A business almost never sets out to steal a brand. It picks a name it likes, hires a designer for the logo, prints the packaging, and goes to market. Then, sometimes years later, a letter arrives. It does not merely accuse the business of infringement; it accuses the business of willful infringement, and it demands the infringer's entire profits, three times the plaintiff's damages, and the plaintiff's attorney's fees. For a company that has built real revenue under the disputed name, that demand is not a nuisance. It is an extinction-level event.

Here is the counterintuitive part. Whether that demand succeeds or collapses often turns less on whether the two marks are confusingly similar than on what was in the defendant's head when it chose the name -- and, more precisely, on the paper trail the defendant created before it ever printed a single label. Two companies can lose the identical lawsuit on the identical facts. One walks away owing a modest royalty and an order to rebrand. The other surrenders its profit stream, trebled damages, and the other side's legal bills. The variable that separates them is good faith, and good faith is something you build on the front end, with documents, long before anyone files suit.

This article explains how that record is built and why it is worth far more than its modest cost. It covers what willful trademark infringement means and the enhanced remedies it unlocks under Section 35 of the Lanham Act, 15 U.S.C. 1117; how the Supreme Court's decision in Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), reshaped the field by removing willfulness as an absolute prerequisite to disgorgement while leaving it the decisive factor; the specific equitable considerations courts weigh before forcing a defendant to give up profits; how good-faith adoption is proved; the protective value of a comprehensive clearance search, a reasoned opinion of counsel, and USPTO examination and registration; the strategic cost of waiving the attorney-client privilege to rely on counsel; the Tea Rose-Rectanus good-faith junior user doctrine; and the willful-blindness and post-notice limits that keep the shield honest. A worked hypothetical ties the doctrine to practice, and a closing FAQ answers the questions brand owners ask most.

One note on terminology. Throughout, invented businesses are labeled clearly -- "Lumen," "Northwind," and the like -- so no real company is implied, and every term of art is explained in plain language the first time it appears.

What "Willful" Trademark Infringement Means

Trademark infringement, at bottom, is the unauthorized use of a mark in a way likely to confuse consumers about the source, sponsorship, or affiliation of goods or services. The Lanham Act supplies the cause of action: Section 32, 15 U.S.C. 1114, for federally registered marks, and Section 43(a), 15 U.S.C. 1125(a), for unregistered (common law) marks and for false designations of origin. Most cases plead both, because a Section 43(a) claim does not depend on a registration and reaches a broader range of conduct. The pivotal question in either is likelihood of confusion: would an ordinary consumer, meeting the defendant's mark in the marketplace, be likely to believe that the defendant's goods come from, or are endorsed by, the plaintiff?

That inquiry is objective. Courts answer it with a multi-factor balancing test that varies a little by circuit but generally tracks the factors first catalogued in Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492 (2d Cir. 1961), and, for registrability, in In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973): the similarity of the marks; the relatedness or proximity of the goods or services; the strength of the plaintiff's mark; evidence of actual confusion; the marketing channels used; the degree of care likely to be exercised by purchasers; the defendant's intent in selecting the mark; and the likelihood that either party will bridge the gap into the other's market. (We unpack each of these in navigating the maze of trademark confusion.) The crucial point for present purposes is that a plaintiff does not need to prove the defendant intended to confuse anyone in order to win on liability. An entirely innocent business that honestly believed its mark was distinct can still be found liable if a court concludes confusion is likely. Intent is one ingredient in the confusion calculus, never a precondition to it.

"Willfulness," by contrast, is a question about the defendant's state of mind. A defendant infringes willfully when it adopts or continues to use a mark with knowledge that the use infringes another's rights, or with reckless or willfully blind disregard of those rights, and typically with an aim of trading on the goodwill the plaintiff has built. The classic markers of willfulness, drawn from cases across the circuits, include knowledge of the plaintiff's prior use before adoption; deliberate copying of a mark's distinctive features; an intent to confuse consumers or free-ride on the plaintiff's reputation; and -- this one matters enormously -- continued use after receiving credible notice of the conflict. "Good faith," its mirror image, describes a defendant who chose the mark for legitimate reasons unrelated to the plaintiff, investigated potential conflicts before adopting, sought and followed professional advice, and responded reasonably when a problem surfaced.

The vocabulary is not perfectly uniform, and a careful litigant minds the differences. Courts and the statute variously speak of "bad faith," "willfulness," "intentional" infringement, and "exceptional" cases. These concepts overlap heavily but are not interchangeable, and a given remedy may answer to a particular one of them. The unifying idea is that culpability sits on a spectrum -- from innocent good-faith adoption, through negligence, through reckless disregard, up to deliberate counterfeiting -- and where a defendant lands on that spectrum decides how much money changes hands.

Why Willfulness Matters: The Enhanced Remedies at Stake

Liability and remedy are two different questions, and conflating them is the most common error lay clients make. A defendant can lose on liability -- be found to infringe and be enjoined -- yet owe little or nothing in money if it acted in good faith. That is because the painful monetary remedies under the Lanham Act are discretionary and governed by principles of equity, and the defendant's intent is the dominant equitable consideration.

Section 35(a), 15 U.S.C. 1117(a), authorizes a prevailing plaintiff, "subject to the principles of equity," to recover three categories of monetary relief: the defendant's profits; any damages the plaintiff sustained; and the costs of the action. The statute then layers on two enhancements. In assessing damages, the court "may enter judgment, according to the circumstances of the case, for any sum above the amount found as actual damages, not exceeding three times such amount" -- the treble-damages provision -- so long as the enhancement functions as compensation, not a penalty. And "in exceptional cases," the court "may award reasonable attorney fees to the prevailing party." Each enhancement turns, in practice, on culpability.

A word on the hierarchy of remedies before we reach the money. The remedy courts care about most is usually the injunction, not damages, because the harm to a brand -- consumer confusion eroding the goodwill the owner spent years building -- is the kind monetary relief cannot adequately repair. See Century 21 Real Estate Corp. v. Sandlin, 846 F.2d 1175, 1180 (9th Cir. 1988). An injunction issues without any finding of bad faith and without proof of actual damage; the good-faith infringer gets enjoined right alongside the bad-faith one. Good faith does its work at the next layer down, where the dollars live.

Disgorgement of profits. The most consequential remedy in many trademark cases is disgorgement -- forcing the infringer to surrender the profits it earned while using the mark. It rests on two ideas: that an infringer should not be unjustly enriched by another's goodwill, and that infringement should not pay. Because it is measured by the defendant's gain rather than the plaintiff's loss, disgorgement can dwarf any provable damages. It is the difference between a parking ticket and a foreclosure. As we discuss below, willfulness is no longer a strict precondition to a profits award in an ordinary infringement case, but it remains the single most important factor in whether a court orders disgorgement at all.

Treble (enhanced) damages. Courts have discretion under Section 1117(a) to award up to three times the plaintiff's actual damages, but in practice they reserve enhanced awards in infringement actions for willful conduct. A good-faith infringer found liable may owe compensatory damages -- a reasonable royalty for the use, or demonstrable lost sales -- but is unlikely to face a trebled judgment.

Attorney's fees in "exceptional" cases. Since the Supreme Court's patent decision in Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014), the courts of appeals have largely imported its standard into the Lanham Act's "exceptional case" inquiry, because the fee provisions are textually parallel. Under that standard, an exceptional case is simply one that "stands out from others" with respect to the substantive strength of a party's litigating position or the unreasonable manner in which the case was litigated, judged on the totality of the circumstances. Willful or bad-faith infringement is a paradigm exceptional case warranting fees. A defendant who can show it acted in good faith hands the court a ready reason to call the case unexceptional and deny fees.

Counterfeiting and dilution: where willfulness is still a gate. Two corners of the statute keep willfulness as a hard requirement, not merely a factor. For use of a counterfeit mark, 15 U.S.C. 1117(b) makes treble damages (or, at the plaintiff's election, statutory damages under 1117(c)) and attorney's fees mandatory absent extenuating circumstances when the violation is intentional. And for federal dilution claims under 15 U.S.C. 1125(c), monetary relief is available only where the defendant willfully intended to trade on the famous mark's recognition or to harm its reputation, per 15 U.S.C. 1125(c)(5). In those settings, proving or disproving willfulness is not about the size of the award; it is the gateway to any monetary recovery at all.

The practical upshot bears repeating because clients so often miss it. Two defendants can lose the identical infringement case on the identical facts. The good-faith defendant may owe a modest reasonable royalty and be ordered to rebrand. The willful defendant may owe its entire profit stream, trebled damages, and the plaintiff's legal bills. Good faith, in trademark law, is worth more than most insurance policies -- and unlike insurance, you cannot buy it after the loss.

The Equitable Factors Courts Actually Weigh Before Ordering Disgorgement

It is one thing to say disgorgement is "equitable" and "discretionary." It is more useful to know what, concretely, the equities consist of. Courts assessing whether to make a defendant give up profits in an infringement case do not flip a coin; they run through a recognizable, if circuit-flavored, set of considerations. The recurring factors include:

  • The defendant's intent -- the centerpiece. Willfulness is no longer a precondition after Romag, but it remains the heaviest thumb on the scale, and a documented record of good faith is the strongest counterweight available.
  • Whether the plaintiff has shown actual consumer confusion, which signals real diversion of trade and strengthens the case for disgorgement.
  • Whether the defendant has been unjustly enriched -- did the defendant in fact profit from the use of the mark, and is there a causal link between the infringement and those profits?
  • Whether the plaintiff is otherwise adequately compensated by an injunction or by actual damages, in which case a profits award may overcompensate.
  • The need for deterrence, both of the particular defendant and of others.
  • Whether the plaintiff delayed in asserting its rights (a laches-flavored equitable consideration, discussed below).
  • The closeness of the case on the merits, which bears on whether the defendant's conduct was culpable or merely a reasonable bet on a debatable question.

Three equitable rationales animate this list -- compensation, unjust enrichment, and deterrence -- and courts often say so explicitly. The reason intent dominates is that two of the three (unjust enrichment and deterrence) speak directly to the defendant's culpability; an innocent good-faith adopter is, by hypothesis, not the unjustly enriched bad actor the doctrine is designed to disgorge. That is exactly why Justice Sotomayor's Romag concurrence, discussed next, warned against profits awards for innocent infringement -- it would sever disgorgement from the equitable theory that justifies it.

Two further rules constrain the money. First, a profits award and an actual-damages award cannot be combined where the combination would overcompensate the plaintiff -- the recovery measures the harm or the gain, not a stacking of both. See Big O Tire Dealers, Inc. v. Goodyear Tire & Rubber Co., 408 F. Supp. 1219, 1241 (D. Colo. 1976). Second, where disgorgement is unwarranted, courts routinely turn to alternative, compensatory measures that are far less punishing: a reasonable royalty for the unauthorized use, Sands, Taylor & Wood Co. v. Quaker Oats Co., 34 F.3d 1340, 1343-44 (7th Cir. 1994), or the cost of corrective advertising to dispel confusion, Big O Tire Dealers, 408 F. Supp. at 1232-33. For a good-faith infringer, the realistic exposure is usually one of these modest figures, not the headline profits number in the demand letter.

Romag Fasteners v. Fossil: What Changed and What Did Not

For decades the federal circuits split over a single question: must a plaintiff prove the defendant infringed willfully before it can recover the defendant's profits? Several influential circuits, including the Second and the Ninth -- which together hear an enormous share of the nation's trademark litigation -- treated willfulness as an absolute prerequisite to disgorgement. Others treated it as one factor among many. The same conduct could thus yield a multimillion-dollar profits award in one courthouse and nothing in another, an intolerable state of affairs for a national statute.

The Supreme Court resolved the split in Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020). Romag made magnetic snap fasteners and licensed Fossil to use them in Fossil's handbags. Romag discovered that factories in China making Fossil products were using counterfeit fasteners stamped with the ROMAG mark. A jury found Fossil had infringed and had acted with "callous disregard" of Romag's rights -- but it specifically found Fossil had not acted willfully. Because controlling Second Circuit precedent made willfulness a precondition to disgorgement, the lower courts denied Romag any of Fossil's profits.

The Supreme Court reversed unanimously. Writing for the Court, Justice Gorsuch turned to the text. Section 35(a) conditions a profits award for a dilution violation under Section 43(c) on the violation being "willful," yet the same provision imposes no parallel willfulness condition on a profits award for ordinary infringement under Sections 32 or 43(a). When Congress writes a mental-state requirement into one part of a statute and leaves it out of a neighboring part, courts read the omission as deliberate. The Court refused to graft onto the infringement remedy a precondition Congress had declined to write, noting the statute's "considerable care" elsewhere in spelling out mental states made the silence here telling. Romag, 590 U.S. at 217-19.

But -- and this qualification is indispensable -- Romag did not make willfulness irrelevant. The Court was emphatic that "a trademark defendant's mental state is a highly important consideration in determining whether an award of profits is appropriate." Id. at 219. Justice Alito, joined by Justices Breyer and Kagan, concurred to underscore that willfulness is "a highly important consideration" even if not an "absolute precondition." And Justice Sotomayor, concurring in the judgment, warned against reading the decision as an invitation to disgorge from innocent actors: awarding profits for "innocent or good-faith trademark infringement," she wrote, would not be "consonant with the 'principles of equity'" that Section 35(a) expressly invokes.

So the post-Romag rule states cleanly. Willfulness is not a strict prerequisite to disgorgement in an infringement case, but it remains the most important equitable factor, and a defendant who can affirmatively prove good faith stands on very strong ground to argue that disgorgement, treble damages, and fees are all inappropriate. The reform cuts in two directions at once. It stripped away the categorical shield that good-faith defendants in some circuits used to enjoy automatically, while preserving an equitable framework in which a demonstrated record of good faith is powerfully exculpatory. The lesson for businesses is bracing: good faith is no longer something the burden of proof will protect for free. It is something you must be able to prove, with contemporaneous documents, because the plaintiff no longer has to disprove it as a threshold matter. The defendant who walks into discovery without a clearance file is, post-Romag, naked.

(A practical caveat on circuit variation: Romag abolished only the willfulness precondition. The circuits still differ on adjacent points -- whether a defendant's profits must be apportioned to the infringement, how causation is shown, and how heavily a given court weighs deterrence. Counsel should check the law of the relevant circuit, and the Practical Law Trademark Infringement Profit and Damage Awards Standards by Circuit Chart is a useful map.)

How Good-Faith Adoption Is Established

If good faith is the prize, how is it won? Courts look for affirmative, contemporaneous evidence that the defendant chose and used its mark honestly -- not after-the-fact rationalizations assembled for litigation, but a real record made at the time of adoption. Four kinds of evidence carry the most weight.

The first is a legitimate, independent reason for selecting the mark. A defendant whose mark came from a founder's surname, a descriptive feature of the product, a coined term generated in a branding exercise, or any source unconnected to the plaintiff has gone a long way toward negating an inference of free-riding. The opposite fact pattern -- a mark that imitates a famous brand, paired with similar trade dress and overlapping goods -- invites the opposite inference all on its own.

The second is investigation before adoption -- the clearance search, detailed below. A business that pays to hunt for conflicting marks before committing to a name is, almost by definition, trying to avoid a collision, not engineer one.

The third is seeking and following professional advice -- the opinion of counsel, also below. Consulting a qualified trademark lawyer, receiving a reasoned opinion, and proceeding in reliance on it is close to the antithesis of conscious wrongdoing.

The fourth is reasonable conduct after notice. Good faith is not frozen at the instant of adoption; it is evaluated across the whole course of the defendant's conduct. A business that promptly investigates and reassesses when a credible claim arrives -- and rebrands if the claim has merit -- preserves its posture. A business that receives a well-supported cease-and-desist letter and shrugs may forfeit the protection a clean adoption would otherwise have earned, at least as to the period after notice.

These four threads are woven through the case law, and one decision braids them better than any other. The Seventh Circuit's opinion in Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947 (7th Cir. 1992), is the touchstone. Quaker, having acquired Gatorade, ran a national campaign built around the slogan "Gatorade is Thirst Aid." A small Vermont company, Sands, Taylor & Wood, owned the registered mark THIRST-AID for beverages and sued. The district court found infringement, pinned a bad-faith label on Quaker, and awarded ten percent of Gatorade's profits -- over twenty-four million dollars -- plus fees and interest. The Seventh Circuit affirmed liability but was openly skeptical of the bad-faith finding, remarking that the "evidence of bad faith in this case . . . is pretty slim." Quaker's in-house counsel had concluded that "Thirst Aid" was being used descriptively -- to say the drink quenches thirst -- rather than as a source identifier, a reading later confirmed by outside counsel. The court announced the principle that has echoed through trademark law ever since: "a party who acts in reasonable reliance on the advice of counsel regarding a close question of trademark law generally does not act in bad faith." It reversed the profits award as inequitable and remanded for a damages calculation built around a reasonable royalty rather than disgorgement. Sands, Taylor & Wood Co. v. Quaker Oats Co., 34 F.3d 1340 (7th Cir. 1994).

The "close question" qualifier deserves emphasis, because it is where the doctrine has teeth. Trademark disputes routinely involve genuinely debatable issues of similarity and relatedness on which reasonable lawyers disagree. When the legal question is close, a defendant's reliance on one reasonable answer should not be branded bad faith. When the question is not close -- when the defendant copied a famous mark for nearly identical goods -- reliance on counsel buys far less, and a court may treat a permissive opinion as proof that something was wrong rather than proof of good faith.

The Clearance Search as the First Layer of the Shield

A trademark clearance search is the structured investigation a business runs, before committing to a name, to determine whether a mark is available to use and to register without infringing someone else's rights. The process usually proceeds in two stages. A preliminary or "knockout" search scans the USPTO register and a few obvious sources to discard clearly unavailable candidates quickly and cheaply. If a candidate survives, the business commissions a full or comprehensive search -- typically through a specialized commercial search vendor -- sweeping federal registrations and applications, all fifty states' trademark registers, common-law sources (trade directories, industry publications, the open web), domain registrations, and business-name databases. A trademark attorney then analyzes the report, applies the likelihood-of-confusion factors to each potential conflict, and renders an availability assessment. Our companion piece, how to conduct a comprehensive trademark clearance search, walks through each step.

The evidentiary value of having done this work is considerable. The very act of clearing a mark before adoption is powerful circumstantial proof that the defendant was trying to avoid a conflict rather than create one. A bad-faith actor bent on hijacking an established brand does not usually begin by hiring a lawyer to hunt down the exact mark it means to misappropriate. Courts and commentators recognize that clearance searching is precisely the kind of low-cost, socially useful behavior the law should encourage when it weighs intent, because it generates exactly the contemporaneous evidence that distinguishes the careful adopter from the calculating copyist.

What matters, though, is not merely that a search was run but what it found and what the business did about it. A search that quietly omits the plaintiff's class of goods, that was deliberately narrowed to keep an obvious conflict from surfacing, or that the business ran and then ignored, offers little protection and can backfire spectacularly -- it proves the business had the information and chose to disregard it. The most protective record is one in which the search was genuinely comprehensive, the conflicts it surfaced were squarely analyzed, and the decision to proceed rested on that analysis. (When the search instead reveals a real conflict, the business may need to design around it, narrow its goods, or pick a new name -- the kind of freedom-to-operate calculus product teams know from the patent world.)

A handful of recurring mistakes routinely gut the protective value of a search, and counsel should guard against each:

  • Searching too narrowly. A search limited to identical marks in a single international class misses confusingly similar variants and related goods. Confusion does not require identity; phonetic, visual, and connotative similarity all count, as do closely related goods and services.
  • Skipping common-law and state sources. Federal registration is not a prerequisite to trademark rights -- rights arise from use. A search confined to the federal register can miss a senior common-law user whose rights, though geographically limited, are senior and enforceable. This omission is also what later defeats a Tea Rose-Rectanus good-faith claim, discussed below.
  • Failing to clear for the actual goods and the actual geography. A mark cleared for one product line or one region may not be clear for an expanded use later.
  • Treating the search as a formality. A search that is run, filed, and never analyzed is worse than useless.

The Opinion of Counsel: The Second Layer, and the Privilege Tradeoff

A clearance search gathers facts; an opinion of counsel -- often called a clearance or availability opinion -- supplies the legal judgment. In it, a qualified trademark attorney analyzes the search results, applies the likelihood-of-confusion factors to the meaningful conflicts, and reaches a reasoned conclusion about whether the proposed mark can be adopted, used, and registered at an acceptable level of risk. A good opinion is candid: it identifies the closest conflicts, acknowledges the risks honestly, explains why those risks are or are not acceptable, and, where appropriate, recommends mitigations -- narrowing the goods description, tweaking the mark, or choosing a different candidate.

For all the reasons Sands, Taylor & Wood articulated, a reasoned opinion the business actually relied on is among the strongest available evidence of good faith. Bad faith is, at bottom, conscious disregard of another's rights. A business that spots a potential conflict, asks a qualified lawyer to evaluate it, receives an opinion that no infringement exists, and proceeds in reliance is doing the opposite of disregarding the senior user -- it is investigating and respecting that user's rights. Even if the opinion later proves wrong, reliance on it negates the inference that the business meant to trade on the plaintiff's goodwill.

Not every opinion earns this credit. Courts probe the quality of the advice and the reality of the reliance, and several conditions recur:

  • It must come from qualified, ideally independent, counsel. An opinion from a non-lawyer, or from a generalist with no trademark experience, carries little weight. Outside counsel usually counts for more than in-house counsel, though Sands, Taylor & Wood shows in-house opinions can count too.
  • It must actually engage the relevant marks and issues. A boilerplate letter that never analyzes the senior mark or the specific confusion factors offers thin protection.
  • It must be reasonable, not advocacy. Courts distinguish genuine analysis from a results-driven letter manufactured to paper over conduct the lawyer knew or should have known was problematic. An opinion that finds no risk whatsoever, ignores obvious similarities, or misstates the law can wound a defendant rather than help it.
  • The defendant must have genuinely relied on it. Getting an opinion and then doing the opposite of what it advised, or commissioning it only after the decision to proceed was already made, undercuts the reliance the doctrine rewards. Reliance is best memorialized contemporaneously -- in board minutes, an internal memo, or an email reflecting that the go-ahead rested on counsel's conclusion.

Here the practitioner must confront a genuine strategic tradeoff: asserting reliance on counsel waives the attorney-client privilege over the subject matter of that advice. An opinion letter and the analysis behind it are ordinarily protected by the privilege and, in part, by the work-product doctrine. A defendant cannot wield the opinion as a sword to prove good faith while keeping it sheathed as a privileged secret. By placing reliance on counsel's advice in issue, the defendant waives privilege -- not only over the favorable opinion but over related communications on the same subject. The plaintiff may then discover what the lawyer was told, what the lawyer said, what facts the lawyer was given or denied, the drafts and earlier versions, and the follow-up communications. If those materials show the opinion rested on a skewed or incomplete record -- that the business hid an inconvenient fact from its lawyer, or shopped for a favorable answer after getting an unfavorable one -- the supposed shield becomes a magnet for the worst evidence in the case.

This waiver dynamic is most fully developed in patent law, where the advice-of-counsel defense to willful infringement has been litigated for decades; the same logic governs trademark cases. (Our comprehensive guide to patent infringement litigation traces the parallel rules.) The practical consequences are several. A defendant should decide early whether it will rely on counsel and should preserve a "clean" opinion file, knowing it may become discoverable. Many sophisticated parties deliberately separate opinion counsel -- the lawyer who renders the clearance opinion -- from litigation counsel, so that asserting the defense waives privilege over the clearance work without exposing trial strategy. And no one should assume an opinion can be produced selectively; once reliance is asserted, the scope of waiver is the court's to define and is often broad. The defense is potent, but it is not free, and the choice to invoke it should be made with eyes open.

USPTO Examination and Registration: The Third Layer

The third element of the shield comes not from the defendant or its lawyers but from a disinterested arm of the federal government. When an applicant files with the United States Patent and Trademark Office, an examining attorney reviews the application for the statutory requirements of registration. One of the examiner's principal jobs is to search the register for conflicting marks and to refuse registration under Section 2(d) of the Lanham Act, 15 U.S.C. 1052(d), if the applied-for mark "so resembles" a registered or prior-pending mark "as to be likely . . . to cause confusion." If the examiner finds a conflict, the examiner issues an Office Action -- a written refusal explaining the basis -- and the applicant may respond with argument and evidence. Persuade the examiner and the refusal is withdrawn; fail and the refusal becomes final, with an appeal available to the Trademark Trial and Appeal Board (TTAB). If no conflict is found, the application is approved for publication and, absent a successful opposition, proceeds to registration. (For the mechanics, see how to file a trademark application with the USPTO and the federal trademark application checklists.)

The scenario most favorable to a defendant is one where the examiner was aware of the senior user's registration and approved the application anyway. This happens constantly when parties operate in different industries, because the trademark system not only tolerates but embraces the coexistence of identical marks for unrelated goods. DOVE soap and DOVE chocolate; DELTA faucets and DELTA airlines: these pairs coexist because no reasonable consumer thinks a soap maker also makes candy. When an examiner reviews an application, sees the prior registration, and concludes confusion is unlikely because the goods are unrelated, that determination is meaningful evidence that the applicant's adoption was reasonable and made in good faith -- an independent governmental validation of the same conclusion the clearance search and the opinion reached.

That said, the weight courts give USPTO determinations in later infringement litigation varies, and a realistic litigant calibrates expectations. Some courts treat examiner approval as significant evidence, reasoning that examining attorneys assess confusion daily and their institutional judgment deserves respect. Others give it little weight, for an understandable reason: examination is ex parte. The senior user does not participate; the examiner sees no marketplace evidence, no consumer surveys, no testimony about actual confusion -- only the application and the register. That makes examination a comparatively "low-level" determination that may not capture the fuller picture litigation produces. Most courts land in between, treating registration as one supporting factor for good faith rather than proof of non-infringement. The evidentiary value is highest where the file wrapper shows the examiner specifically considered and rejected a Section 2(d) conflict -- identified the senior mark, analyzed the relationship of the goods, and concluded confusion was unlikely -- and lowest where the examiner simply never found the senior mark at all. This is a practical reason to make a substantive record at the PTO: if an Office Action cites a prior mark, respond on the merits and distinguish it, so the file reflects the examiner's considered conclusion rather than a conflict the examiner overlooked.

Federal registration carries a second benefit beyond evidence of good faith. Under 15 U.S.C. 1057(b), a certificate of registration is prima facie evidence of the mark's validity, of the registrant's ownership, and of the registrant's exclusive right to use the mark for the goods specified. After five years of continuous use, a registration may become incontestable under 15 U.S.C. 1065, hardening those presumptions further. While these presumptions chiefly arm plaintiffs enforcing their own marks, a defendant that holds its own registration can also point to it: the government examined the mark, found it entitled to protection, and necessarily found no bar from prior registrations. That is not a defense to infringement -- two registered marks can still be found to collide -- but it reinforces the good-faith narrative.

The Tea Rose-Rectanus Doctrine: Good Faith and the Junior User's Territory

Good faith does more than mitigate remedies; in one important doctrine it secures substantive rights. The Tea Rose-Rectanus doctrine -- named for the companion Supreme Court decisions Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916) (the "Tea Rose" case), and United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918) -- governs disputes between two parties using the same or a similar mark in different geographic areas, where neither holds a federal registration conferring nationwide constructive priority.

The doctrine flows from the nature of common-law trademark rights, which arise from use in commerce, not registration, and which extend only to the markets where the mark has actually been used (plus a reasonable zone of natural expansion). A good-faith junior user -- a business that adopts a mark in a remote area without knowledge of a senior user's prior use elsewhere -- can acquire its own enforceable common-law rights in its own territory. The senior user, never having used the mark in that remote area, cannot displace the junior user there. Two innocent parties can thus coexist, each owning the mark in its own region.

The hinge of the doctrine is good faith, which here means lack of knowledge of the senior user. A junior user who adopts a mark knowing of the senior user's prior use is not in good faith, gets no protected territory, and must yield. This is precisely where clearance discipline pays off: a junior user who actually ran a diligent search and genuinely found nothing has a strong claim to good-faith, knowledge-free adoption -- while one who skipped the common-law and state sources has handed the senior user an argument that the conflict was discoverable. The Ninth Circuit's decision in Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017), drives the point home: the junior user there had actual knowledge of the senior user's mark -- it had copied the senior user's logo from the senior user's own materials -- and that knowledge destroyed any good-faith defense. We analyze the case in depth in Stone Creek v. Omnia: knowledge destroys good faith under the Tea Rose-Rectanus doctrine.

Two refinements matter. First, federal registration changes everything: under 15 U.S.C. 1072, registration provides constructive notice of the registrant's claim, so a party that adopts a mark after the senior user has federally registered it cannot claim to be an innocent, knowledge-free junior user in any territory -- the registration puts the world on notice as a matter of law. This is one of the most valuable practical benefits of registering, and a reason to register early. Second, the "good faith" the doctrine demands is sometimes framed as more than mere ignorance; some courts ask whether the junior user adopted with an intent to benefit from the senior user's reputation. But the dominant inquiry remains knowledge: a junior user who knew of the senior mark cannot hide behind a clearance search it never truly relied on.

The Limits of the Shield: Willful Blindness, Post-Notice Conduct, and Equitable Defenses

A shield that protected everyone would protect no one, and the good-faith defense has real limits.

The most important is willful blindness, which prevents a business from manufacturing good faith by deliberately avoiding knowledge it suspects exists. Imported into intellectual-property law through the Supreme Court's decision in Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011), and applied in the trademark context through Tiffany (NJ) Inc. v. eBay Inc., 600 F.3d 93 (2d Cir. 2010), the doctrine holds that a party who strongly suspects a fact and deliberately avoids confirming it is treated as if it knew. A business cannot insulate itself by ordering a search, learning of a glaring conflict, and then telling counsel to look no further -- or by structuring the clearance process so the inconvenient mark never surfaces. Such contrivances are the antithesis of good faith. The law rewards genuine diligence, not the appearance of it staged for a future jury. Willful blindness also means an opinion obtained on a deliberately incomplete factual record provides no shelter: the gaps, once exposed through the privilege waiver discussed above, convert the opinion from a shield into a confession.

The second great limit is post-notice conduct. Good faith is assessed across the entire course of behavior, not frozen at adoption. A pristine clearance record does not license a business to ignore a credible cease-and-desist letter. When a senior user delivers notice -- especially notice backed by evidence not available during the original clearance, such as proof of actual confusion or evidence of the mark's fame -- the junior user is expected to take it seriously and reassess. A defendant that promptly investigates, obtains updated advice, and rebrands if the claim has merit preserves its good-faith posture. A defendant that doubles down may be found to have infringed willfully as to the post-notice period, even if its initial adoption was blameless. The Sands, Taylor & Wood litigation itself illustrates the dynamic: the reasonable reliance on counsel insulated the original adoption, while it was the conduct after the dispute crystallized that drove the remedy fight.

This does not mean a business must capitulate to every demand letter; many cease-and-desist letters overreach, and some are outright bullying. If counsel reviews the new claim and reasonably concludes it lacks merit, the business may justifiably continue -- but that conclusion should itself be documented in a reasoned written analysis, because it will be scrutinized later. The guiding principle is reasonableness in light of what the defendant knew and when. For how to read, weigh, and answer such letters, see our guides to drafting a trademark cease-and-desist letter (which shows what a credible demand looks like from the sender's side) and responding to a trademark cease-and-desist letter.

The third limit is the flip side worth knowing: the plaintiff's own conduct can supply equitable defenses that cut against the very remedies willfulness would otherwise unlock. A plaintiff that sat on its rights may face laches -- unreasonable delay plus prejudice to the defendant -- which can bar or shrink both monetary and injunctive relief. The catch is symmetrical: laches commonly fails where the likelihood of confusion is strong or the infringement was intentional, so a willful infringer cannot lean on the plaintiff's delay the way a good-faith one can. Acquiescence, waiver, and estoppel operate similarly. We treat this family of doctrines in understanding equitable defenses. The lesson for our purposes is that good faith and the plaintiff's delay reinforce one another: the good-faith defendant who also benefited from a slow-moving plaintiff is in the strongest position of all, while the willful infringer forfeits both arguments at once.

Three final caveats complete the picture. First, none of this defeats liability. Good faith mitigates remedies; it does not defeat likelihood of confusion. A good-faith infringer can still be enjoined and forced to rebrand. Second, good faith does not always eliminate compensatory relief; a good-faith infringer may still owe a reasonable royalty or actual damages, even as it escapes disgorgement and trebling. Third, the quality of the evidence governs its force. A perfunctory search, a results-driven opinion, or a registration the examiner granted without ever seeing the senior mark provides far less protection than thorough, candid, well-documented work.

A Worked Example: Lumen Analytics and the Northwind Conflict

Consider a hypothetical that ties the doctrine together. (It is entirely invented; no real company is intended.) Lumen Analytics, Inc. is a startup building data-visualization software. Before adopting the name LUMEN, its founders engage trademark counsel, who runs a knockout search and then commissions a full clearance search from a commercial vendor. The report surfaces a registered mark, LUMEN, owned by Northwind Lighting Co., a regional maker of LED light fixtures, registered for "electric lighting fixtures" in International Class 11.

Counsel prepares a written opinion. Applying the confusion factors, counsel concludes that although the marks are identical in spelling and sound, the goods are unrelated (enterprise software versus light fixtures), the channels of trade do not overlap (Northwind sells through electrical distributors; Lumen sells SaaS subscriptions to corporate IT departments), and the relevant purchasers are sophisticated. Counsel opines that the question is "reasonably close on the identical-marks factor but, on balance, confusion is unlikely," recommends a narrowly drawn goods description limited to "downloadable and cloud-based data-analytics software," and advises Lumen to monitor Northwind for any move into software. Lumen's board approves the name in reliance on the opinion, and the minutes say so.

Lumen files. The USPTO examining attorney identifies the Northwind registration, weighs a Section 2(d) refusal, but approves the application without issuing one, reasoning the goods are unrelated. The mark registers. Lumen builds a successful business over four years.

Then Northwind, having pivoted into smart-building software that competes with Lumen's analytics product, sends a cease-and-desist letter alleging willful infringement and demanding Lumen's profits, treble damages, and fees. How does Lumen fare?

On liability, the answer is genuinely uncertain. Northwind's entry into software narrows the gap between the goods and could support a finding of likely confusion that did not exist at adoption. Lumen may lose on confusion. But on remedies, Lumen is in a commanding position. It conducted a comprehensive clearance search before adoption; it obtained a reasoned, candid opinion that engaged the closest conflict head-on; its board actually relied on that opinion; and the examiner specifically considered the Northwind registration and approved anyway. That record forecloses every standard bad-faith argument: Lumen cannot be accused of failing to investigate, of ignoring an obvious conflict, of proceeding without advice, or of hiding its mark from the PTO. Running the equitable-disgorgement factors above, Lumen is not the unjustly enriched bad actor the remedy targets; deterrence cuts weakly; and the case on the merits was close at adoption. Invoking Romag and Justice Sotomayor's concurrence, Lumen argues that disgorging its profits would not be "consonant with the principles of equity," that any monetary exposure should be capped at a reasonable royalty or corrective-advertising costs, and that the case is not "exceptional" for fee purposes.

The flashpoint becomes Lumen's post-notice conduct. If, on receiving Northwind's letter, Lumen promptly returns to counsel, evaluates whether Northwind's market entry now creates confusion, and either secures a reasonable updated opinion or begins a rebranding plan, it preserves its good faith throughout. If instead Lumen ignores the letter and aggressively pushes the LUMEN brand into Northwind's new territory, it risks a finding of willful infringement as to the post-notice period -- and the very disgorgement it had earned the right to resist. To assert reliance on counsel, Lumen will waive privilege over its clearance opinion file; because that opinion was candid and its facts complete, the waiver helps rather than hurts. The example shows the shield working as designed. It does not guarantee a win on liability, but it can be the difference between a manageable rebrand and a company-ending judgment.

Best Practices: Building the Record Before You Need It

The time to build a good-faith record is before adoption, not after a demand letter lands. The following program, drawn from the doctrine above and from established clearance guidance, gives counsel and clients a concrete checklist -- not a substitute for tailored advice.

  • Clear before you commit. Run a knockout search early to drop obviously unavailable candidates, then commission a full search -- federal, state, common law, domain, and business-name sources -- for any serious candidate. Document the scope and preserve the report.
  • Get a real opinion, and make it candid. Have qualified trademark counsel analyze the meaningful conflicts under the confusion factors and render a written opinion that names risks honestly. A balanced opinion is more credible, and more protective, than a clean bill of health that ignores obvious problems.
  • Document genuine reliance. Memorialize the go-ahead in board minutes or an internal memo that references counsel's conclusion, so the reliance is contemporaneous and provable.
  • Draw the goods description with care. Narrow, accurate identifications avoid manufacturing conflicts with marks that pose no real-world problem and shrink the surface area for a Section 2(d) refusal.
  • Make a clean record at the USPTO. If the examiner cites a prior mark, respond substantively, distinguishing the marks and goods, so the file wrapper reflects a considered conclusion rather than a missed conflict.
  • Register early, and where it matters. Federal registration provides constructive notice under 15 U.S.C. 1072 and the validity presumptions of 15 U.S.C. 1057(b), and it forecloses a later adopter's Tea Rose-Rectanus good-faith claim.
  • Plan the privilege decision in advance. Decide whether you may later rely on counsel, keep the opinion file clean and complete, and consider separating opinion counsel from litigation counsel so that asserting the defense does not expose trial strategy.
  • Respond reasonably to notice. When a credible claim arrives, investigate promptly, obtain updated advice, document any reasoned decision to continue, and rebrand without delay if the claim has merit. Good faith at adoption is squandered by stubbornness afterward.
  • Avoid willful blindness. Never narrow a search, or instruct counsel to stop looking, to dodge a conflict you already suspect. The law rewards genuine diligence, not its appearance.

For the operational detail behind several of these steps, see how to conduct a comprehensive trademark clearance search and the federal trademark application checklists.

Key Takeaways

Willful trademark infringement is about state of mind, and state of mind controls the remedies that hurt: disgorgement of profits, treble damages under 15 U.S.C. 1117(a), and attorney's fees in "exceptional" cases. After Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), willfulness is no longer an absolute precondition to disgorging profits, but it remains the most important equitable factor courts weigh -- which means a defendant must be ready to prove good faith affirmatively rather than rely on the plaintiff's failure to disprove it.

Three layers of pre-adoption diligence build that proof. A comprehensive clearance search shows the business tried to avoid a conflict. A candid opinion of counsel, on which the business genuinely relied, shows it sought and respected professional guidance on a close question -- though asserting reliance waives the attorney-client privilege, so the opinion file must be clean and complete. And USPTO examination and registration -- especially where the examiner saw the senior mark and approved anyway -- supply independent governmental validation, with the bonus of the registration's validity presumptions and constructive notice. In the geographic-conflict setting, good faith does even more under the Tea Rose-Rectanus doctrine, where a knowledge-free junior user can secure protected territory of its own.

The shield has limits. It does not defeat liability; a good-faith infringer can still be enjoined and forced to rebrand. It does not survive willful blindness or a deliberately incomplete record. And it can be lost going forward by ignoring a credible cease-and-desist letter. But for the careful business, the modest cost of clearing a mark properly is the best available insurance against the catastrophic remedies that willful infringement unlocks -- often the difference between a manageable rebrand and a ruinous judgment.

Frequently Asked Questions

Does a clearance search and an attorney opinion mean I cannot be sued for infringement? No. They do not prevent a lawsuit and do not defeat liability. Likelihood of confusion is an objective inquiry; a good-faith adopter can still be found to infringe and ordered to stop using the mark. What this diligence does is establish good faith, the dominant factor in whether a court awards the painful monetary remedies -- the infringer's profits, treble damages, and attorney's fees.

After Romag, is willfulness still worth worrying about? Yes, more than ever. Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), held only that willfulness is not an absolute precondition to disgorging profits in an infringement case. The Court and the concurrences stressed that mental state remains "highly important," and that awarding profits against a good-faith infringer would not be "consonant with the principles of equity." Because the plaintiff no longer must prove willfulness as a threshold, the practical burden has shifted: defendants should be ready to prove good faith with documents.

If disgorgement is off the table, what might a good-faith infringer still owe? Usually one of the compensatory measures: a reasonable royalty for the unauthorized use, or the cost of corrective advertising to dispel confusion. See Sands, Taylor & Wood Co. v. Quaker Oats Co., 34 F.3d 1340 (7th Cir. 1994); Big O Tire Dealers, Inc. v. Goodyear Tire & Rubber Co., 408 F. Supp. 1219 (D. Colo. 1976). And almost always an injunction to rebrand. These figures are typically a fraction of the profits demand in the original cease-and-desist letter.

If I rely on my lawyer's opinion as a defense, do I have to turn it over to the other side? Generally yes. Asserting reliance on counsel waives the attorney-client privilege over that advice and related communications on the same subject. The opposing party can then discover what your lawyer was told, what the lawyer said, the drafts, and the follow-ups. This is why the opinion should rest on a complete and accurate factual record, and why many parties separate opinion counsel from litigation counsel.

How much weight will a court give to the USPTO's approval of my mark? It varies. Some courts treat examiner approval as significant evidence of good faith; others give it little weight because examination is ex parte and based on a limited record. The weight is highest when the examiner specifically identified the senior mark, analyzed the goods, and concluded confusion was unlikely -- and lowest when the examiner simply never found the senior mark. In all events, registration also confers the validity presumptions of 15 U.S.C. 1057(b) and constructive notice under 15 U.S.C. 1072.

What is the Tea Rose-Rectanus doctrine, and how does good faith fit in? It is the rule, from Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916), and United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918), that a good-faith junior user who adopts a mark in a remote area without knowledge of a senior user's prior use can acquire enforceable common-law rights in that area. Good faith here means lack of knowledge; a junior user who knew of the senior mark gets no protected territory, as Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017), illustrates. Federal registration cuts off the defense by providing constructive notice.

Can I ignore a cease-and-desist letter if my original clearance was thorough? No. Good faith is judged across your entire course of conduct. Continuing to use a mark after credible notice, without investigating or getting updated advice, can support a finding of willful infringement for the post-notice period even if your adoption was blameless. Investigate promptly, get updated advice, document any reasoned decision to continue, and rebrand if the claim has merit. See responding to a trademark cease-and-desist letter.

Related Articles

Selected Authorities

Cases. Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020); Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947 (7th Cir. 1992), and 34 F.3d 1340 (7th Cir. 1994); Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014); Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916); United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918); Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017); Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011); Tiffany (NJ) Inc. v. eBay Inc., 600 F.3d 93 (2d Cir. 2010); Big O Tire Dealers, Inc. v. Goodyear Tire & Rubber Co., 408 F. Supp. 1219 (D. Colo. 1976); Century 21 Real Estate Corp. v. Sandlin, 846 F.2d 1175 (9th Cir. 1988); In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973); Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492 (2d Cir. 1961).

Statutes. Lanham Act Section 2(d), 15 U.S.C. 1052(d) (likelihood-of-confusion bar to registration); Section 7(b), 15 U.S.C. 1057(b) (presumption of validity); Section 15, 15 U.S.C. 1065 (incontestability); Section 22, 15 U.S.C. 1072 (constructive notice); Section 32, 15 U.S.C. 1114 (registered-mark infringement); Section 34, 15 U.S.C. 1116 (injunctive relief); Section 35, 15 U.S.C. 1117(a)-(c) (remedies, profits, treble damages, fees, counterfeiting, statutory damages); Section 43(a), (c), 15 U.S.C. 1125(a), (c) (unfair competition; dilution).

Secondary. J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition (5th ed.); U.S. Patent and Trademark Office, Trademark Manual of Examining Procedure (TMEP); Practical Law, Trademark Infringement and Dilution Claims, Remedies, and Defenses; Practical Law, Trademark Infringement Profit and Damage Awards Standards by Circuit Chart.

This article is provided by mclaw.io for general informational purposes only. It is not legal advice, it does not create an attorney-client relationship, and it may not reflect the most current legal developments in every jurisdiction. Trademark disputes turn on specific facts and on the law of the relevant circuit. Readers facing an actual or threatened trademark dispute, or adopting a new mark, should consult qualified trademark counsel about their particular circumstances.