A Japanese confectioner spent decades building a slim, chocolate-dipped cookie stick called Pocky into a cultural icon, registered its shape as trade dress, and then watched a competitor sell a nearly identical stick for thirty years. When the confectioner finally sued, it had everything a brand owner is supposed to have: federal registrations, fame, and a copyist who plainly knew exactly what it was imitating. It lost anyway. In 2020 the Third Circuit held that because almost every feature of the Pocky stick — the bare handle that keeps chocolate off your fingers, the slim diameter that fits your mouth, the length that lets a box hold many sticks to share — did something useful, the whole design was functional and free for anyone to copy. Ezaki Glico Kabushiki Kaisha v. Lotte International America Corp., 977 F.3d 261 (3d Cir. 2020). Fame did not save it. Registration did not save it. Three decades of brand investment did not save it.

That result is not a glitch. It is the system working exactly as designed. In the American economy, copying is the default and protection is the exception, and the features of a useful article not covered by a valid patent are, as a starting presumption, free for anyone to imitate. Everything in this corner of intellectual-property law is a negotiation between that baseline freedom and the legitimate interest a producer has in not being confused with its rivals. Get the negotiation wrong — claim too much, advertise the wrong things, lean on the wrong regime — and you end up like Ezaki Glico, holding paper that protects nothing.

This article is the policy-and-strategy installment in a set of related pieces on trade dress and product design. Where our companion articles work through the case law in depth and compare specific regimes head to head, this one steps back to ask the harder questions: Why does the law make product configuration so difficult to protect? Which of the overlapping intellectual property regimes should a company actually use, and in what combination? How do the proof burdens and registration practices play out in real disputes? And where is the doctrine heading as products migrate from physical shelves into digital storefronts, additive manufacturing, and virtual worlds? Readers who want the granular doctrinal history should also consult our comprehensive case law analysis of trade dress for product design and configuration, which traces the four foundational Supreme Court decisions and the circuit splits in detail; our primer on balancing brand identity and functional design, which walks element-by-element through how to plead and prove a trade dress claim; and our side-by-side comparison of design patents versus trade dress for product configurations, which weighs the two appearance regimes feature against feature. This piece deliberately stays in a different lane: the competition policy that explains why those doctrines look the way they do, and the strategy that follows from taking that policy seriously.

A note on terminology before we begin, because this field is dense with terms of art. Trade dress is the total image and overall appearance of a product or its packaging — everything that signals "this is a thing made by a particular company" other than the brand name or logo. Product configuration (used interchangeably here with "product design") is the subset of trade dress consisting of the shape and appearance of the product itself, as opposed to its packaging. Trademark law, of which trade dress is a branch, exists to protect source identification — to prevent consumers from being deceived about who made what. Functionality is the legal doctrine that strips protection from features that are useful rather than merely identifying. Secondary meaning (also called acquired distinctiveness) is the consumer association that must develop before a non-distinctive feature can be protected at all. Each is explained in plain terms below the first time it does real work.

The First Principle: The Right to Copy

To understand product configuration protection, you have to start somewhere counterintuitive — with the proposition that copying is lawful and often socially valuable. The American patent system is built on a bargain. An inventor who discloses a new, useful, and non-obvious invention to the public receives, in exchange, a time-limited monopoly. Utility patents run twenty years from filing; design patents fifteen years from issuance for applications filed on or after May 13, 2015. When that term expires, the bargain's other half kicks in: the invention enters the public domain and everyone is free to make it. That free-to-copy endpoint is not an accident or an oversight. It is the consideration the public receives for having tolerated the monopoly in the first place.

The Supreme Court made this principle the law of the land in a pair of 1964 decisions usually cited together, Sears, Roebuck & Co. v. Stiffel Co., 376 U.S. 225 (1964), and Compco Corp. v. Day-Brite Lighting, Inc., 376 U.S. 234 (1964). In both, a manufacturer of a lamp or lighting fixture tried to use state unfair-competition law to stop a competitor from copying an article that was not protected (or no longer protected) by a valid patent. The Court held that the federal patent laws preempted those efforts. "An article on which the patent has expired," the Court wrote in Sears, "is in the public domain and may be made and sold by whoever chooses to do so." 376 U.S. at 231. To let states forbid copying of unpatented designs would, the Court reasoned, hand a perpetual monopoly to articles that Congress had decided should either earn a limited patent or be free for all.

The Court later softened the sharpest edges of Sears and Compco. In Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141 (1989), it struck down a Florida statute that banned a specific cheap method of copying boat hulls (direct molding), but it confirmed that states retain room to regulate deception — to require labeling, to police passing off, to prevent consumer confusion about source. The line Bonito Boats draws runs through this entire field: a state, or federal trademark law, may protect against confusion about who made the product, but it may not, under the banner of unfair competition, grant a patent-like right to exclude others from making the product itself. As the Court put it, "free exploitation of ideas will be the rule, to which the protection of a federal patent is the exception." 489 U.S. at 151. Bonito Boats also supplies a phrase worth keeping: the Court warned against allowing a state to "offer patent-like protection to ideas deemed unprotected under the present federal scheme." That is the recurring nightmare of this body of law — protection that quacks like a patent without paying the patent system's price of disclosure and a fixed term.

This is why a trade dress plaintiff cannot simply say, "I designed this; you copied it; therefore you owe me." Copying, standing alone, is not a wrong. The plaintiff must show that the copied feature identifies source and that protecting it will not hand the plaintiff a monopoly over something useful. Those two requirements — distinctiveness and non-functionality — are the load-bearing walls of product configuration law, and the Supreme Court has spent three decades reinforcing both in ways that make protection harder, not easier, to obtain.

The Channeling Doctrine: Each Right Has Its Lane

Underneath the right to copy sits an organizing idea that legal scholars call channeling: the principle that the law tries to route each kind of value to the one intellectual-property regime designed for it, and to keep the regimes from bleeding into each other. Function is the domain of patent law, which grants strong rights but only after rigorous examination and only for a fixed term. Source identification is the domain of trademark and trade dress law, which can last forever but protects only against confusion. Original expression is the domain of copyright. Channeling does the housekeeping that keeps the bargain honest: it stops a producer from using trademark's potentially perpetual term to capture functional value that patent law says must expire, and it stops copyright from swallowing utilitarian design that belongs to the patent system.

You can see channeling at work in every doctrine discussed below. The functionality bar is channeling between trademark and patent. The useful-article exclusion in copyright is channeling between copyright and patent. TrafFix's rule that an expired utility patent is strong evidence of functionality is channeling enforced at the seam between two regimes a single producer is trying to straddle. Once you see the pattern, the cases stop looking like a grab bag of technical rules and start looking like a single immune system rejecting attempts to put the wrong kind of value in the wrong kind of right.

How Trade Dress Came to Cover Product Shapes at All

Trademark law began with words and symbols — brand names, logos, the marks burned into cattle or stamped on barrels. The conceptual leap to protecting a product's appearance came gradually, as courts recognized that consumers often identify a maker by look as much as by name. The federal trademark statute, the Lanham Act, never used the phrase "trade dress," but its broad protection of unregistered marks under Section 43(a), 15 U.S.C. § 1125(a), proved capacious enough to reach a product's total visual presentation. The Supreme Court confirmed the breadth of what can serve as a source identifier in Qualitex Co. v. Jacobson Products Co., 514 U.S. 159 (1995), holding that even a single color — there, the green-gold of a dry-cleaning press pad — can be a protectable mark if it has acquired source significance and is not functional. "It is the source-distinguishing ability of a mark — not its ontological status as color, shape, fragrance, word, or sign — that permits it to serve" the trademark function, the Court explained. 514 U.S. at 164.

For a brief stretch in the early 1990s, it looked as though trade dress protection might be relatively easy to obtain. In Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763 (1992), the Court held that trade dress — there, the festive décor and layout of a Mexican restaurant — could be inherently distinctive, meaning protectable immediately upon use without proof that consumers had come to associate it with a single source. That was a generous rule. But Two Pesos involved the look of a business, akin to packaging or trade-show dress, not the configuration of a tangible product. When the Court finally confronted product design itself, it pulled back sharply.

Wal-Mart and the Secondary-Meaning Wall

The decisive case is Wal-Mart Stores, Inc. v. Samara Brothers, Inc., 529 U.S. 205 (2000). Samara made a line of children's clothing with distinctive appliqués; Wal-Mart sold knockoffs sewn from photographs of the Samara line. Samara won at trial on a product-configuration trade dress theory. The Supreme Court reversed, and in doing so drew a bright line that governs the field to this day: product design trade dress can never be inherently distinctive; it is protectable only on proof of secondary meaning. 529 U.S. at 216.

The Court's reasoning is worth understanding because it explains the strategic landscape. Consumers, Justice Scalia wrote for a unanimous Court, do not ordinarily take the shape of a product as a signal of its maker. They take it as the product. A spoon shaped like a spoon, a coat shaped like a coat, a bottle shaped to hold liquid — the "predictable" consumer reaction is to see utility or aesthetics, not a brand. Because design almost always serves purposes other than source identification, the Court refused to let courts guess at inherent distinctiveness and instead required real-world proof that consumers had come to treat the design as a brand. The Court also offered a candid, anti-litigation rationale: requiring secondary meaning gives competitors and lower courts a clearer, harder-to-game rule, discouraging "strike suits" by which a producer could use a vague trade dress claim to harass a lawful copyist. 529 U.S. at 213-14. That passage matters for the competition theme of this article — the Court was explicitly worried that loose configuration rights would chill lawful competition, and it built the doctrine to prevent that.

Wal-Mart also resolved an ambiguity that Two Pesos had left open: what to do when it is genuinely unclear whether something is "product design" or "product packaging." The Court instructed lower courts to err on the side of treating ambiguous cases as product design, which means requiring secondary meaning. 529 U.S. at 215. The classic illustration the Court itself gave is a Coca-Cola bottle. Is the bottle "packaging" (it contains the soda) or "product configuration" (the bottle is itself a distinctive shape sold as part of the product experience)? The Court suggested the bottle is more like packaging and could be inherently distinctive, but conceded the line is not always crisp. For practitioners, the lesson is that anything resembling the shape of the salable article will be treated as configuration and will demand the heavier proof.

The Coca-Cola contour bottle is also the field's best illustration of what a successful configuration right looks like. Registered as a mark and recognized worldwide, the fluted "hobble-skirt" silhouette lets Coca-Cola exclude confusingly similar bottle shapes, anchor a unique visual identity, and use the bottle itself as a branding asset. But its scope is bounded by the same principles that constrain every configuration claim: Coca-Cola cannot bar competitors from using any bottle shape, only those likely to cause consumer confusion as to source. Other shapes have walked the same road with varying success. The triangular prism of a Toblerone bar is instantly recognizable and trades on a sense of Swiss-heritage premium quality, illustrating how a distinctive shape can carry both source significance and brand personality. By contrast, configuration claims in fast-moving design industries are perennially contested: in Apple Inc. v. Samsung Electronics Co., Apple asserted trade dress (alongside design and utility patents) in elements of the iPhone such as its rounded-rectangle front face and grid of icons, and although Apple prevailed on portions of its design-patent claims, its trade dress theories ran into exactly the obstacles this article describes — proving that the asserted features were non-functional and had acquired secondary meaning in a market saturated with similar rectangular touchscreens. Automakers face the same tension: Jeep has long defended trade dress in the Wrangler's boxy silhouette, flat sides, and seven-slot grille, a brand-defining look that nonetheless must be policed against the argument that much of a vehicle's shape is dictated by aerodynamics, safety, and packaging function. The through-line is that fame and investment buy nothing unless the claimed features clear the non-functionality and secondary-meaning bars.

What Secondary Meaning Requires, and Why It Is Hard

Secondary meaning exists when, in the minds of the relevant consumers, the primary significance of the feature is to identify the source of the product rather than the product itself. Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 851 n.11 (1982); Two Pesos, 505 U.S. at 766 n.4. The party claiming the right bears the burden of proving it, generally by a preponderance of the evidence, and courts repeatedly emphasize that the evidentiary showing for a product shape is substantial. General Motors Corp. v. Lanard Toys, Inc., 468 F.3d 405, 417-18 (6th Cir. 2006); Leapers, Inc. v. SMTS, LLC, 879 F.3d 731, 741 (6th Cir. 2018).

In practice, courts weigh a familiar cluster of factors, no one of which is dispositive:

  1. Direct consumer evidence, above all a properly conducted survey showing that an appreciable share of the relevant market associates the design with a single (even if anonymous) source. Surveys are expensive and methodologically fragile, but they are the gold standard.
  2. Length and exclusivity of use. A design used by one producer alone for many years is more likely to have acquired source significance than one used briefly or shared with others. Exclusivity is doing quiet but crucial work here: if competitors have used similar shapes, the consumer association cannot have formed in the first place.
  3. Amount and manner of advertising, and in particular "look-for" advertising that affirmatively directs consumers to recognize the design as a brand cue ("look for the curved bottle").
  4. Sales volume and market share, as circumstantial evidence that many consumers have encountered the design.
  5. Unsolicited media coverage that treats the design as iconic of the producer.
  6. Intentional copying by the defendant, which some courts treat as probative of secondary meaning on the theory that a copyist copies what consumers value as a source signal — though, consistent with Sears and Compco, copying alone proves nothing, and many courts give it limited weight.

The difficulty is structural. To prove secondary meaning in a shape, a producer must show that consumers have effectively learned to read the shape as a logo. That is a high bar, and it is why so many configuration claims fail at the threshold before functionality is even reached. Our detailed treatment of acquired distinctiveness in the filing context appears in how to file a trademark application with the USPTO and the federal trademark application checklists.

A subtle trap deserves emphasis, because it is where confident plaintiffs most often stumble. Consumer association caused by a monopoly is not the kind of secondary meaning the law rewards. This is the doctrine of de facto secondary meaning: where consumers associate a feature with one producer only because that producer held a patent and was the sole lawful source, the association does not create trademark rights. A.J. Canfield Co. v. Honickman, 808 F.2d 291, 297 (3d Cir. 1986). A producer cannot bootstrap a patent-derived monopoly into perpetual trade dress merely because, during the patent term, consumers had no one else to buy from. De facto secondary meaning is the recognition half of the same channeling concern that functionality enforces on the utility half: both prevent the patent system's temporary exclusivity from metastasizing into a permanent trademark monopoly once the patent lapses.

The Functionality Doctrine: The System's Safety Valve

If secondary meaning is the first wall, functionality is the second and, for policy purposes, the more important one. The functionality doctrine is the mechanism by which trademark law refuses to become a back door to a perpetual patent. The Lanham Act codifies it twice over: a functional feature cannot be registered, 15 U.S.C. § 1052(e)(5), and functionality is a complete defense even to an incontestable registration, 15 U.S.C. § 1115(b)(8). No matter how famous a product shape becomes, no matter how strong its secondary meaning, if the feature is functional it cannot be monopolized through trademark law. As the leading practice authority frames it, functionality is the doctrine that "strikes the balance" between a producer's interest in its design and two public interests on the other side of the scale: the interest in not extending exclusive rights beyond the parameters patent law sets, and the interest in letting competitors use functional features to compete effectively. Hold onto those two rationales — they decide hard cases.

The Inwood Test and the Two Faces of Functionality

The Supreme Court's basic definition comes from Inwood Laboratories: a product feature is functional, and thus unprotectable, "if it is essential to the use or purpose of the article or if it affects the cost or quality of the article." 456 U.S. at 850 n.10. This is utilitarian functionality, and every circuit has adopted it. Put plainly: if a feature is the reason the product works, or makes it cheaper or better to make, the feature is functional and free for all.

Courts also recognize a second, more contested category, aesthetic functionality: a feature that is not mechanically useful but whose exclusive use would put competitors at a significant non-reputation-related competitive disadvantage. Qualitex, 514 U.S. at 165. The paradigm is a feature consumers demand for its own sake — a particular color of outdoor furniture that matches everything, or, in older cases, the color black for boat motors because it coordinates with any hull and makes motors look smaller. Some circuits embrace aesthetic functionality robustly; others are skeptical of it or reject it outright as a redundant or unworkable concept. The Second Circuit's red-sole litigation, Christian Louboutin S.A. v. Yves Saint Laurent America Holding, Inc., 696 F.3d 206 (2d Cir. 2012), shows the doctrine's live edge: the court upheld Louboutin's red lacquered outsole as a valid mark as applied to shoes whose uppers contrast with the sole, but refused to let the brand monopolize red on a monochromatic red shoe, where the color is something competitors plausibly need. Both species of functionality share a single rationale: patent law already protects utilitarian features, trademark law must not extend exclusive rights beyond patent's bounds, and competitors must remain free to use functional features in order to compete. The interplay between aesthetic functionality and ornamental design protection is explored further in our companion piece on the evolution of design patent functionality.

TrafFix: Alternative Designs Will Not Save You

The most consequential functionality decision for strategy is TrafFix Devices, Inc. v. Marketing Displays, Inc., 532 U.S. 23 (2001). Marketing Displays had held now-expired utility patents on a dual-spring mechanism that kept road signs upright in the wind. When the patents expired and a competitor copied the dual-spring design, MDI sued for trade dress infringement, arguing the spring configuration identified its product. The Court rejected the claim and, in doing so, made two moves every practitioner must internalize.

First, the Court held that an expired utility patent is "strong evidence that the features therein claimed are functional." 532 U.S. at 29-30. A party trying to assert trade dress in a feature that was the subject of a utility patent "must carry the heavy burden of showing that the feature is not functional, for instance by showing that it is merely an ornamental, incidental, or arbitrary aspect of the device." This is the single most important sentence for anyone choosing between patent and trade dress: the utility patent you obtain today may be the exhibit that defeats your trade dress claim tomorrow. The patent's specification, by touting the utilitarian advantages of the very feature you later call a source identifier, becomes a confession of functionality.

Second, and just as important, the Court held that once a feature is functional under the Inwood test, courts should not proceed to ask whether alternative designs are available. 532 U.S. at 33-34. Before TrafFix, many courts had treated the existence of equally good alternative designs as evidence that a feature was not functional — the theory being that if competitors could compete with other shapes, exclusivity caused no harm. TrafFix rejected that approach for utilitarian functionality: "There is no need ... to engage ... in speculation about other design possibilities .... The dual-spring design is not an arbitrary flourish ...; it is the reason the device works." The competitor-need/alternative-designs inquiry survives, the Court indicated, only in the aesthetic functionality context. The practical upshot is that defendants and USPTO examiners attack utilitarian functionality first and hardest, because if they win there, the plaintiff's catalog of alternative designs is legally irrelevant.

Ezaki Glico (Pocky): "Useful" Means "Useful," Not "Essential"

TrafFix settled the role of alternative designs but left a question that Inwood's phrasing made tempting to argue: does "essential to the use or purpose" mean a feature is functional only if the product cannot work without it? Plaintiffs love that reading, because almost no single design choice is strictly indispensable. The Third Circuit shut it down in Ezaki Glico Kabushiki Kaisha v. Lotte International America Corp., 977 F.3d 261 (3d Cir. 2020) — the Pocky case from this article's opening.

Ezaki Glico held registered trade dress in the configuration of its thin, chocolate-dipped cookie sticks and sued a competitor who had sold a near-identical stick for decades. Ezaki argued that a feature is functional only if it is essential to the product. The court rejected that, reading the Supreme Court's precedents to mean that any useful feature is functional — not merely an indispensable one. A feature is functional if it makes the product work better in any respect, and the court catalogued how nearly every aspect of the Pocky stick was useful: the uncoated end is a handle that keeps chocolate off the fingers; the slim diameter lets you eat it without opening your mouth too wide; the length and shape let many sticks pack into a box for sharing. Because "every aspect of [the] design is useful," the configuration as a whole was functional and unprotectable, "no matter how brand-defining it had become." The court also emphasized the channeling rationale directly: trade dress protects features that identify source, while useful features fall "within the sphere of patent law."

Ezaki Glico is the practical face of TrafFix's policy, and it is sobering for brand owners. It means a product shape can be famous, registered, exclusively used for thirty years, and deliberately copied — and still be free for the taking — if its features pull their weight functionally. It also disposes of the comforting intuition that a design must be source-identifying because a copyist bothered to copy it. The copyist may simply want the working shape. For configuration strategy, Ezaki Glico underscores a brutal lesson: before you invest in a configuration as a brand, ask honestly whether each feature is doing a job or merely doing identification. If the answer is "a job," patent law is your only lane, and trade dress will not bail you out.

The Morton-Norwich Factors in Operation

Because Inwood's "essential to use or purpose / affects cost or quality" formula is abstract, courts and the Trademark Trial and Appeal Board (TTAB) apply a set of objective indicators drawn from In re Morton-Norwich Products, Inc., 671 F.2d 1332, 1340-41 (C.C.P.A. 1982). The four Morton-Norwich factors ask whether:

  1. a utility patent discloses the utilitarian advantages of the design (the strongest single factor after TrafFix);
  2. the owner's advertising touts the utilitarian advantages of the design (a self-inflicted wound, since marketing copy boasting that a shape "grips better" or "pours faster" is an admission of function — and one that sank Ezaki Glico, whose ads promoted Pocky's convenient design);
  3. alternative designs are available to competitors (relevant chiefly to aesthetic functionality and as a secondary consideration after TrafFix; note that the converse — showing other workable designs exist — does not by itself make a design non-functional, as Ezaki Glico confirmed); and
  4. the design results from a comparatively simple or cheap method of manufacture — i.e., whether the feature affects cost.

One refinement worth flagging is that the utility-patent factor is not satisfied just because the owner holds a utility patent somewhere in the vicinity. The patent's claims must actually overlap with the claimed trade dress features. In Ezaki Glico the court noted that the company's utility patent covered a method of making the stick rather than the stick's shape, so the patent did not directly prove the shape's functionality — yet the design was functional anyway on the other factors. The takeaway is two-edged: an off-point patent will not save you, and a patent whose claims do read on the shape will bury you.

These factors give the strategist a checklist that runs in both directions. A company that wants to preserve a configuration as trade dress will, ideally, avoid claiming the feature's utility in patents and ads, will be able to point to many competing shapes, and will resist describing the shape as cheaper to make. A company attacking a configuration will mine the owner's patent portfolio and marketing for admissions. For a deeper dive into how the same functionality logic now influences design-patent validity, see our piece on the intersection of design and utility patents.

Who Bears the Burden

Burden allocation matters enormously and turns on registration status. For unregistered trade dress asserted under Section 43(a), the Lanham Act places the burden of proving non-functionality squarely on the party asserting the right. 15 U.S.C. § 1125(a)(3); TrafFix, 532 U.S. at 29. For registered trade dress, registration creates a presumption of validity, so the challenger must come forward with evidence of functionality — but that presumption can be overcome, and functionality remains available as a defense even against an incontestable registration. The lesson for portfolio planning is that a federal registration is worth obtaining not only for its enforcement advantages but because it flips the functionality burden onto the copyist. (It is, however, no guarantee: Ezaki Glico held two registrations and still lost, because once a challenger marshals real functionality evidence, the presumption evaporates.)

One subtlety worth flagging is that functionality is asymmetric as a litigating position. A defendant may assert that the plaintiff's claimed trade dress is functional and therefore invalid — the ordinary defense. But a defendant cannot assert that its own allegedly infringing use is "functional" and therefore non-infringing; functionality is a doctrine about the validity of the asserted right, not a license to copy. The Fourth Circuit made this explicit in Rosetta Stone Ltd. v. Google, Inc., 676 F.3d 144, 162-63 (4th Cir. 2012), rejecting the argument that a defendant's purportedly "functional" use of another's mark could immunize it from liability. For the configuration strategist, the point is that functionality is a sword aimed at the plaintiff's claimed right, and the entire contest is fought over the validity of that right, not over the defendant's conduct.

A Worked Hypothetical: Litigating Functionality

The following is a hypothetical, offered to illustrate how the burdens and factors interact. GripCo sells a hex-shaped pencil grip and holds an unregistered trade dress claim in the hexagonal cross-section. A competitor, Scrivo, launches an identical hex grip, and GripCo sues under Section 43(a). Because the trade dress is unregistered, GripCo bears the burden of proving non-functionality from the outset — a structural disadvantage. Scrivo's counsel pulls GripCo's old utility patent, which (it turns out) claimed that "the hexagonal profile prevents the grip from rolling off a desk and improves finger purchase." That single sentence is close to dispositive: under TrafFix, the expired patent is strong evidence of functionality, and the patent's own words tout the utilitarian advantage (factor one and factor two of Morton-Norwich, both against GripCo). GripCo responds with a survey showing 70% of teachers recognize the hex grip as "the GripCo one" and a catalogue of alternative round and triangular grips on the market. Under pre-TrafFix law, the alternatives might have carried the day. After TrafFix and Ezaki Glico, they are nearly irrelevant: once the hex shape is shown to do a job — anti-roll, better purchase — the court does not weigh alternatives, and the survey proves only de facto recognition of a functional shape, which the law does not protect. GripCo loses on summary judgment. The lesson is that GripCo's fate was sealed years earlier, in the patent-drafting room and the marketing department, not in the courtroom.

The Overlapping Regimes: A Field Guide

A single product design can, in principle, implicate four distinct bodies of law at once. Understanding what each protects, for how long, and at what cost is the precondition for any rational protection strategy. The table below is a compressed map; the discussion that follows fills in the nuance.

Regime What it protects Key threshold(s) Term Defeated by
Utility patent How a thing works (function) Novelty, utility, non-obviousness; examination 20 years from filing Expiration; invalidity
Design patent Ornamental appearance Novelty, non-obviousness, ornamentality (not dictated by function) 15 years from issuance (post-5/13/2015) Functionality; invalidity
Trade dress Appearance as a source identifier Secondary meaning (always, for configuration) + non-functionality + likelihood of confusion Potentially perpetual, while in use Functionality (absolute); loss of distinctiveness
Copyright Separable artistic features of a useful article Originality + separability Life + 70 years Useful-article exclusion (no protection for shape)

Utility Patents

A utility patent protects new, useful, and non-obvious inventions — how a thing works. 35 U.S.C. §§ 101-103. It is the strongest right in the toolkit: a true twenty-year monopoly (measured from the earliest filing) over the functional invention, enforceable without any need to prove consumer confusion. But it is also the most expensive and slowest to obtain, requires public disclosure of the invention, and — crucially for our purposes — expires absolutely. After expiration the invention is dedicated to the public. A producer who relied on a utility patent to protect a useful shape will, post-TrafFix, find that the very patent forecloses any trade dress fallback. Utility patents protect function and are, by design, temporary. For the basics, see our utility patent basics and patent basics guides.

Design Patents

A design patent protects the "new, original and ornamental design for an article of manufacture." 35 U.S.C. § 171. It covers appearance, not function — the way a product looks, as depicted in the patent's drawings. The term is fifteen years from issuance for applications filed on or after May 13, 2015, with no maintenance fees. Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc), governs infringement under the "ordinary observer" test: infringement exists if an ordinary observer, familiar with the prior art, would be deceived into thinking the accused design is the patented one.

Design patents occupy an interesting middle ground. Like trade dress, they protect appearance; unlike trade dress, they require no proof of secondary meaning, no proof of consumer confusion, and no proof that the design identifies source. A design patent issues on a showing of novelty, non-obviousness, and ornamentality, and it then confers a clean right to exclude for fifteen years. The functionality constraint exists here too — a design "dictated by function" is unpatentable, Automotive Body Parts Ass'n v. Ford Global Technologies, LLC, 930 F.3d 1314 (Fed. Cir. 2019) — but the design-patent functionality test is more forgiving than trademark's, asking whether the overall design is dictated by function rather than whether individual features are useful. Notably, the availability of alternative designs and the mere fact that a design is aesthetically pleasing tend to support patentability — the opposite valence from trade dress, and a striking illustration of how the same word ("functional") carries different operational meaning in different lanes. The obviousness standard for design patents was recently reshaped by the Federal Circuit's en banc decision in LKQ Corp. v. GM Global Technology Operations LLC, which our firm analyzes at length in the LKQ decision: a seismic shift in design patent obviousness analysis. For the litigation mechanics, see our overview of patent litigation: design patents.

The most important practical point about design patents is timing. Because they protect appearance without any secondary-meaning requirement, a design patent is often the only meaningful protection available for a brand-new product whose shape has not yet acquired any consumer association. It buys fifteen years during which the producer can build the very secondary meaning that will later qualify the same shape for (potentially perpetual) trade dress protection. Design patent first, trade dress later, is one of the central layering strategies discussed below.

Trade Dress

Trade dress under the Lanham Act protects the appearance of a product as a source identifier. Its great advantage is duration: trademark rights can last indefinitely so long as the mark continues in use and continues to signify source. Its great disadvantages are the twin thresholds we have already covered — secondary meaning (always required for configuration after Wal-Mart) and non-functionality (always required, never waivable) — plus the need, in any infringement suit, to prove likelihood of confusion under the multifactor test familiar from word-mark cases. Trade dress protects source, potentially forever, but only for shapes that are both distinctive-in-fact and useless-as-function. For how confusion itself is analyzed, see navigating the maze of trademark confusion.

Copyright After Star Athletica

The fourth regime, often overlooked in product-shape planning, is copyright. Copyright protects original works of authorship, but it categorically excludes "useful articles" — objects with an intrinsic utilitarian function — from protection. The doctrinal escape hatch is separability: a pictorial, graphic, or sculptural feature of a useful article is copyrightable if it can be identified separately from, and is capable of existing independently of, the article's utilitarian aspects. 17 U.S.C. § 101.

For decades the courts splintered over how to apply separability. The Supreme Court provided a unified test in Star Athletica, L.L.C. v. Varsity Brands, Inc., 580 U.S. 405 (2017), a case about decorative chevrons, stripes, and color blocks on cheerleading uniforms. The Court held that a design feature is eligible for copyright protection if (1) it can be perceived as a two- or three-dimensional work of art separate from the useful article, and (2) it would qualify as a protectable work — on its own or fixed in some other medium — if imagined apart from the useful article. 580 U.S. at 417. Applying that test, the surface decorations on the uniforms were copyrightable because, imagined as a pattern on a canvas rather than on a uniform, they were original pictorial works.

Star Athletica matters to product-configuration strategy in a specific and bounded way. It expands copyright's reach over the applied ornament on a product — surface patterns, sculptural flourishes, applied artwork — without protecting the product's shape itself. The Court was emphatic that copyright does not extend to the useful article's shape "even if that shape is impressive." A jewelry maker can copyright an original sculptural figurine that happens to function as a paperweight, but cannot copyright the generic shape of a paperweight. For products that carry separable artistic features — fabric prints, ornamental carvings, applied decorative elements — copyright is now a cheap, fast, long-lived (life-plus-seventy-years), and registration-friendly complement to the other regimes. For pure shape, it remains unavailable, and the producer is thrown back on design patents and trade dress. This is channeling again: copyright keeps to ornament and refuses to absorb utilitarian form. Practitioners seeking the mechanics should review how to register a copyright with the US Copyright Office and our overview of the legal protection of software, which addresses analogous separability and overlap problems.

Strategic Selection and Layering of Protections

The regimes above are not mutually exclusive, and the sophisticated approach is to layer them by design rather than stumble into overlap by accident. The right combination depends on what kind of design is at issue and where the product sits in its life cycle.

Consider a hypothetical. Acme Audio Corp. has developed a wireless speaker with a striking, faceted hourglass body. The faceting also happens to disperse sound more evenly — a genuine acoustic benefit. The body is unlike anything on the market. How should Acme protect it?

Start by triaging features by function. The portions of the faceting that are acoustically load-bearing are likely functional and therefore unprotectable as trade dress; if Acme wants exclusivity over the acoustic geometry, its only real option is a utility patent on the sound-dispersion structure, accepting the twenty-year horizon and the TrafFix consequence that the patent will later defeat any trade dress claim over those same facets. Acme should make a clear-eyed decision: claim the acoustic function in a utility patent and surrender trade dress in those facets, or forgo the patent and preserve the option to argue non-functionality later. It cannot have both for the same feature — and after Ezaki Glico, it cannot rescue a useful facet by pointing to alternative speaker shapes.

The ornamental aspects of the body — the proportions, the hourglass silhouette, the non-acoustic facets — are candidates for a design patent, filed before any public disclosure, which gives Acme fifteen years of clean exclusivity over the look without proving secondary meaning. During those fifteen years Acme runs "look-for" advertising (taking care to advertise the silhouette as a brand cue, not as a performance feature), sells in volume, and cultivates press coverage, all building the consumer association that will let the same silhouette mature into trade dress by the time the design patent expires. If the speaker carries an applied surface pattern or a sculptural emblem, that ornament can be copyright-registered under Star Athletica for the longest term of all.

This is the canonical layering sequence: design patent for the early years, trade dress for the long term, copyright for any separable applied art, and a utility patent only where a genuine functional advantage justifies the trade-off and the producer accepts the loss of trade-dress fallback for the patented feature. The sequence exploits each regime's strengths and hedges each one's weaknesses. It also requires discipline: the company's patent counsel and trademark counsel must coordinate, because what one writes can sink the other. A utility patent's enthusiastic description of functional benefits is a gift to a future trade dress challenger; a marketing team's boast about performance is a Morton-Norwich admission; a design patent's drawings define the very scope a later trade dress claim will mirror. Companies building this kind of portfolio should also be running freedom-to-operate analysis for new products and treating the design as one component of a broader IP strategy.

A few cross-cutting strategic principles follow:

  • Decide early whether a feature is "function" or "brand." The same shape cannot be both monopolized as useful (patent) and protected forever as identifying (trade dress) without contradiction. Sort features into buckets at the design stage, before either the patent application or the ad campaign is written.
  • Mind the marketing copy. Advertising that boasts of a shape's utility is admissible against you on functionality. If a feature is meant to be trade dress, advertise it as a brand cue, not as a performance feature.
  • Use registration to flip burdens — but don't treat it as armor. A federal trade dress registration shifts the functionality and secondary-meaning burdens to the copyist and provides nationwide constructive notice, yet a determined challenger can still overcome the presumption with real evidence, as Pocky's owner learned.
  • Sequence, don't stack blindly. Design patents and trade dress dovetail in time; utility patents and trade dress collide. Plan the timeline.
  • Document the design process. Records of alternative designs considered, and of design choices made for non-functional reasons, are powerful non-functionality evidence later (relevant chiefly to aesthetic functionality after TrafFix, but still worth having).
  • Scale the strategy to the company. The cost of surveys, registrations, and enforcement falls hardest on startups and small businesses, which also struggle to build the market presence that secondary meaning demands. For them the most cost-effective entry point is usually a design patent filed before any public disclosure (no secondary-meaning burden, modest cost, fifteen years of clean exclusivity), supplemented by meticulous documentation of distinctive, non-functional design choices and, where available, unregistered or state-law trade dress rights — with the more expensive trade-dress and survey program deferred until the brand and budget can support it. A distinctive, protectable design is itself an asset that can raise a company's valuation, attract investors, and be licensed.

Registration Practice Before the USPTO and TTAB

Many configuration disputes never reach a courtroom; they play out as registration battles before the United States Patent and Trademark Office (USPTO) and its appellate tribunal, the Trademark Trial and Appeal Board (TTAB). The registration record is where functionality and distinctiveness are first contested, and it shapes everything downstream.

When an applicant seeks to register a product configuration as a mark, the examining attorney will almost invariably raise two refusals. The first is functionality under Section 2(e)(5), supported by the Morton-Norwich factors — the examiner will search for utility patents, scrutinize the applicant's own advertising for functional touts, and assess whether the configuration affects cost or quality. The second is non-distinctiveness: because product design can never be inherently distinctive under Wal-Mart, the examiner will refuse registration on the Principal Register unless the applicant proves acquired distinctiveness under Section 2(f), 15 U.S.C. § 1052(f). The applicant must submit the secondary-meaning evidence described earlier — declarations, survey results, advertising figures, sales data, length of exclusive use. The USPTO will sometimes accept five years of substantially exclusive and continuous use as prima facie evidence of acquired distinctiveness for ordinary marks, but for product configuration the Office and the Board generally demand more, treating the five-year provision as insufficient on its own for a shape.

A useful counterpoint — and a vivid reminder that the configuration/packaging distinction is outcome-determinative — is the TTAB's precedential decision in In re Seminole Tribe of Florida, Serial No. 87890892 (T.T.A.B. May 25, 2023). The Seminole Tribe sought to register the three-dimensional shape of its guitar-shaped Hard Rock hotel-and-casino building for casino, hotel, restaurant, and bar services. The examining attorney refused, reasoning that consumers do not perceive a building's exterior as a source identifier. The Board reversed, finding the guitar-shaped building inherently distinctive. The key move was characterization: because the building is the "package" in which casino and hotel services are delivered to consumers — not a tangible product sold to them — the Board treated it as analogous to product packaging, which can be inherently distinctive, rather than product configuration, which cannot. The Board then ran the Seabrook inherent-distinctiveness analysis, asking whether the design was a mere refinement of a commonly adopted, well-known form of ornamentation in the field. Seabrook Foods, Inc. v. Bar-Well Foods, Ltd., 568 F.2d 1342 (C.C.P.A. 1977). Drawing on In re Chippendales USA, Inc., 622 F.3d 1346 (Fed. Cir. 2010), and In re Frankish Enterprises, 113 U.S.P.Q.2d 1964 (T.T.A.B. 2015) (the dinosaur-bodied monster truck), the Board found that no one else used a guitar-shaped building for such services, that the design was therefore unique in the relevant field, and that consumers would immediately rely on it to distinguish the applicant's services — so the design was registrable without any secondary meaning. The case is a vivid demonstration that whether your design is "configuration" or "packaging" can decide the entire dispute — and that trade dress for service venues often escapes the Wal-Mart secondary-meaning wall that traps trade dress for products.

The TTAB is also the forum for oppositions and cancellations in which a competitor attacks a configuration registration on functionality or non-distinctiveness grounds, and a third party who spots a functional application during examination can submit a letter of protest with evidence and argument. The discovery and procedural mechanics of contesting these registrations are addressed in our guide to discovery practice in TTAB trademark proceedings, and the path for challenging adverse examiner decisions is covered in the USPTO trademark application checklists. One often-overlooked benefit of obtaining a configuration registration is its effect on marking and notice; producers juggling patent and trade dress marking obligations should consult our discussion of patent and trade dress marking requirements.

The Competition-Policy Frame: Why the Law Stays Restrictive

Step back and the architecture reveals a single, coherent policy: trademark law will protect a producer against confusion-based harm — being mistaken for a rival — but it will not let a producer convert the public's freedom to copy into a private monopoly over useful or aesthetically demanded features. Every doctrine we have examined enforces that boundary. Secondary meaning ensures protection attaches only where consumers genuinely use the design as a source signal. Functionality ensures protection never attaches to the parts that make the product work or that competitors need. The Sears/Compco/Bonito Boats preemption line ensures states cannot do an end-run around the federal patent bargain. And the heavy burdens of proof ensure that doubtful claims fail rather than chill lawful competition.

It helps to see that this is, at bottom, the same instinct that animates antitrust and unfair-competition law more broadly: the suspicion of monopoly that is not earned by a clear legislative bargain. The functionality doctrine is, in effect, a competition-law principle wearing trademark clothing. The point is sharpened by an unexpected cousin in the antitrust world — the "functional availability" defense under the Robinson-Patman Act, where a seller can defeat a price-discrimination claim by showing that a discount or service was functionally available to all competing buyers on equal terms. The shared intuition is that the law protects access to the functional means of competing, and refuses to let one player fence off what everyone needs to compete. Trademark's functionality bar does the same job for product shapes: a competitor who needs a useful shape to make a competitive product is entitled to use it, full stop.

The same public-interest instinct surfaces at the field's ethical edges, even where current doctrine does not squarely address them. Aggressive configuration claims can shade into appropriation when a design borrows from indigenous, traditional, or culturally significant forms, raising questions of attribution and consent that ordinary trade dress analysis is not built to answer. They can also bear on access: locking up the shape of a medical device, a safety component, or a repairable part can raise costs or foreclose competition in ways that reach beyond the parties to a given dispute. These concerns rarely decide a case on their own, but they inform how courts apply the functionality doctrine and how policymakers weigh the reach of any new protection — and they are a reminder that the freedom to copy serves consumers and the public, not just rival producers.

This restraint is not anti-innovation; it is the innovation policy. Patents reward and then release functional inventions on a schedule Congress set. Free copying of unpatented features drives the price competition and incremental improvement that consumers benefit from — recall that Sears and Compco were, at heart, decisions protecting the public's access to cheap lamps. Trademark protection, confined to genuine source identifiers, lets producers build reputations without commandeering the public domain. The system's critics are right that the lines are sometimes blurry and that aggressive plaintiffs can use vague configuration claims to intimidate smaller rivals — a concern Wal-Mart expressly invoked when it worried about strike suits. But the response the law has chosen is to make protection hard to get and easy to challenge, precisely so that the default of competition is disturbed only where the source-identification rationale is real. For practitioners, the takeaway is humility: a beautiful, expensive, beloved design is not, for that reason, protectable. Protectability turns on distinctiveness-as-source and non-functionality, full stop.

A Glance Abroad: The Same Tension, Different Settlement

The American settlement is not the only possible one, and a brief comparative glance reveals just how much of a choice the U.S. approach is. The European Union protects product appearance through a dedicated registered and unregistered Community design system (now the EU design right), which offers up to twenty-five years of protection for registered designs and a short period of automatic, unregistered protection that arises upon disclosure of the design within the EU — a regime far more generous to product shape than anything in U.S. law, and one that does not demand secondary meaning. The EU also permits registration of three-dimensional shapes as trademarks, generally at a lower distinctiveness threshold than the United States demands and with more emphasis on the design's overall impression than on individual features, and many member states layer robust unfair-competition laws on top. The EU likewise excludes features "solely dictated by technical function," its own channeling rule, but it draws the functionality line differently and more narrowly than TrafFix and Ezaki Glico do.

The major Asian economies offer their own blends. Japan provides design-registration protection for product appearance, allows registration of three-dimensional trademarks for distinctive shapes, and supplements both with an Unfair Competition Prevention Act that can protect famous product configurations against imitation. China grants design patents and permits three-dimensional trademark registration, though establishing the secondary meaning needed to register a product shape remains difficult in practice. South Korea similarly protects appearance through design rights and 3D trademark registration, backed by strong unfair-competition statutes. The contrast underscores the central theme: every system must balance innovation against competition, but the United States has chosen to weight the scale toward the freedom to copy, while much of the rest of the world weights it somewhat more toward the designer. A company selling globally cannot assume its home-jurisdiction strategy travels; a shape that is free to copy in New Jersey may be an enforceable registered design in Munich, Tokyo, or Seoul. International filing strategy for the appearance regimes, including the Hague system for design patents, is a separate planning exercise worth undertaking early — and it should account for the fact that the EU's automatic unregistered design right and Asia's unfair-competition regimes can supply protection that U.S. law would withhold.

Emerging Pressure Points

The doctrine described above was built for a world of physical products sold on shelves. Three developments are now testing whether it can stretch.

3D Printing and Additive Manufacturing

Inexpensive 3D printing collapses the distance between a digital design file and a physical object, letting anyone with a printer reproduce a product shape from a downloadable file. This stresses the system in two directions. On enforcement, it multiplies the number of potential copyists and shifts infringement from factories to individuals and small shops, making detection and litigation impractical. The legal characterization questions are genuinely novel: is the design file itself an infringing article, or only the printed object? Trade dress and design-patent infringement are framed around making, using, and selling the article; a file is arguably none of these until printed, which pushes rights holders toward contributory and inducement theories against those who distribute files, and toward contractual and digital-rights-management controls over file distribution. Functionality analysis is unchanged in principle — a printed copy of a functional shape is just as lawful as an injection-molded one — but the practical reality is that additive manufacturing magnifies the cost of the law's pro-copying default. The same dynamics drive the broader debates over consumer access we examine in the right to repair movement.

Digital and Virtual Product Configurations

As commerce moves into apps, augmented reality, and immersive virtual environments, the question becomes whether the configuration of a virtual object can be protected, and under which regime. A virtual handbag, sneaker, or vehicle in a game or metaverse platform has a "shape," but it has no utilitarian function in the ordinary sense — it does not carry, cushion, or transport anything. That fact cuts in a counterintuitive direction: a shape that would be functional (and thus unprotectable as trade dress) in the physical world may be purely ornamental when rendered as a digital good, because the utility that made the physical shape functional has evaporated. The Pocky stick's anti-roll, easy-to-hold geometry would carry no functional baggage as a virtual collectible. The flip side is that purely virtual configurations may struggle to show secondary meaning as product design, and courts and the USPTO are still working out how trademark and trade dress map onto goods that exist only as software. Smart, connected products complicate the picture from the other direction: when an Internet-of-Things device's behavior — and even its apparent functionality — can change through a software update, the functionality analysis that fixes on a static physical shape fits awkwardly, and protection may have to reach the user-interface and interaction design as much as the housing. A device whose useful functions migrate into firmware may leave its enclosure looking more ornamental than it once was, shifting the configuration calculus over the product's life. Our firm addresses the broader brand-protection questions raised by these markets in trademark challenges in the metaverse and virtual goods and its companion on brand protection strategies for digital-first commerce. The intersection of generative design tools and ownership — when a configuration is produced by an algorithm rather than a human designer — raises authorship and inventorship questions explored in AI-generated inventions: who owns what the machine creates.

Convergence and the Risk of Over-Layering

A final emerging tension is internal to the strategy this article recommends. As producers layer design patents, trade dress, and copyright over the same product, courts and commentators have begun to worry about cumulative over-protection — the prospect that a clever stacking of regimes could achieve, in combination, the perpetual monopoly over useful or basic shapes that each regime alone forbids. The doctrines push back at the seams: functionality polices the trade dress and design-patent overlap; the useful-article exclusion polices the copyright boundary; and TrafFix polices the patent-to-trade-dress handoff. But the watchful practitioner should expect courts to scrutinize aggressive layering, especially where the practical effect would be to lock up a shape the public should be free to copy. Layering is a legitimate and powerful strategy; layering as a device to evade functionality is not. The channeling principle, in other words, is not just a set of entry rules for each regime — it is also a backstop against using the regimes in combination to smuggle in the very monopoly each was designed to deny.

Key Takeaways

  • Copying is the default. Under Sears, Compco, and Bonito Boats, unpatented product features are presumptively free to copy. Protection is the exception and must be justified by a genuine source-identification interest.
  • Channeling explains everything. The law routes function to patents, source-identifiers to trademarks, and expression to copyright, and it polices the seams to stop value from leaking into the wrong (often longer-lived) regime.
  • Product configuration always requires secondary meaning. After Wal-Mart v. Samara, product-design trade dress can never be inherently distinctive. The producer must prove that consumers read the shape as a brand — a demanding, evidence-intensive showing — and association caused by a former patent monopoly (de facto secondary meaning) does not count.
  • Functionality is an absolute bar, and "useful" is enough. Under Inwood, TrafFix, and Ezaki Glico (Pocky), a feature that is useful — essential to use, or affecting cost or quality — cannot be protected as trade dress no matter how famous, registered, or deliberately copied. Alternative designs do not rescue a utilitarian-functional feature; an expired utility patent whose claims read on the shape is strong evidence of functionality.
  • The four regimes do different jobs. Utility patents protect function (twenty years, then public domain); design patents protect appearance without secondary meaning (fifteen years); trade dress protects appearance as a source identifier (potentially forever, but hard to obtain); copyright protects separable applied ornament after Star Athletica (life plus seventy years) but never the useful shape itself.
  • Layer deliberately, sequence carefully. Design patent first, trade dress later, copyright for separable art, and a utility patent only when a real functional advantage justifies surrendering the trade-dress fallback for that feature.
  • Registration matters but is not armor. A Principal Register registration flips the functionality and secondary-meaning burdens onto the copyist and provides nationwide notice; a determined challenger with real evidence can still win, as Pocky shows.
  • Characterization can decide the case. Whether a design is "product configuration" or "product packaging" — as the guitar-building Seminole Tribe decision shows — often determines whether secondary meaning is required at all.
  • Watch the frontier. 3D printing magnifies enforcement difficulty and raises file-versus-article questions; virtual goods invert the functionality analysis; over-layering invites judicial scrutiny; and foreign regimes (e.g., the EU design right) strike the innovation-competition balance differently.

Frequently Asked Questions

Can I protect the shape of my product as a trademark the moment I launch it? Generally no, if the shape is product configuration. Wal-Mart v. Samara holds that product design can never be inherently distinctive, so you must first build and then prove secondary meaning — a consumer association between the shape and your company. That typically takes years of exclusive use, advertising, and sales. A design patent, by contrast, can protect the appearance immediately, without any secondary-meaning showing, which is why many companies start there.

If my design is genuinely beautiful, famous, and expensive to develop, isn't that enough to protect it? No. Beauty, fame, cost, and consumer affection are not the legal test. The Pocky case (Ezaki Glico v. Lotte) is the cautionary tale: a famous, registered, decades-old design lost because its features were useful. Trade dress turns on whether the design identifies your company as the source and whether it is non-functional. A gorgeous, beloved shape that consumers perceive as "just the product," or one whose features make the product work better or cheaper, is not protectable as trade dress regardless of how much it cost to create.

Doesn't the fact that a competitor copied my design prove it's a source identifier? Not by itself. Copying is the lawful default for unpatented features, and a copyist often wants your working shape, not your brand signal. Some courts treat deliberate copying as weak circumstantial evidence of secondary meaning, but after Sears, Compco, and Ezaki Glico, copying alone proves nothing and cannot overcome a finding that the feature is functional.

What is the single biggest mistake companies make? Letting their patent strategy and marketing undermine their trade dress. A utility patent that touts a shape's functional benefits, or advertising that brags about how a shape improves performance, becomes the key exhibit proving functionality. TrafFix makes an expired utility patent strong evidence that the patented features are functional. Coordinate patent counsel and trademark counsel — and the marketing team — from the start.

How long does each form of protection last? A utility patent lasts twenty years from filing; a design patent fifteen years from issuance (for applications filed on or after May 13, 2015); copyright generally the life of the author plus seventy years; and trade dress potentially forever, so long as the mark stays in use and continues to identify source. The indefinite term is the chief allure of trade dress — and the chief reason the law makes it so hard to obtain for product shapes.

Does copyright protect my product's shape after Star Athletica? Only its separable artistic features, not the shape itself. Star Athletica allows copyright for a design element that can be perceived as a standalone work of art and would qualify for protection on its own — surface patterns, applied ornament, sculptural flourishes. The utilitarian shape of the article remains outside copyright "even if that shape is impressive."

Can I protect the configuration of a purely virtual product? The law is unsettled. A virtual object's shape often lacks the physical utility that triggers the functionality bar, which can make protection easier on functionality grounds; but proving secondary meaning and fitting virtual goods into existing trademark and trade dress categories remains an evolving area. Brand owners entering these markets should plan for uncertainty and consider the metaverse-specific strategies discussed in our companion articles.

Is the U.S. approach the same everywhere? No. Many jurisdictions, notably the European Union with its registered and unregistered design right, protect product appearance more readily than U.S. law and do not require secondary meaning for shapes. A global product strategy should account for the fact that a shape free to copy in the United States may be an enforceable registered design abroad.

Conclusion

Product configuration protection is best understood not as a grant of rights but as a carefully bounded exception to a freedom. The freedom — to copy unpatented articles — is the engine of competition that American law has chosen to favor, from the cheap lamps of Sears and Compco to the chocolate sticks of Ezaki Glico. The exception — protection for designs that genuinely identify their source and that take nothing functional from the public — exists to prevent consumer deception, not to reward design investment as such. Every doctrine in the field, from Wal-Mart's secondary-meaning wall to TrafFix and Ezaki Glico's treatment of functionality to the Sears/Compco right to copy, is calibrated to keep that exception narrow, and the channeling principle ties them all together.

For businesses, the practical path forward is to treat the overlapping regimes as a coordinated portfolio rather than a grab bag: a design patent to bridge the early years when no secondary meaning yet exists, trade dress to carry distinctive non-functional appearance into the long term, copyright to capture separable applied ornament after Star Athletica, and a utility patent reserved for features whose genuine function justifies the trade-offs. The discipline that separates a durable strategy from an expensive failure lies in honestly sorting features into "function" and "source," in keeping patents and marketing from sabotaging trade dress, and in respecting — rather than trying to evade — the functionality doctrine that keeps the whole system honest. Ezaki Glico did none of those things consistently, and thirty years of brand-building bought it nothing enforceable. As products migrate into additive manufacturing and virtual environments, the underlying questions will not change, even as their answers must. Distinctiveness as a source identifier, and freedom for the functional, will remain the twin poles around which this entire body of law turns.

Related Articles

This article is provided by mclaw.io for general informational purposes only and does not constitute legal advice, nor does it create an attorney-client relationship. Intellectual property law is fact-specific and varies by jurisdiction and over time. Readers should consult qualified counsel before making decisions about protecting or contesting product configuration or trade dress rights.