by Dennis Crouch
In a previous post, I examined the important issue of comparison prior art that emerged from the dispute between Columbia Sportswear and Seirus. This post will focus on another key issue from the case – the relevance of logos in design patent infringement analysis.
The Federal Circuit’s pair of decisions provide guidance on how logos factor into the design patent infringement inquiry, and begin to tease-out differences in policy concerns underlying design patent law versus trademark law. As the court made clear, while logos are often key for avoiding consumer confusion about product source in the trademark sense, the absence of source confusion does not necessarily preclude a finding of design patent infringement. Still, ornamental logos found on the accused product can still be relevant as visual distractors in the process of evaluating similarities and differences between the claimed design and accused design.
Design Patent Infringement vs. Trademark Infringement
The standards for proving design patent infringement and trademark infringement differ significantly regarding the relevance of consumer confusion about product source.
For trademark infringement under the Lanham Act, likelihood of consumer confusion is a key requirement. Moseley v. V Secret Catalogue, Inc., 537 U.S. 418 (2003). The touchstone is whether ordinary consumers in the relevant market are likely to be confused about the source of the goods. Logos and other source identifiers play an important role in this analysis. A logo prominently displayed on the accused goods that is clearly distinct from the plaintiff’s mark can weigh heavily against finding a likelihood of confusion. See, e.g., Groeneveld Transp. Efficiency, Inc. v. Lubecore Int’l, Inc., 730 F.3d 494 (6th Cir. 2013) (“[The accused infringer] has in fact scrupulously avoided such confusion by choosing a starkly different logo that it prominently displays on its [products] and on all its sales and marketing literature.”).
For design patent infringement, however, likelihood of consumer confusion is not directly relevant. See Unette Corp. v. Unit Pack Co., 785 F.2d 1026 (Fed. Cir. 1986) (“Likelihood of confusion as to the source of the goods is not a necessary or appropriate factor.”). The Federal Circuit has made clear that the sole test is whether the claimed and accused designs appear substantially similar to an ordinary observer, not whether consumers will be confused about source. See Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665(Fed. Cir. 2008) (en banc) (“[I]n accordance with Gorham and subsequent decisions, we hold that the ‘ordinary observer’ test should be the sole test for determining whether a design patent has been infringed.”).
In essence, while both analyses invoke an “ordinary consumer”, the design patent test focuses solely on visual similarity absent any real-world marketplace considerations and asks whether the design has been replicated. The trademark test deeply analyzes real-world conditions and actual consumer viewpoints on source confusion. This leads to divergent treatment of logos and potentially divergent results.
On their face, both the design patent and trademark infringement tests invoke principles of confusion. The ordinary observer test asks whether the resemblance between designs would “deceive” an observer, “inducing him to purchase one supposing it to be the other.” Gorham Co. v. White, 81 U.S. 511 (1871); see also 35 U.S.C. § 289 (“colorable imitation”). And the likelihood of confusion test directly asks whether consumers would likely be confused about the source of goods. The overt references to “deception,” “imitation” and “confusion” create risk that the separate doctrines could be conflated or misapplied. This underscores the need for precision in explaining that design patent infringement does not consider real-world consumer perspectives, while consumer viewpoints are paramount in trademark analysis. The linguistic parallels make delineating these distinctions even more crucial.
While the Federal Circuit has characterized design patents and trademarks as distinct doctrines, in practice they substantially overlap in protecting product designs that serve as source identifiers. Certain product design features may function simultaneously as ornamental, novel designs eligible for design patent protection and as source-identifying trade dress. For example, a unique shape for a handbag or sneaker may qualify for a design patent based on its ornamentality while also indicating source as trade dress.
What that means is that design patents are often used to protect brand identifiers as commercial source signifiers. But, are able to do so without needing to satisfy the requirements of trade dress or its more substantial fair use and non-functionality doctrines. In the case of the wavy reflective inner lining — that clearly has trademark meaning in my mind (as a purchaser of winter gloves). And, it is that marketplace meaning that likely serves as the true basis of the long running lawsuit. So design patents will sometimes grant backdoor trademark-like protection, encompassing design aspects that convey source identity and marketplace meaning. Ultimately, while the doctrines have distinct requirements on paper, in practice their protections substantially overlap for certain product designs functioning as visual source identifiers.
Columbia Sportswear v. Seirus
This distinction emerged as a key issue in the dispute between Columbia Sportswear and Seirus. Columbia accused Seirus of infringing its design patent covering a wavy pattern design for use on heat reflective material. Seirus argued that differences between its design and Columbia’s patented design – especially Seirus’s prominent and repeated logo – should weigh against finding infringement.
On summary judgment, the district court disregarded Seirus’s logo entirely based on L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117 (Fed. Cir. 1993), and found infringement. In Columbia I, the Federal Circuit held this was error:
Our precedent does not prohibit the fact finder from considering an ornamental logo, its placement, and its appearance as one among other potential differences between a patented design and an accused one.
Columbia Sportswear N. Am., Inc. v. Seirus Innovative Accessories, Inc., 942 F.3d 1119 (Fed. Cir. 2019).
The appellate court explained that, while labeling a copy does not avoid design patent infringement, logos still factor into the holistic comparison of the claimed and accused design appearances. Id. The court thus vacated summary judgment and remanded.
On remand, the district court instructed the jury on the ordinary observer test for design patent infringement. But, the district court declined Columbia’s request to instruct that consumer confusion was irrelevant and that likelihood of confusion need not be found. The jury returned a verdict of non-infringement.
In Columbia II, the Federal Circuit affirmed the jury instructions given were legally correct. Columbia Sportswear N. Am., Inc. v. Seirus Innovative Accessories, Inc., No. 2021-2299 (Fed. Cir. Sept. 15, 2023). The court sympathized with concerns that the ordinary observer test’s language referencing deception risked conflation with concepts of consumer confusion. But it found no reversible instructional error, reiterating that logos remain relevant to the overall similarity analysis:
[J]ust because consumers might not be confused about an accused product’s source, that alone would not preclude an ordinary observer from deeming the claimed and accused designs similar enough to constitute design-patent infringement.
Id. (emphasis in original; citing Braun Inc. v. Dynamics Corp. of Am., 975 F.2d 815 (Fed. Cir. 1992)).
These decisions offer important insight and guidance for attorneys in the trenches, and also highlight a need to reconcile the doctrines to ensure that they are serving important societal goals.