Pom Wonderful LLC v. Hubbard, No. 14-55253 (9th Cir. Dec. 30, 2014)
The Ninth Circuit is the source of some of IP’s highest highs and its lowest lows. The reader is invited to consider where this decision fits. The district court found that Pom was unlikely to succeed on the merits of its trademark infringement claim against defendant (d/b/a Pur) for Pur’s “pŏm” pomegranate-flavored energy drink. The court of appeals found an abuse of discretion and therefore reversed and remanded, leaving Pom only with the Herb Reed hurdle. (Query: what effect will the delay caused by the appeal have on Pom’s burden of showing irreparable harm under Herb Reed? Even though the delay wasn’t Pom’s fault, it would arguably provide an opportunity to develop evidence of irreparable harm that is often unavailable in a preliminary injunction context.)
Pom owns a number of POM registrations, used with various goods including pomegranate juice beverages. POM also occasionally licenses the use of POM marks to other companies. Before 2002, when Pom first began using the POM marks for beverages, no one in the industry used “pom.” Since then, Pom has sold more than 190 million bottles of pomegranate juice. It’s the leading seller 100% pomegranate juice in US supermarkets, with sales over $60 million. It spent $24 million in 2011 and 2012 on promoting its juices (not clear whether this is yearly or aggregate).
Pom discovered that Pur was selling a pomegranate-flavored energy drink, “pŏm,” sued, and moved for a preliminary injunction, which was denied. The court of appeals declined to exercise de novo review. “Although legal standards certainly inform a district court’s application of the Sleekcraft factors, Ninth Circuit precedent requires us to review the district court’s Sleekcraft-factor findings for clear error.” Nonetheless, clear error there was.
Pom showed an ownership interest in the standard character mark POM for fruit juices. This established “Pom Wonderful’s exclusive right to use the mark in connection with fruit juices.” And, because it was a standard character mark, it covered all design variations. (Citing only Federal Circuit/registration precedents.) “Therefore, Pom Wonderful’s exclusive right to use its ‘POM’ standard character mark is extremely broad, covering the word in all types of depictions.”
OK, this sounds like a troubling mingling of the registration/infringement inquiries, but then the court takes it back by saying that ownership is only part one of the inquiry; the rest is confusion.
Sleekcraft is a “fluid” test, and a plaintiff doesn’t need to satisfy every factor if there are strong showings on some. It’s always the totality of facts that matters. Here, though the district court correctly weighed mark strength, relatedness of goods, and consumer care in favor of Pom, it incorrectly weighed mark similarity, marketing channels, actual confusion, intent, and product expansion against Pom. Absent that error, likely confusion was clear.
The district court was right to weigh mark strength in Pom’s favor. The mark was merely suggestive, because “pom” “is not ascribed independent pomegranate-related meaning by conventional dictionaries” and “requires customers to use some additional imagination and perception to decipher the nature of Pom Wonderful’s goods.” (Aaaaaaargh. That’s not what “suggestive” is supposed to mean. “Quik” for a print shop doesn’t tell you the nature of the shop’s goods. The question is, once you know the mark and the goods, is imagination required to understand their relationship? Here, the answer is plainly “no.”) Suggestive marks are “presumptively weak” (which at least undoes the damage caused by the initial mistake). However, Pom showed substantial market strength, making it strong.
Juice beverages were also related to energy drinks. “Both beverages are pomegranate-based or pomegranate-flavored, single-serve, and marketed for their healthful properties.” And consumer care was likely to be low, given that the beverages at issue cost between $1.99 and $2.49.
The district court clearly erred in weighing the similarity of the marks against Pom. This factor is always important because when “marks are entirely dissimilar, there is no likelihood of confusion,” while “as the similarities between two marks increase, so too does the likelihood of confusion.” Basic rules: evaluate similarity by sight, sound, and meaning; evaluate similarity as it appears in the marketplace; and similarities weigh more heavily than differences.
Pom argued that the trade dress should be ignored in assessing mark similarity. The court of appeals disagreed. Marks aren’t considered in the abstract (as they are in the registration context) but rather as they’re encountered in the marketplace. Moreover, although Pom’s mark is registered in standard character form without design elements, the court considered the Pom design’s heart-shaped “O” in evaluating the visual similarities because “the only images of Pom Wonderful’s products included in the preliminary injunction record [show that] the ‘O’ in ‘POM’ is in the shape of a heart.” (Again, correcting much of the potential mischief from the holding on what it is that Pom owns, as long as other courts notice this.)
The marks’ many visual similarities were obvious:
Most significantly, each mark is comprised of the same three letters. These three letters are presented in the same order, with a stylized second letter (i.e., the “o” in “POM” is heart-shaped, and the “o” in “pŏm” has a breve over it). In addition, the letters in both marks are uniformly cased (i.e., they are either all uppercase, or all lowercase) and presented in a simplistic, white front that is offset by a dark maroon background.
There were also some visual dissimilarities in terms of prominence on the package, font, size, and capitalization, and the middle letter was stylized differently in each:
But sound and meaning were identical. Semantic similarity was particularly noteworthy because meaning can itself substantiate similarity. Moreover, nonvisual similarities were particularly important where, as here, there was no evidence that consumers encountered the marks side by side in the marketplace. If you asked a friend to buy you a ‘POM’ drink, unless your friend already knew your preferences, she could easily return with Pur’s pŏm.
Considering the “many visual similarities, perfect aural similarity, and perfect semantic similarity more heavily than the marks’ visual dissimilarities,” similarity weighed “heavily” in Pom’s favor. Moreover, a lesser degree of similarity is required when the accusing mark is strong. This made the district court’s holding clear error.
There was also clear error in the district court’s finding that marketing channel convergence weighed against Pom. Both companies used parallel market channels: supermarkets across the country. “In addition to selling their products in at least one overlapping state (Texas), both companies sell their products in an overlapping supermarket chain (Albertson’s).” It wasn’t clear whether they were sold in the same Albertson’s stores, “a channel of trade is not limited to identical stores or agents.” Plus, the products were highly similar. “Although Pom Wonderful advertises its products using a much broader range of outlets than Pur uses, the similarities between the products suggest an overlapping general class of consumers.” The district court mistakenly required Pom to prove that its drinks were sold in the very same brick and mortar stores as Pur’s. “Although such proof would surely increase the likelihood of consumer confusion, the absence of identical channels does not, by itself, undermine Pom Wonderful’s likelihood of proving that the marketing channels converge.”
As for the absence of actual confusion evidence, lack of evidence of bad intent, and lack of evidence of Pom’s intent to expand product lines, the district court erroneously weighed these against Pom rather than treating them as neutral, as it properly should have done in this case. Difficulty in gathering evidence of actual confusion makes its absence generally unsurprising, especially at the preliminary injunction stage. Nor is bad intent or product expansion required; Pom’s failure to present evidence on these “neither undermines nor advances its ability to prove likelihood of confusion.”
Five factors weighed in favor of Pom and none weighed in favor of Pur. Sleekcraft isn’t a counting exercise, though. In this case, the district court clearly erred by (1) mistakenly weighing differences in the marks more than similarities, (2) requiring Pom to prove too much in terms of convergent marketing channels, and (3) weighing three neutral factors against Pom. Thus, the district court’s likelihood-of-confusion holding left the court of appeals with the “definite and firm conviction that a mistake has been committed.”
On remand, the district court could allow more evidence on the Herb Reed factors.