has invested significantly in television, online/social media, print, and in-store advertising as part of the product rollout:

The bottom of each container consists of a clear or transparent section. The transparent portion is designed to allow consumers to see the liquid product. A spruce green portion predominates most of the product packaging. “The Miracle-Gro green is a brighter green with a glossy finish that resembles a freshly cut lawn on a sunny day, while the green on the Spruce packaging has a matte finish and is darker, more like a pine (or spruce) tree in a shadowy forest.” On most packages, a round, yellow dandelion image (with an even darker green background) traverses the clear and dark green portions, intended to depict a half-living, half-dehydrated-and-dying, dandelion. The Spruce trademark appears in bold white text, with a yellow “violator” containing the text “Visible Results in 1 HOUR.” I learned: “A graphic violator is a visual element used in product design that sellers use to draw the consumer’s attention to certain messaging the seller wants to emphasize.”

Many third-party lawncare products similarly use green and yellow color combinations: 

Although market presence for all of these wasn’t shown, Scotts admitted that Preen Weed Preventer Plus Plant Food product (leftmost) is “widely sold in the lawn-and-garden marketplace” (perhaps outselling the Miracle-Gro weed preventer product) and at times “shelved right next to” that Miracle-Gro product. That is true also of Spectracide (center), which Scotts admitted is “a leading weed killer product.”

Plaintiff’s witness Sass had worked for Scotts for over 20 years. He testified about the 12 distributor declarations and 110 consumer declarations submitted to the PTO for the Registration. The declarants each said something like: “when I see packaging which is green on top and yellow on the bottom in connection with plant food products, I interpret the packaging design as an indication that the goods come from a single source, i.e., the makers of Miracle-Gro.”

Testifying about differences from other products on the market, Sass emphasized the importance of the proportions (“typically one-third green on top, two-thirds yellow on the bottom”) and a dark circle element for the Scott products. The court concluded that the proportions were “perhaps more important” than Scott argued.

Meanwhile, P&G’s witness Croswell testified that P&G settled on Spruce’s dark green because P&G believed it would make Spruce distinctive in the weed-killer market and because it invoked the namesake of the brand (i.e., Spruce trees). “[D]uring development, P&G and one of the third-party marketing companies it used identified concerns about whether a certain version of the Spruce design may have been too similar to a particular competitive product. But at no point during that process did anyone raise a concern that any version of the proposed Spruce design was too similar to the Miracle-Gro line of products.” And P&G has no plans to expand the Spruce brand into other product categories in the lawn and garden space.

Nobody was aware of instances of actual confusion.

Winning my heart, the court began its confusion analysis by cautioning that it would not allow Scotts to extend the benefits of incontestability to the common-law trade dress, and that incontestability and likely confusion are two different questions.

Strength of the mark: Miracle-Gro’s trade dress likely acquired distinctiveness through secondary meaning, even though its PTO declarations were only directed to a rectangular box and it had no survey evidence. Length of time on the market, advertising, sales volume, and market leadership favored secondary meaning nonetheless.

The trade dress was also probably nonfunctional.

Without evidence of actual confusion, “it basically comes down to the Court’s assessment of the objective likelihood of confusion based on the products and packaging, along with the evidentiary value of the competing consumer surveys the parties tendered.”

Given that strength of the plaintiff’s mark and similarity of the marks are the most important, Scotts lost primarily because of dissimilarity.

Miracle-Gro’s trade dress had substantial commercial strength, but its conceptual strength was unclear, especially given the definitional questions (are proportions key to the trade dress, or not?). The court noted that, on all the products it saw, the one-third/two-thirds division was the same, and the green was above the yellow. With that, plus the “circular horizontally centered” black circle at the dividing line between the colors, there was likely some conceptual strength.

“But when you start subtracting individual elements from that combination, the distinctiveness quickly vanishes.” There was nothing particularly distinct about using green and yellow for packaging in the lawn care industry: they “are the colors of sunshine and plants.” Although the burden is on the defendant to show what actually happens in the market, P&G did so, showing that several other strong market performers use green and yellow. It’s not that those others are confusing—it’s that reasonable consumers wouldn’t just rely on seeing green and yellow to attribute source given the market.

The dissimilar Miracle-Gro variants also sapped some of the conceptual strength of the trade dress. “[T]he more consumers come into contact with Miracle-Gro products with a different style of packaging, and in particular different color combinations, the less likely they are to look for the green and yellow combination as identifying their favorite lawn and garden product.” (But the black circle abides.)

Nonetheless, this factor overall tilted towards Scotts.

Relatedness of the goods: One of the products bearing the Scotts trade dress is a “Weed Preventer.”

That didn’t move the needle much (herbicide is not “weed preventer” but killer, and you’d use the weed preventer on a flowerbed but not the weed killer, and vice versa for weeds sprouting between bricks), but the products were somewhat related insofar as they are all in the lawn and garden category.

Similarity of the marks: a “defendant’s resounding success on this factor makes the plaintiff’s burden of prevailing on the seven other Frisch’s factors effectively insurmountable.” Similarity doesn’t depend on a side-by-side, element-by-element comparison; it is based on the overall impression arising from the combination of elements. Even going element by claimed element, there was substantial dissimilarity.

Color: Very distinct shades of green, and Spruce was matte (and transparent in part) while Miracle-Gro was glossy and entirely opaque.

Separate horizontal bands of color with top smaller: Scotts has the one-third/two-third ratio, and Spruce uses a clear, bottom portion (about one-fifth), then dark green predominates over most of the rest. The yellow portion, it is relatively small and is used to highlight a message—“Visible Results in 1 HOUR.”

True, on both packages, the colors “shar[e] a common border that runs horizontally along the package” in the form of “a straight line dividing the two colored bands.” “But these visual elements are wholly unremarkable and add little to the overall visual impression of each product.”

Likewise, both products contain a “horizontally centered graphic element.” But on one, it’s the Miracle-Gro logo, which is white text on a black circle with some additional features. Spruce, has a circular yellow dandelion (with different graphics on each half) overlaid on a dark green background. Moreover, the circular graphics are “in different places on the package ([top] v. bottom).” “The dissimilarity on this element could not be more stark.”

There were other dissimilarities as well, including in the actual containers—with five Spruce configurations versus the entire Miracle-Gro product line, “none of them even remotely resemble each other in shape.” Scotts didn’t have text in the top portion; P&G did. The graphics were “meaningfully” different: photorealistic images of vegetation versus graphic design-like elements (e.g., an outlined paw print). And the Spruce trademark creates its own distinct visual impression, serving as a house mark.

The trade dresses at issue are “clearly distinguishable and would appear so to all but the most obtuse consumer.”

Scotts tried to change this result with survey evidence. Its expert, Dr. Wind, conducted a Squirt survey—one that presents survey respondents with both of the conflicting marks and “do[ ] not assume that the respondent is familiar with the senior mark.” Potential purchasers of Miracle-Gro and Spruce were broken into three groups, Home Depot, Lowe’s, or Meijer, each with a test and control cell. After telling respondents to imagine they were considering purchasing a lawn and garden product, the survey showed respondents in each group in-store displays from the stores to which they were assigned (except the Lowe’s, which was mistakenly shown Home Depot; the court found this rendered the survey “suspect and deserving of little weight” as to this subgroup). E.g., Home Depot respondents saw these: 

Then test respondents were shown some of the same photos containing Spruce, with red lines surrounding the Spruce products, and asked how they would describe those products to a friend.

Finally, test respondents were shown the in-store display that included Miracle-Gro products along with various other third-party products (the right-most photo in the initial photo array above) and were asked: “Do you believe that any of these products or product lines on this plant food display were made by the same company that manufacturers the products you saw that were circled in red?” Respondents were asked some follow-up questions (e.g., the reason they selected the products).

The survey repeated the process for (1) asking whether respondents thought any of the products or product lines in the display with the Miracle-Gro “ha[d] a business affiliation or connection with the company that manufactures the products you saw that were circled in red” [I note that there was no training on what a “business affiliation” is, and there probably should be]; and (2) asking whether respondents thought any of the products or product lines in the display with the Miracle-Gro “gave permission or approval to the company that manufactures the products you saw that were circled in red.”

Control groups saw the same images and stimuli, except the colors on the Spruce products were black, white, and silver.

If a respondent who answered positively mentioned green and yellow in connection with Spruce, the coders tagged that respondent as “confused.” Dr. Wind calculated net confusion rates, “[d]ue to explicit reference to the green and yellow packaging” of 16.2% for the Lowe’s subgroup, 9.1% for Home Depot, and 17.7% for Meijer.

P&G objected to (1) the Squirt survey format; (2) the design; and (3) what Wind counted as “confusion.”

Squirt: P&G argued that Miracle-Gro and Spruce do not appear side-by-side in the marketplace and that an Eveready survey is the appropriate tool to use where one of the marks at issue (here Miracle-Gro) is a strong mark. The court agreed with this criticism. “The products at issue are typically not displayed side-by-side in a retail setting, nor was there a sufficient showing that the typical consumer sees them sequentially,” and Miracle-Gro is commercially strong. The court quoted McCarthy to the effect that “Squirt methodology is inappropriate unless there are ‘a significant number of real world situations in which both marks are likely to be seen in the marketplace sequentially or side-by-side.’”

Design: P&G argued that Squirt surveys have an inherently leading nature (seems true), which was amplified by stimuli unreflective of true market conditions. This was even more problematic than choosing Squirt in the first place. First, there was the Lowe’s error. Second, in the Home Depot image, nearly half of the “plant food” display shown to respondents was dominated by a pallet of Miracle-Gro potting mix. “[T]he Court finds it unlikely that large pallets of Miracle-Gro potting mix typically sit directly in front of Home Depot’s plant food shelves (or at least, that customers typically would stand behind such a pallet while selecting something on the plant food shelf). Simply put, the Home Depot photo was highly suggestive.”

Identification of confused respondents: Dr. Wind “classified any respondent ‘confused’ for simply describing the products as ‘green and yellow’—even if they mentioned nothing about Scotts or Miracle-Gro.” That is, if a respondent accurately noted that the packaging for Spruce products contained the colors green and yellow, that would be coded as reflecting “confusion.” This the court found most troublesome of all. “P&G identified a significant number of responses that clearly should not have been coded in that manner—namely, respondents who referenced “Spruce” in their answers, and who did not mention “Miracle-Gro” or “Scotts” at all, but who happened to mention that the Spruce bottle was green and yellow (which it is).”

P&G offered its own survey by Dr. Simonson: an “aided Eveready survey.” An Eveready format assumes that survey respondents “are aware of the [senior] mark from their prior experience.” This “format is especially useful when the senior mark is readily recognized by buyers in the relevant universe.” Respondents are shown the allegedly infringing products,, then asked:

• Who do you think makes or puts out this product?

• Does the company that makes this product put out any other products?

• Does the company that makes this product have a business affiliation or connection with any other company? [Again, no definition/training.]

• Did the company that makes this product receive permission or approval from another company?

However, Simonson used the typical Eveready questions, but displayed multiple products from the marketplace (as would occur in a Squirt survey), instead of the single, allegedly infringing product. Each respondent viewed a picture array of products, like so:

They were asked to review all the products “as they would if they were considering purchasing a weed preventer at an online store.” The respondents then saw one of the four images below, with the Spruce product (or a control version of the Spruce product, bottom) blown up on the left-hand side: 

They were then asked variations of the four standard Eveready questions along with follow-up probing questions as necessary. The “control” “had a different trade dress, but still incorporated green and yellow elements as well as the language and small icons used on Spruce.” Simonson found that “only 2.9% of the Test group respondents … mentioned either Miracle-Gro or Scotts.” Although the court didn’t rely on the Simonson survey, it didn’t like the control.

Here, the characteristic being assessed was the color combination. But instead of altering solely Spruce’s color, as Dr. Wind did for his control, Dr. Simonson created an entirely new shape, maintained the colors green and yellow (but making white the most prominent color), and added a circular graphic element to the top portion of the packaging. In many ways, Dr. Simsonson crafted a control that was more similar to the Miracle-Gro’s Trade Dress than Spruce’s current packaging, which may explain why the control group displayed greater confusion than the test group.

(The court  did reject Scotts’ criticism that the answers “Ortho,” “RoundUp,” “fertilizer,” or the like should have been coded as confused. “This case is about Miracle-Gro; not every brand Scotts uses. And Scotts certainly does not have a monopoly on the word ‘fertilizer.’”)

Remaining factors: Marketing channels favored Scotts; degree of purchaser care was not very significant/it was dependent on mark similarity. Intent: (1) P&G considered other packaging designs with other color schemes; (2) some third-party reports prepared for P&G, as part of Spruce’s packaging development process, featured images of Miracle-Gro products; (3) Scotts sent P&G a letter expressing concerns about confusing similarity between the products’ designs in May 2024. This was “attenuated at best” intent evidence. “It seems natural to the Court that a product development team might consider different colors and designs, then test those options before going to market.” Nor was a study’s inclusion of “a few images of Miracle-Gro products (along with many, many other lawn and garden care products)” evidence of intentional copying. “Scotts is the category leader; you would expect some of its products to appear in any report about the market.” Finally, the Scotts letter had no bearing on intent—the packaging design was nearly finalized by then. No weight.

Likely product line expansion: Not likely; no weight.

Dilution: In the Sixth Circuit, “[t]he ‘similarity’ test for dilution claims is more stringent than in the infringement milieu.” Given the high level of dissimilarity here, that was fatal.