A utility patent is a wall of text. A design patent is a set of pictures. That single difference—claims drawn rather than written—changes almost everything about how these cases are tried, and it is the reason design patent litigation rewards lawyers who think like art critics as much as engineers. The question in a design case is not "does the accused product perform the claimed function?" It is closer to "would an ordinary buyer, looking at these two things, be deceived into thinking they are the same design?" That is a question about appearance, deception, and the human eye, and the law has spent more than 150 years learning to answer it with rigor.

What makes the field worth your attention is the mismatch at its heart. A design patent is cheap to obtain, fast to issue, and famously narrow—it covers one appearance and little more. Yet within that narrow band it can be the single most punishing weapon in the entire patent system, because one design patent on one component can, by statute, capture an infringer's total profit on an entire product. Narrow scope, enormous leverage. Hold that paradox in mind; it runs through every section below.

This guide is a practitioner's tour of that body of law, organized the way a court moves through a case: what a design patent is and what makes it valid, how a court construes a claim built of figures, how a fact finder proves or defeats infringement under the famous ordinary-observer test, and then the part that makes general counsel sit up straight—damages. We close with willfulness, marking, injunctions, the International Trade Commission, and the strategic choices that separate a winning case from an expensive one. Along the way you will meet a chalk holder shaped like a pencil, a flotation device with armbands, a Chevrolet fender, a "Y"-patterned chair that turned out to be a basket, and a famous rounded-corner smartphone. By the end you will understand not just the rules but why they exist and where they remain genuinely unsettled.

If you want the broader patent-litigation backdrop—Markman hearings, summary judgment, trial mechanics—see our comprehensive guide to patent infringement litigation. If you are still deciding whether a design patent is the right tool at all—or whether you should pair it with a utility patent or trade dress—start with the intersection of design and utility patents and design patents vs. trade dress protection for product configurations. This article assumes you have chosen design patents and now have to fight about them.

What a Design Patent Is (and the Statute That Makes It Hard)

A design patent protects "any new, original and ornamental design for an article of manufacture." That phrase comes straight from 35 U.S.C. § 171, and every word is a potential battleground. "Ornamental" means the design must be about appearance, not function. "Article of manufacture" means the design must be applied to a tangible thing—a chair, a phone, a graphical user interface displayed on a screen. "New" and "original" import the novelty and non-obviousness requirements (35 U.S.C. §§ 102 and 103) that govern all patents.

Section 171(b) does something quietly important: it provides that the provisions relating to patents for inventions "shall apply to patents for designs, except as otherwise provided." So a design patent borrows the entire utility-patent apparatus—infringement under § 271, the validity defenses, the presumption of validity under § 282, marking under § 287, injunctions under § 283—unless a special rule displaces it. The displacements matter enormously. The two biggest are term and damages. A design patent runs fifteen years from grant (35 U.S.C. § 173, as amended by the Patent Law Treaties Implementation Act for applications filed on or after May 13, 2015), with no maintenance fees—shorter than a utility patent's twenty-years-from-filing term but cheaper to keep alive. And § 289 gives design patentees a remedy no utility patentee enjoys: the infringer's total profit, which we will spend real time on below.

Another displacement is the claim itself. A design patent contains exactly one claim, and it almost always reads, in full: "The ornamental design for [the article], as shown and described." The "as shown" does the work. The drawings are the claim. Solid lines show what is claimed; broken (dashed) lines show environment that is disclaimed.

That single-picture claim is why Judge Giles Rich—one of the great patent jurists of the twentieth century and a principal architect of the 1952 Patent Act—observed that "[d]esign patents have almost no scope. The claim at bar, as in all design cases, is limited to what is shown in the application drawings." In re Mann, 861 F.2d 1581, 1582 (Fed. Cir. 1988). He meant it as a description of the genre, not an insult. A design patent is narrow by nature; an accused product that looks meaningfully different escapes. The flip side, again, is the leverage.

One more structural point that surprises newcomers: a design must be tied to an article of manufacture. The law has never recognized a design "in the abstract"—surface ornamentation floating free of any product. The Federal Circuit drove this home in Curver Luxembourg, SARL v. Home Expressions Inc., 938 F.3d 1334 (Fed. Cir. 2019), discussed below, holding that the claim language can confine a design patent's scope to a particular article even when the figures themselves show only the pattern. We will return to Curver because it is the doctrinal twin of Lanard, the pencil-shaped chalk-holder case.

A quick vocabulary primer for non-lawyers. "Prior art" means everything publicly known before the patent's filing—earlier products, patents, and publications. "Claim construction" (sometimes a Markman ruling) is the court's job of deciding what the patent covers before a jury compares it to the accused product. "Anticipation" means a single piece of prior art already showed the claimed design, destroying its novelty. "Obviousness" means the design, while not identically shown before, would have been an obvious variation to a designer of ordinary skill. Keep those four terms handy; the rest of this guide turns on them.

Validity: The Ways to Kill a Design Patent

In litigation, the accused infringer almost always counterattacks the patent's validity, and it must do so by clear and convincing evidence to overcome the § 282 presumption. There are several principal lines of attack, and a smart defendant runs all of them. The three workhorses are functionality, anticipation, and—newly muscular after 2024—obviousness. A fourth, less common but worth knowing, is the indefiniteness/non-enablement attack on the drawings themselves.

Functionality: Is the Design Really About How It Looks?

Section 171 protects ornamental designs. If the appearance of the article is "dictated by" its function—if the thing looks the way it does only because that is the way it has to work—the design is unprotectable and the patent invalid. The classic teaching example is a key blade: its profile is dictated by the lock it must open, so its shape cannot be a protectable ornamental design. The Federal Circuit applied that logic in Richardson v. Stanley Works, Inc., 597 F.3d 1288, 1293–94 (Fed. Cir. 2010), where a multi-function carpentry tool's elements were dictated by use.

The doctrine is subtler than that crisp example suggests, and design litigation lives in the subtlety. The Federal Circuit does not ask whether individual elements have a function; almost everything does. It asks whether the design as a whole is primarily functional rather than primarily ornamental, often by asking whether alternative designs could perform the same function. If a competitor could have built an equally good widget that looked different, the chosen appearance was a design choice, not a functional inevitability, and it is protectable. Courts weigh a cluster of factors—often called the Berry Sterling factors after Berry Sterling Corp. v. Pescor Plastics, Inc., 122 F.3d 1452 (Fed. Cir. 1997)—including whether the protected design represents the best design, whether alternative designs would adversely affect utility, whether there are concurrent utility patents, whether advertising touts particular features as functional, and whether the design's elements are dictated by use.

Two principles cut in the patentee's favor. First, "[t]he presence of alternative ways to achieve the article's function, although not dispositive, increases the likelihood that a design is primarily ornamental," and second, crucially, "[t]he aesthetic appeal of a design to consumers does not render a design functional." Automotive Body Parts Ass'n v. Ford Global Technologies, LLC, 930 F.3d 1314 (Fed. Cir. 2019). Automotive Body Parts is itself instructive: an aftermarket trade group argued that Ford's design patents on replacement hoods and headlamps were functional because consumers want a repaired car to match the original. The Federal Circuit rejected the argument squarely—the desirability of a matching part is an aesthetic consideration, not a functional one, and so cannot defeat ornamentality. (That holding is one of the flashpoints in the broader right to repair debate, where design patents on collision parts are a central battlefield.)

Two more cases show how functionality plays out in real disputes, and they pull in opposite directions in a way that is instructive. In Sport Dimension, Inc. v. Coleman Co., 820 F.3d 1316 (Fed. Cir. 2016), the patent claimed the design of a personal flotation device with armbands and a tapered torso. The district court, finding those features functional, simply deleted them from the claim during construction. The Federal Circuit reversed: even functional features contribute to the overall ornamental appearance, and a court cannot surgically remove them from a design claim. The right move is to recognize that an abundance of functional features makes the overall claim scope narrow, not nonexistent. As the court put it, "the armbands and side torso tapering serve a functional purpose, so the fact finder should not focus on the particular designs of these elements when determining infringement, but rather focus on what these elements contribute to the design's overall ornamentation. Because of the design's many functional elements and its minimal ornamentation, the overall claim scope of the claim is accordingly narrow."

The companion lesson came a year earlier in Ethicon Endo-Surgery, Inc. v. Covidien, Inc., 796 F.3d 1312 (Fed. Cir. 2015), involving the contoured handle and knobs of ultrasonic surgical shears. The district court had construed the design patents to "cover nothing"—reading out every ornamental aspect as functional. The Federal Circuit vacated, explaining that although the underlying trigger, torque knob, and activation button are functional articles, "there is no evidence in the record that any of the ornamental designs adorning those underlying articles are essential to the use of the article." The patents therefore had some scope—"the particular ornamental designs of those underlying elements." The unifying principle: a court may narrow a design claim because of functionality, but it may not erase it.

There is one more functionality wrinkle worth knowing, because it surprises trademark lawyers. The "aesthetic functionality" doctrine—familiar from trademark and trade dress law, where a feature can be "functional" if protecting it would put competitors at a non-reputation-related disadvantage—does not apply to design patents. In Shure Inc. v. ClearOne, Inc., 2020 WL 6074233 (D. Del. 2020), the accused infringer argued that the perforated front face of a ceiling-tile array microphone was functional because it let the device blend in and let sound through. The court rejected the argument: making a product "unobtrusive" is about appearance, not utility, and the fact that a design is attractive or blends in does not make it functional under patent law. That distinction—aesthetic functionality bars trademark protection but is irrelevant to design patents—is one of the cleanest lines between the two regimes, and it is why a single product look can sometimes be protected by a design patent even when trade dress would fail. We unpack that contrast at length in design patents vs. trade dress protection for product configurations and the deeper evolution of design patent functionality.

Anticipation: Did Someone Already Show This Design?

A design patent is anticipated—and therefore invalid under § 102—if a single prior-art reference shows the same design to the eye of an ordinary observer. The test deliberately mirrors the infringement test: if an ordinary observer would think the prior art and the patented design are substantially the same, the patent is anticipated. The Federal Circuit confirmed this symmetry in International Seaway Trading Corp. v. Walgreens Corp., 589 F.3d 1233 (Fed. Cir. 2009) (the foam clog case), holding that the ordinary-observer test is the sole test for anticipation just as it is for infringement. That symmetry is a strategic gift: the very prior art a defendant gathers to argue non-infringement can double as anticipation ammunition, and vice versa.

This rule has a famous trap built into it: views matter. The smartphone wars supply the canonical example. In Apple, Inc. v. Samsung Electronics Co., 678 F.3d 1314 (Fed. Cir. 2012), the district court found Apple's D'087 design patent likely anticipated by a prior-art tablet—but it had compared only the front views. The Federal Circuit held that was error. The D'087 patent claimed a partial side view showing a bezel extending from the front to the sides of the device, a feature the prior art lacked. By ignoring the side view, the district court compared an incomplete picture and reached an unsustainable anticipation finding. The lesson is concrete and tactical: a design patent is the sum of all its figures. If you are attacking validity, your prior art must match every claimed view; if you are defending, you scour the figures for a distinctive perspective—a bevel, a contour, a partial view—that no single reference captures.

Obviousness After LKQ: The Ground Just Moved

For more than forty years, design patents enjoyed a relatively patent-friendly obviousness test. Under In re Rosen, 673 F.2d 388 (CCPA 1982), and Durling v. Spectrum Furniture Co., 101 F.3d 100 (Fed. Cir. 1996), a challenger first had to find a single "primary reference"—a prior design "basically the same" as the patented design—before any combination of references could even be considered. If no such near-twin existed, the obviousness inquiry was over and the patent survived. That rigidity made design patents hard to invalidate for obviousness and made obviousness rejections rare in examination.

That world ended on May 21, 2024. Sitting en banc in LKQ Corp. v. GM Global Technology Operations LLC, 102 F.4th 1280 (Fed. Cir. 2024), the Federal Circuit overruled the Rosen-Durling test as "improperly rigid" and inconsistent with the Supreme Court's flexible obviousness command in KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007). The case arose, fittingly, over the ornamental design of a Chevrolet Equinox front fender. In place of Rosen-Durling, the court installed the same four-factor Graham v. John Deere Co., 383 U.S. 1 (1966), framework that has long governed utility patents: (1) the scope and content of the prior art; (2) the differences between the prior art and the claim; (3) the level of ordinary skill—here, the perspective of "an ordinary designer in the field of the article of manufacture"; and (4) secondary considerations such as commercial success, industry praise, and copying.

What changed for litigators is the threshold. A challenger no longer needs a primary reference that is "basically the same" as the claimed design; it needs the "closest prior art"—"the prior art design that is most visually similar to the claimed design." But the change is not a free-for-all. The Federal Circuit built in a sliding scale: "the more different the overall appearances of the primary reference versus the secondary reference(s), the more work a patent challenger will likely need to do to establish a motivation to alter the primary prior art design ... without the aid of hindsight." Early post-LKQ practice confirms that primary references still matter. In Dynamite Marketing, Inc. v. The WowLine, Inc., Nos. 2024-1523, 2024-1525 (Fed. Cir. Sept. 12, 2025) (non-precedential), the Federal Circuit affirmed a finding of non-obviousness in part because the challenger failed to establish a primary reference at all and offered only "a general hindsight-based conclusion" rather than record-supported reasons to combine. The PTAB, applying LKQ in cases like Next Step Group, Inc. v. Deckers Outdoor Corp., IPR2024-00525 (P.T.A.B. Aug. 6, 2024), has likewise declined to institute several early post-LKQ design IPRs.

The practical upshot is that obviousness has gone from a near-dead defense to a live one, and discovery now reaches motivation-to-combine evidence, designer testimony, and secondary-considerations proof that Rosen-Durling made almost irrelevant. Plaintiffs should expect more invalidity contentions and should build their secondary-considerations record—commercial success, copying, praise—early. Defendants should retain a design expert who can speak as "an ordinary designer in the field" and articulate, without hindsight, why the claimed look was an obvious adaptation. Because this is the single fastest-moving area in design law, we devote an entire companion article to it: the LKQ decision: a seismic shift in design patent obviousness analysis. For the underlying § 103 mechanics it now borrows from utility practice, see overcoming obviousness rejections: a comprehensive guide to Section 103 analysis.

A Fourth Front: Indefiniteness and Enablement of the Drawings

Because a design claim is its drawings, sloppy figures can sink a patent under §§ 112 and 171. A design claim must be definite—its scope clear with reasonable certainty—and enabling—the specification must let "an average designer" make the design. In practice these collapse into one inquiry: do the figures, read together, convey a single, consistent design? In In re Maatita, 900 F.3d 1369, 1375–76 (Fed. Cir. 2018), the Federal Circuit held a single-plan-view shoe-sole design was not indefinite even though it left the depth dimension open, because the design (like a rug or a fabric pattern) could be understood in two dimensions. But inconsistent or internally contradictory figures—a feature that appears in one view and vanishes in another, lines that cannot be reconciled across perspectives—are a recognized invalidity hook. The defense lesson: before you fight about infringement, lay the patent's figures side by side and look for contradictions a draftsman missed years ago.

Claim Construction: Reading a Claim Made of Pictures

Before a jury compares the patent to the accused product, the judge usually must construe the claim. In a utility case, claim construction means deciding what disputed words mean. In a design case there are essentially no words to construe—so what is there to do?

The Federal Circuit answered that in its landmark design opinion, Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc), which we meet again under infringement. Egyptian Goddess cautioned trial courts against translating a design into a detailed verbal description, because words inevitably distort a visual claim—emphasizing some features, flattening others, and inviting the jury to focus on elements in isolation rather than on the design as a whole. The preferred approach is for the court to let the drawings speak and supply only such guidance as is needed: identifying the role of the broken lines (disclaimed environment), pointing out which features are functional and therefore contribute only to the overall impression, and resolving genuine disputes about what the figures depict. Construction in a design case is less translation than curation.

That said, words on the face of the patent can still matter. In Curver Luxembourg, SARL v. Home Expressions Inc., 938 F.3d 1334 (Fed. Cir. 2019)—an issue of first impression—the patent claimed an overlapping "Y" pattern. Its title, claim, and figure descriptions all recited a "pattern for a chair," but the figures depicted only the pattern, with no chair anywhere in sight. (During prosecution the examiner had rejected the vaguer original title under 37 C.F.R. § 1.153 and MPEP § 1503, so the applicant amended in the word "chair" to get the patent issued.) When Curver sued a company selling baskets bearing a similar pattern, the Federal Circuit held the patent limited to chairs and affirmed dismissal: a design patent has never protected surface ornamentation "in the abstract," and where the figures fail to tie the pattern to an article, the claim language that does so controls. Curver is a double-edged sword in litigation—it can rescue a patentee whose figures are ambiguous, or it can shrink an over-broad patent to nothing when the accused product is simply a different article.

Three other disputes recur often enough to deserve names.

First, what do the lines mean? Drawing conventions carry legal weight. Solid lines are claimed; broken lines are not. But parties sometimes fight over whether a line represents a three-dimensional contour (a curving surface) or a two-dimensional ornamental feature (a seam, a groove, a painted stripe). In simplehuman, LLC v. iTouchless Housewares & Products, Inc., 2020 WL 7342739 (N.D. Cal. 2020), the parties disputed whether certain lines on a trash-can design were shading or seams. The court studied the figures, noticed the lines appeared in views (like a top-down view) where shading would be impossible, and concluded they were ornamental seams that were part of the claimed design. The case is a reminder that careless patent drawings create litigation fights years later, and that good design-patent prosecution is itself a form of risk management. See how to prepare an invention disclosure for your patent attorney for the upstream side of this.

Second, should the court verbalize a special feature? Defendants love to extract a single feature and ask the court to write it into the claim, because a narrowed, feature-focused claim is easier to design around. Egyptian Goddess counsels against it. In Catalyst Lifestyle Ltd. v. Elago Co., involving waterproof cases for AirPods, the defendant asked the court to construe the claim to require a "two-piece" design because two of the ten figures showed an exploded, two-piece view. The court refused, warning that singling out one feature risks the fact finder "improperly concentrating on these features in isolation rather than considering the design as a whole." The design-as-a-whole principle is not a slogan; it is an operational rule that decides motions.

Third, how do you handle functionality during construction? As Sport Dimension and Ethicon taught above, the court may instruct the jury to treat heavily functional elements as contributing only to the overall impression—but it may not delete them. Some courts, like Lifted Limited, LLC v. Novelty Inc., 2020 WL 2747814 (D. Colo. 2020), refine this by excluding only those aspects that are "solely" functional (there, the precise shape and size of an oval hole dictated entirely by the component it had to fit) while keeping everything that contributes to ornamentation. The same opinion declined to read invalidity-for-functionality into claim construction and rejected the argument that white space in the drawings limited the design to smooth surfaces—a useful reminder that drawing conventions, not litigant intuition, govern.

Infringement: The Ordinary-Observer Test

Now the heart of the case. How does a fact finder decide whether the accused product infringes?

The answer is the ordinary-observer test, and its pedigree runs to the Supreme Court's 1871 decision in Gorham Co. v. White, 81 U.S. (14 Wall.) 511 (1871)—a dispute over the ornamental design of silverware handles. There the Court asked whether, "in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same"—such that the resemblance deceives the observer, "inducing him to purchase one supposing it to be the other." Id. at 528. Infringement turns on deception of an ordinary buyer, not on a side-by-side checklist of features parsed by an expert. Gorham also rejected the defendant's invitation to judge the designs through the eyes of an expert who could catalog tiny differences—the ordinary observer is the retail purchaser, not the connoisseur. That choice of vantage point is doing quiet work in every modern case.

For decades after Gorham, the Federal Circuit layered a second requirement on top: the "point of novelty" test, which required the patentee to prove that the accused design appropriated the specific feature that distinguished the patented design from the prior art. In practice it devolved into expert-driven, element-by-element fights about which feature was "the" point of novelty, and it pulled fact finders away from the holistic comparison Gorham demanded. In Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc)—a case about, of all things, the ornamental design of a nail buffer—the Federal Circuit abolished the point-of-novelty test as a separate requirement. The ordinary-observer test, the court held, is the sole test for design patent infringement.

But Egyptian Goddess did something cleverer than merely simplifying. It folded the concern behind the old test—that minor features distinguishing a crowded field should matter—into the ordinary-observer analysis itself, through prior art. The ordinary observer, the court explained, is "deemed to view the differences between the patented design and the accused product in the context of the prior art." When the patented and accused designs are close, the fact finder compares both against the prior art; differences that "might not be noticeable in the abstract can become significant to the hypothetical ordinary observer who is conversant with the prior art." In a crowded field of similar designs, an ordinary observer pays attention to small differences and a patent's scope is correspondingly narrow. In an empty field the same observer is more easily deceived and scope is broader.

This "three-way visual comparison"—patented design, accused design, and the prior art—is the engine of modern design infringement. Several cases show it running.

In Arc'Teryx Equipment Inc. v. Westcomb Outerwear Inc., a dispute over a curvilinear jacket zipper (U.S. Design Patent No. D513,715), the district court did exactly what Egyptian Goddess prescribes. It construed the claim from the illustrations rather than words; defined the ordinary observer as a discerning outdoor-apparel buyer (a nice example of the observer being tailored to the actual market); identified several visual differences between the patented and accused zipper sections; and placed those differences in the context of the prior art, finding that the accused zipper was closer to a prior-art reference than to the patent. Because an ordinary observer familiar with the prior art would not be deceived, the court granted summary judgment of non-infringement. The prior-art comparison did the decisive work.

Lanard Toys Ltd. v. Dolgencorp LLC, 958 F.3d 1337 (Fed. Cir. 2020)—the chalk holder shaped like a No. 2 pencil—shows the same machinery and a recurring temptation. The Federal Circuit affirmed summary judgment of no infringement because so many overall concepts of the "pencil" design already existed in the prior art that the patent's protectable scope shrank to its specific proportions and ornamentation, and the accused product's proportions differed. The court held this approach did not secretly revive the point-of-novelty test; comparing the designs in light of the prior art is precisely what Egyptian Goddess requires. Lanard also rejected a clever argument worth flagging, and it dovetails with Curver: Lanard wanted the written description ("a chalk holder") to enrich its claim, arguing it had cleverly applied a pencil's ornamentation to a different article. The court said no—a design patent protects visual appearance, and merely applying an existing visual design to a new article (changing the function but not the look) is not a patentable design invention. The function of the article does not expand the scope of the design.

Two practical points fall out of this. First, when the accused and patented designs are "plainly dissimilar," the fact finder need not consult the prior art at all; the comparison ends. Egyptian Goddess and Ethicon both recognize this shortcut—in Ethicon, the accused shears had a linear handle versus the patent's contoured one, "plainly dissimilar," and summary judgment of non-infringement followed. Second, the prior art does double duty: it narrows scope at the infringement stage and supplies the ammunition for anticipation and obviousness at the validity stage. A defendant's prior-art search is the most valuable hour in a design case, full stop. For the broader principle of how courts handle the line between claimed and accused subject matter, compare our overview of what constitutes patent infringement.

A note on terminology that trips people up: design patents have no separate "doctrine of equivalents" analysis the way utility patents do. The ordinary-observer test already asks about substantial similarity, so it absorbs the function the doctrine of equivalents serves elsewhere. There is one visual question, asked once. Infringement is a question of fact, which is why so many design fights are won or lost on summary judgment, when the figures and the prior art let a court say the designs are plainly alike—or plainly not.

Damages: The Section 289 "Total Profit" Bombshell

Here is where design patents stop being a quiet corner of IP and become a boardroom topic. For utility patents, damages under 35 U.S.C. § 284 are "in no event less than a reasonable royalty"—you license the patented feature and award its value, with apportionment to isolate the patented contribution. Design patents have all of that available too. But they also have something extra and extraordinary: 35 U.S.C. § 289.

Section 289 says that whoever applies a patented design "to any article of manufacture for the purpose of sale" shall be liable to the patent owner "to the extent of his total profit." Read that again. Total profit. Not the profit attributable to the design; the infringer's entire profit on the article of manufacture, computed without apportionment—no Georgia-Pacific royalty analysis, no "value of the patented feature," just the infringer's books, the cost accounting, and the total profit on the relevant article. See Nike, Inc. v. Wal-Mart Stores, Inc., 138 F.3d 1437, 1442 (Fed. Cir. 1998) (total profits "based on the entire article of manufacture"). The statute even provides that the patentee "shall not twice recover the profit made from the infringement," which tells you Congress expected this remedy to be large. There is also a statutory floor of $250—a relic of the provision's nineteenth-century origins, when Congress enacted it in 1887 specifically to overrule Supreme Court decisions (the Dobson line) that had made design-infringement remedies effectively worthless by demanding the patentee prove how much profit was attributable to the design alone.

Why does this exist? Because for many products, the design is the product. People buy a particular chair, lamp, watch, or sneaker because of how it looks. Congress decided that an infringer who steals the design that drives the sale should disgorge the profit on the sale, not just a notional royalty. It is a disgorgement remedy of unusual breadth, unique to design patents.

Samsung v. Apple: What Is "the Article of Manufacture"?

For most of § 289's life, "total profit on the article of manufacture" was simple because the article was the whole product. A patented design for a carpet meant total profit on the carpet. Then smartphones arrived, and the question got hard.

In the smartphone wars, a jury found that several Samsung phones infringed Apple design patents covering, among other things, a black rectangular front face with rounded corners (the D'677 patent) and a colorful grid of icons on the home screen (the D'305 patent). Applying § 289 as the Federal Circuit then understood it, the jury awarded Apple roughly $399 million—Samsung's entire profit on the infringing phones. Samsung's argument was intuitive: the patents covered the shape of the front face and the icon layout, not the processor, battery, antenna, camera, software, or the dozens of other components that make a phone valuable. Why should infringing a design for the glass face cost Samsung the profit on the entire device?

The Supreme Court agreed to look, and in Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2016), it unanimously rejected the Federal Circuit's rule that the "article of manufacture" must always be the end product sold to consumers. Justice Sotomayor, writing for the Court, parsed § 289's two operative terms. An "article of manufacture," she explained, is simply "a thing made by hand or machine"—broad enough to encompass both a product sold to a consumer and a component of that product. Nothing in § 289 limits "article of manufacture" to the finished product. So when a design is applied to one component of a multicomponent product, the relevant "article of manufacture"—the base against which total profit is calculated—may be that component rather than the whole product. Id. at 60. The Court pointedly declined to set out a test for how to identify the relevant article, leaving that to the lower courts, and remanded.

The remand is as important as the decision, and it is where the doctrine actually crystallized. On remand the district court adopted a four-factor test (drawn from the Solicitor General's amicus brief) for identifying the "article of manufacture" to which the design has been applied: (1) the scope of the design claimed in the patent, including the drawings and written description; (2) the relative prominence of the design within the product as a whole; (3) whether the design is conceptually distinct from the product as a whole; and (4) the physical relationship between the patented design and the rest of the product—for example, whether the component can be physically separated from the product. A retrial followed in 2018. With the article-of-manufacture question now in play, a jury again awarded Apple a large sum—reportedly about $539 million—after which the companies settled, ending nearly seven years of litigation.

For litigators, Samsung v. Apple added a discrete, high-stakes phase to every multicomponent design case. The plaintiff argues the article of manufacture is the whole product (maximizing the profit base); the defendant argues it is a narrow component (minimizing it). The four factors are now litigated with expert accounting testimony, product-teardown evidence, and consumer-survey data about what drives purchasing. The choice of base can swing a verdict by an order of magnitude. And the burden allocation matters: courts have generally placed an initial burden on the patentee to identify the article of manufacture and prove the infringer's total profit on it, with the burden of production then shifting to the infringer to come forward with evidence of an alternative, narrower article and its deductible costs. A design defendant that fails to put on component-level accounting evidence may be stuck with the whole-product base by default—an expensive silence.

Putting It Together: A Worked Example (Hypothetical)

Suppose Acme Furniture Co. holds a design patent on the distinctive sculpted shell of an office chair—the curved seat-and-back silhouette—claimed in solid lines, with the base, casters, and gas cylinder shown in broken lines (disclaimed). A competitor, Zephyr Seating, Inc., sells a chair with a strikingly similar shell. Acme sues. (This scenario is hypothetical and offered only to illustrate the doctrine.)

At infringement, the jury, as an ordinary observer familiar with the crowded field of ergonomic office chairs, compares Acme's claimed shell to Zephyr's shell and to the prior art, and finds the two shells substantially the same and the differences trivial against the prior art. Infringement established.

At damages, Acme demands Zephyr's total profit. Zephyr argues that under Samsung v. Apple the "article of manufacture" is only the molded shell—a component it buys for, say, $40 and that represents a fraction of the chair's value—so total profit should be computed on the shell, not the $600 chair. Acme argues the article is the entire chair, because consumers buy the chair for its sculpted look and the shell is not sold separately. The court runs the four factors: the design claim covers only the shell (factor one favors Zephyr); the shell is the visually dominant, defining feature of the chair (factor two favors Acme); the shell is conceptually distinct from the mechanism but is the thing buyers see and choose (factors three and four split). The jury, on this record, could plausibly land on either base—and the gap between "total profit on a $40 shell" and "total profit on a $600 chair" is the gap between a nuisance verdict and a company-threatening one. That is § 289 in a sentence: the doctrine that makes design patents worth far more than their narrow scope would suggest.

A final damages note: § 289 is an election of remedies. The patentee cannot stack § 289 total profit on top of § 284 lost profits or a reasonable royalty for the same sales (see the "shall not twice recover" proviso; Braun Inc. v. Dynamics Corp. of America, 975 F.2d 815, 824 (Fed. Cir. 1992)). The patentee chooses the larger recovery. Reasonable-royalty damages under § 284 remain available as a floor and are sometimes the better path when the article of manufacture is a small component or when total profit is hard to prove. And the design owner can still seek a permanent injunction under 35 U.S.C. § 283, subject to the four-factor equitable test of eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006)—a meaningful remedy when the goal is to clear knockoffs off the shelf rather than collect money.

Willfulness, Enhanced Damages, and the Marking Statute

Willfulness and Enhancement

Design patents borrow the utility-patent enhancement provision. Under 35 U.S.C. § 284, a court "may increase the damages up to three times the amount found or assessed." Since Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016), the standard for enhancement is egregious, willful misconduct—the "wanton and malicious pirate"—judged under the totality of the circumstances rather than the old rigid two-part Seagate test. Note the interaction with § 289: enhancement is keyed to the § 284 damages found, so a patentee weighing enhanced reasonable-royalty damages against unenhanced § 289 total profit has a genuine election to model.

In design cases, willfulness often turns on direct copying evidence: design defendants frequently did copy, sometimes proudly, and copying is also a secondary consideration cutting against obviousness. So the same internal email that proves the defendant set out to clone the plaintiff's look can do double duty—supporting willfulness on damages and supporting validity against the post-LKQ obviousness attack. Plaintiffs should hunt for it; defendants should assume the plaintiff will find it.

Marking

The marking statute, 35 U.S.C. § 287, applies to design patents and is a quiet trap for plaintiffs. If a patentee (or its licensees) sells products embodying the patented design without marking them—either physically ("U.S. Patent D000,000") or via virtual marking (a "Patent" notice on the product directing to a freely accessible website that lists the patent)—the patentee generally cannot recover damages for infringement that occurred before the infringer received actual notice of the patent (a cease-and-desist letter or the complaint). For a product-design business that sells in volume and forgets to mark, this can erase years of damages: marking is constructive notice, and without it damages may start only at the date of the lawsuit. Design patentees should build marking into packaging and product tooling from day one and document it. We cover the mechanics, the virtual-marking option, and the litigation consequences in understanding patent and trade dress marking requirements and understanding patent marking requirements.

Because § 289 can produce enormous recoveries, the difference between damages running from the date of first infringement (marked) versus the date of the complaint (unmarked) is, again, potentially the whole case. Marking is the cheapest litigation insurance a design-patent owner can buy.

The International Trade Commission: A Powerful Alternative Forum

A design patent owner facing infringing imports has a second, very different forum: the U.S. International Trade Commission (ITC), under Section 337 of the Tariff Act of 1930 (19 U.S.C. § 1337). The ITC does not award damages. What it offers is an exclusion order—a directive to U.S. Customs and Border Protection to bar infringing products at the border—and cease-and-desist orders against domestic inventory. For a brand fighting cheap overseas knockoffs of a popular product design, an exclusion order can be more valuable than any money judgment, because it removes the infringing goods from the market entirely. (And note that eBay's injunction limits do not constrain the ITC, which issues exclusion orders on a public-interest standard rather than the four-factor equitable test.)

Three features make the ITC distinctive for design cases. First, it is fast: investigations typically resolve in roughly 16 to 18 months, far quicker than district court. Second, the complainant must prove a "domestic industry"—an economic investment in the United States in articles protected by the patent—which is usually straightforward for an established product but can be a real hurdle for a non-practicing entity. Third, the same substantive validity and infringement law applies: the ordinary-observer test from Gorham and Egyptian Goddess, the functionality bar, and post-LKQ obviousness all govern at the ITC just as in district court. Many design owners run a parallel track—an ITC action for the exclusion order and a district court suit for damages, with the district court case often stayed pending the ITC investigation under 28 U.S.C. § 1659. For the broader strategic map of cross-border IP enforcement, see global patent litigation strategies and, on the practical problem of reaching foreign infringers, serving legal documents to Chinese defendants in IP litigation.

Strategy: How These Cases Are Actually Won and Lost

Pulling the doctrine into a game plan.

For the plaintiff (the design owner). Your case is built on three pillars: a clean, well-drawn patent; a tight visual comparison; and a maximized § 289 base. Before filing, audit your own drawings for the ambiguities that sank patentees in simplehuman (are those lines seams or contours?), the over-functional designs that narrowed scope in Sport Dimension, and the article-of-manufacture mismatch that doomed Curver (does your claim language confine you to the wrong product?). Choose a defendant whose product is a close visual match—design patents are narrow, and a "close but not that close" target invites the prior-art-narrowing defense that doomed the plaintiffs in Lanard and Arc'Teryx. Develop your secondary-considerations record (commercial success, copying, industry praise) early, both to defeat post-LKQ obviousness and to support willfulness. Invest in a § 289 damages expert who can prove total profit on the broadest defensible article of manufacture. And confirm your marking is clean so damages run from first infringement, not the complaint.

For the defendant (the accused infringer). Your best friends are the prior art and the figures. A thorough prior-art search does triple duty: it narrows the patent's scope under Egyptian Goddess (helping non-infringement), it supplies anticipation references (validity), and it supplies the primary and secondary references for the now-revived obviousness attack under LKQ. Look hard for a distinctive claimed view the prior art lacks—or, on offense, for prior art that matches every view (the Apple v. Samsung side-view lesson cuts both ways). Inspect the figures for internal contradictions that support an indefiniteness defense under Maatita. Press functionality not to make the claim "cover nothing" (that argument lost in Ethicon) but to narrow scope and to argue your differing functional implementation reads differently to the ordinary observer. And on damages, do not sleep on Samsung v. Apple: put on real component-level accounting evidence to argue a narrow article of manufacture, because if you stay silent the plaintiff's whole-product base may stand by default.

For both sides. Decide early between district court and the ITC (or both). Understand that design patents are cheap to obtain, fast to issue, and—because of § 289—can carry outsized settlement value, which is exactly why they have become a favored tool against knockoffs and why defendants take them seriously. And because the obviousness landscape is changing month to month, keep current: the post-LKQ case law is still being written, and a defense that was hopeless in 2023 may be viable in 2026. For the strategic interplay between design and utility patents on the same product—and the double-patenting and term issues that creates—see the intersection of design and utility patents.

Key Takeaways

A design patent protects appearance, not function; its single claim is its drawings, so litigation is fundamentally a visual exercise governed by the eye of an ordinary observer descended from Gorham Co. v. White. Validity is attacked through functionality (which can narrow but not erase a claim, and which is indifferent to mere aesthetic appeal), anticipation (where every claimed view counts), obviousness (which after the 2024 en banc LKQ decision now runs on the flexible Graham framework and has become a meaningfully stronger defense), and occasionally indefiniteness in the drawings themselves. Infringement is decided under the sole test of Egyptian Goddess v. Swisa—the ordinary observer, viewing the designs in the context of the prior art—with the abandoned point-of-novelty test now absorbed into that comparison. And damages are the wild card: § 289's "total profit" remedy, reshaped by Samsung v. Apple's "article of manufacture" inquiry, can transform a narrow design patent into a company-threatening claim. Add willfulness, the marking trap, injunctions, and the ITC's border power, and you have a field where small pictures carry enormous stakes.

Frequently Asked Questions

Is the test for design patent infringement really just "do they look alike"? Almost—but with rigor. The ordinary-observer test, which traces to Gorham Co. v. White, 81 U.S. 511 (1871), and was made the sole test in Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc), asks whether an ordinary purchaser, giving the attention a buyer usually gives, would find the two designs substantially the same, such that the resemblance deceives. The twist is that the observer is "deemed conversant with the prior art," so in a crowded field of similar designs, small differences matter and the patent's scope is narrow. It is a holistic visual comparison, performed in the context of what came before—not a free-floating impression.

How can a design patent on a phone's front face cost the infringer the profit on the entire phone? Because of 35 U.S.C. § 289, which awards the patentee the infringer's "total profit" on the "article of manufacture" bearing the design—no apportionment. That nearly cost Samsung $399 million for infringing Apple's design patents. But in Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2016), the Supreme Court held that the "article of manufacture" can be a single component rather than the whole product, so the profit base may sometimes be much smaller. Courts now use a four-factor test (claim scope, prominence of the design, conceptual distinctness, and physical separability) to identify the right article.

What changed with the LKQ decision in 2024? LKQ Corp. v. GM Global Technology Operations LLC, 102 F.4th 1280 (Fed. Cir. 2024) (en banc), overruled the rigid Rosen-Durling obviousness test, which had required a single prior-art reference "basically the same" as the claimed design before any combination could be considered. In its place the court adopted the flexible four-factor Graham v. John Deere framework used for utility patents. The practical effect: obviousness is now a much more viable way to invalidate a design patent, though challengers must still identify a close primary reference and avoid hindsight, as Dynamite Marketing v. WowLine (Fed. Cir. 2025) confirmed.

Can a functional product look ever be protected by a design patent? Yes, as long as the overall design is primarily ornamental rather than dictated by function. Individual elements can have functions; what matters is whether alternative designs could achieve the same utility, and the fact that consumers find a design attractive does not make it functional (Automotive Body Parts Ass'n v. Ford, 930 F.3d 1314 (Fed. Cir. 2019)). Functionality can narrow a design claim—as in Sport Dimension v. Coleman (Fed. Cir. 2016)—but a court cannot read a design claim down to "nothing," as the Federal Circuit warned in Ethicon v. Covidien (Fed. Cir. 2015). And the trademark "aesthetic functionality" doctrine does not apply to design patents at all.

Does the wording of the claim ever matter if the drawings are the claim? Sometimes, yes. In Curver Luxembourg, SARL v. Home Expressions Inc., 938 F.3d 1334 (Fed. Cir. 2019), the figures showed only a "Y" pattern, but the claim language tied it to "a chair." The Federal Circuit held the patent limited to chairs and did not reach baskets, because a design patent has never protected ornamentation "in the abstract." So while the drawings define the visual design, claim and title language can confine that design to a specific article of manufacture.

Do I have to mark my products to recover design-patent damages? Usually, yes. Under 35 U.S.C. § 287, if you sell products embodying the patented design without marking them (physically or via a virtual-marking website), you generally cannot recover damages for infringement before the infringer received actual notice—often meaning damages start only at the date of the complaint. Given § 289's potential size, that lost period can be the whole case, so mark from day one.

Is the ITC worth using for a design patent? Often, especially against imported knockoffs. The International Trade Commission cannot award damages, but it can issue an exclusion order barring infringing products at the U.S. border under Section 337, and it moves faster than district court (roughly 16–18 months). You must prove a domestic industry, and the same validity and infringement law applies. Many design owners run an ITC action and a district court damages suit in parallel.

How is a design patent case different from a utility patent case? The claim is a set of drawings rather than words, so claim construction is visual curation rather than word-definition; infringement is judged by the ordinary-observer test rather than element-by-element claim mapping; there is no separate doctrine-of-equivalents step; the term is fifteen years from grant with no maintenance fees; and damages can reach total profit under § 289, a remedy utility patents lack. The validity defenses (anticipation, obviousness, plus the design-specific functionality bar) are conceptually parallel but visually applied. See the intersection of design and utility patents for the full comparison.

Related Articles

This article is general information, not legal advice. Design patent litigation is fact-specific and the law—especially design patent obviousness after LKQ—is evolving rapidly; consult qualified patent counsel about your particular situation.