Thursday, January 05, 2012

Fresh step misstep

Church & Dwight Co., Inc. v. Clorox Co., --- F.Supp.2d ----, 2012 WL 8466 (S.D.N.Y.)

Taking time out from slamming the SEC and the financial industry, Judge Rakoff granted a preliminary injunction against Clorox’s “smart cats” ad, finding that Clorox’s tests were insufficiently reliable to support its claims.  C&D makes cat litter prominently featuring baking soda.  Clorox makes Fresh Step, which uses carbon instead.

In Fall 2010, Clorox began airing the predecessor ads, showing cats choosing litter boxes with Fresh Step over C&D’s Super Scoop.  A voiceover explained that “cats like boxes ... with Fresh Step litter inside ... because Fresh Step's scoopable litter with carbon is better at eliminating odors than Arm & Hammer.”  Last January, Clorox began showing “clever” cats. The voiceover said: “Cats are smart. They can outsmart their humans. Their canines. Unlock doors. They're also smart enough to choose the litter with less odors.”

C&D’s test found that of 158 cats, only 6 rejected a Super Scoop litter box, while 8 rejected a Fresh Step box.  C&D thus sued Clorox, claiming literal falsity.  Clorox discontinued the ads and C&D dismissed the complaint last February.

Clorox then began airing a new “clever” ad (versions here and here at least for now, along with a variant here): “We get cats. They're smart. They can outsmart their humans. And their canines.” Then a cat is seen entering a litter box and pawing through the litter as the voiceover continues, “That's why they deserve the smartest choice in litter.”  The ad then shows two lab beakers.  The Fresh Step beaker has a black substance labeled “carbon,” and the other beaker has a white substance labeled “baking soda.”  Though it’s not labeled with a brand, C&D’s is the only major cat litter brand that uses baking soda.  The ad then shows green gas floating through the beakers.  Voiceover: “So we make Fresh Step scoopable litter with carbon, which is more effective at absorbing odors than baking soda.”  The green gas in the Fresh Step beaker then rapidly evaporates while the gas level in the baking soda beaker barely changes.  The small super says that Clorox's claims are “[b]ased on [a] sensory lab test.”

C&D argued literal falsity based on the insufficiency of the lab test.  Clorox’s test was done in-house.  Clorox prepared separate containers of: (i) fresh cat feces covered with carbon; (ii) fresh cat urine covered with carbon; (iii) fresh cat feces covered with baking soda; (iv) fresh cat urine covered with baking soda; (v) uncovered feces; and (vi) uncovered urine. (Fun job!)  The sealed containers sat for 22-26 hours and then they were placed in sensory testing booths, one each for each type of sample.  Eleven panelists rated the samples on a 0-15 scale, and this was done four times.  The results: carbon reduced odor from 2.72 to 0 while baking soda reduced odor only from 2.72 to 1.85 (a 32% decrease, represented precisely in the ad).

Clorox trained the panelists to use a specific scale, which Clorox alleged had been subject to peer review and used at many universities.  Panelists were trained with identical smells at different levels of intensity in order to develop a common metric for pungency and with different litter products, both with and without cat excrement, in order to learn to discriminate between cat malodor and other odors present in litter. Panelists were trained to give a rating of zero when they could not detect a certain olfactory stimuli, which happened here whenever cat excrement was treated with carbon.

C&D argued that the test didn’t support the broad claim that Fresh Step outperforms baking soda products in eliminating odor.  It also argued that the findings of zero odor 44 times were so suspicious as to render the test unreliable. The court agreed.

The necessary implication was that Fresh Step better eliminated odor than C&D litters that use baking soda.  Clorox argued that one reasonable interpretation of the ad was a general comparison of the odor-reducing properties of the two substances, but the court rightly dismissed that: “Clorox has not identified any basis for believing that any consumer who pays attention to its commercial reasonably cares about how effectively carbon works compared with baking soda outside the context of cat litter and competing litter products.”

With that out of the way, Clorox’s test couldn’t support the necessary implication of litter superiority.  “In actual practice, however, cats do not seal their waste, and smells offend as much during the first twenty-two hours as they do afterwards. Thus, the Jar Test's unrealistic conditions say little, if anything, about how carbon performs in cat litter in circumstances highly relevant to a reasonable consumer.”  Moreover, substantiation of the claims in the ad would require proof not just that carbon eliminates odors in open cat litter, but also that it outperforms baking soda in that task and that baking soda eliminates only 32% of those odors.  “Given that the Jar Test says little about how substances perform in litter as opposed to jars, it cannot possibly support Clorox's very specific claims with regard to litter.”  Thus, the ad was literally false by necessary implication.  (In my opinion, the epicycles of Lanham Act falsity parsing are now doing more harm than good; necessary implication is a useful concept to defeat obviously ridiculous strained readings of ads by defendants trying to force plaintiffs to do surveys, but it would be a lot better for the doctrine if courts could just up and say: the clear message of the ad is that carbon cat litter outperforms baking soda litter; there’s no point in making the claim in an ad for cat litter otherwise; end of story.)

The court separately agreed with C&D’s criticism of the methodology.  The panelists’ uniformity in finding zero malodor in carbon-treated litter was highly implausible, “and more likely reflects flaws in their in-house training or objectivity than any reliable result.”  C&D submitted evidence that humans are “noisy instruments,” and, “for neurological reasons, perceive the exact same thing differently at different times and report the presence of olfactory stimuli even where they do not exist…. [G]iven the variation even among the same person's reports at different times, C & D argues that eleven different people almost certainly would not uniformly report the same experience forty-four times.” 

Clorox’s own studies supported this contention.  In an internal report, Clorox’s panelists gave an average score of greater than zero to a box of litter with no cat excrement in it.  In an earlier iteration of the jar test, 18% of trials resulted in a report of some malodor in jars of excrement treated with carbon.  Clorox’s expert also agreed that humans are “noisy.” 

Clorox responded that none of this precluded a uniform rating of zero if there was actually no malodor.  The court found this unpersuasive.  “[I]t is highly implausible that eleven panelists would stick their noses in jars of excrement and report forty-four independent times that they smelled nothing unpleasant.”  If the average false alarm rate was 14%, a number taken from published work, the probability of Clorox’s results was 0.13%.  (Hi, statistics!  Nice to see you!)  Thus, the results of the test weren’t sufficiently reliable to conclude with reasonable certainty that they established the proposition for which they were cited.

Having shown likely success on the merits, C&D needed to show irreparable harm.  Recently, courts have been presuming irreparable harm where a plaintiff shows likely success on the claim that a comparative ad is literally false as long as, given the nature of the market, the audience would obviously receive the message that the plaintiff was being targeted even if not identified by name.  Clorox argued that eBay disapproved the use of presumptions in the context of establishing irreparable harm, but the court noted that the Second Circuit used this language in the Time Warner case after eBay.  Anyway, “legal presumptions aside,” C&D proved the likelihood of irreparable harm.  C&D provided evidence that consumers shopping for cat litter overwhelmingly identify baking soda with C&D’s products, and the ad shared themes with Clorox’s former ads recalling those ads’ explicit references to C&D.  This was enough to persuade the court that consumers would undoubtedly understand that C&D was being criticized.

As the court summarized: “Put simply, Clorox, cloaking itself in the authority of ‘a lab test,’ made literally false claims going to the heart of one of the main reasons for purchasing cat litter. In such circumstances, where the misrepresentation is so plainly material on its face, no detailed study of consumer reactions is necessary to conclude inferentially that Clorox is likely to divert customers from C & D's products to its own unless the offending commercial is enjoined. Thus, C & D has successfully shown a high likelihood of irreparable harm.

Clorox was enjoined from airing the commercial.  There’s a question here on the standard to be applied were C&D to seek to suppress the copies to which I linked above assuming they’re posted by people with no relationship to Clorox and thus not commercial speech; I think C&D’s showing of falsity sufficient to trigger the Lanham Act is plainly not enough to force a takedown on disparagement/defamation grounds, but I don’t know if anyone would want to fight.  One obvious move would be to have Clorox transfer the copyright as part of a settlement and C&D could then send takedown notices based on copyright instead of on falsity, but that’s a DMCA 512(f) claim waiting to happen, especially if someone (e.g., me) wanted to illustrate the concept of falsity by necessary implication by using this as an example.

No comments: