A trademark case is, at bottom, a fight about a single sentence: consumers are likely to be confused. Everything else—the discovery, the depositions, the dueling survey experts, the trial—exists to prove or disprove that one proposition. And in the Second Circuit, the place where that proposition most often lives or dies short of a verdict is summary judgment.

Here is the paradox that makes the topic worth a guide. The Second Circuit has said, again and again, that likelihood of confusion is a fact-intensive question ill-suited to resolution on a paper record. In the same breath, it has affirmed summary judgment for defendants in case after case, holding that no reasonable jury could find confusion likely. Both statements are true. The art of litigating these motions—and the reason a smart movant wins and a careless one gets reversed—lies in understanding which factors a court can responsibly resolve without a trial and which it generally cannot. That is the subject of this article.

We approach the question the way a litigator does: factor by factor, with attention to how each one behaves under Rule 56, illustrated by the cases the Second Circuit actually cites. If you want the broader doctrinal picture of how the confusion test operates outside the summary judgment posture, start with our overview, Navigating the Maze of Trademark Confusion. This piece narrows the lens to one high-stakes procedural moment and stays there.

Why summary judgment is the battleground

Most trademark infringement claims rise or fall on likelihood of confusion. It is the touchstone of every infringement claim under the Lanham Act, whether the asserted mark is federally registered under Section 32 (15 U.S.C. § 1114) or protected as an unregistered mark under Section 43(a) (15 U.S.C. § 1125(a)). It is also the test for many trade dress and false-association claims. If you are new to the underlying vocabulary—marks, distinctiveness, infringement—our Trademark Basics primer and our overview of infringement and related rights lay the groundwork.

Summary judgment matters enormously because it can end a case—or a discrete claim—before trial, sparing the parties the cost, delay, and unpredictability of putting the question to a jury. For a defendant, a ruling of no likelihood of confusion as a matter of law is a complete victory on the infringement claim. For a plaintiff, a ruling that confusion is so plain that no reasonable jury could find otherwise is equally decisive (though, as we will see, that direction is considerably rarer). And because the Second Circuit—covering New York, Connecticut, and Vermont, and home to the Southern and Eastern Districts of New York—handles an outsized share of the nation's trademark litigation, its approach to summary judgment on the Polaroid factors is among the most consequential bodies of trademark law in the country.

There is also a strategic reason summary judgment looms so large. As Thomson Reuters' Practical Law observes, courts "generally disfavor summary judgment in trademark cases because these cases often involve genuine factual disputes," yet the motion remains "a useful tool to resolve trademark claims, defenses, or issues when the facts cannot reasonably be disputed." A defendant who files and wins gets out of the case. A defendant who files and loses has shown the plaintiff its full hand, taught the court the plaintiff's strongest evidence, and sometimes handed the plaintiff a favorable ruling on the factors it does win. The decision to move is itself a piece of trial strategy, not a routine box to check.

The test: eight factors from a 1961 camera case

The framework comes from Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492 (2d Cir. 1961), an opinion by the legendary Judge Henry Friendly. The dispute pitted Polaroid—then synonymous with cameras, film, and optics—against Polarad, a maker of electronic and microwave equipment. The products did not directly compete, which posed a problem the older case law had not cleanly solved: when may the owner of a strong mark stop a newcomer from using a similar mark on different goods? Judge Friendly answered by setting out a list of considerations for evaluating the likelihood of confusion in exactly that situation. The court ultimately resolved the case on laches grounds, so the famous list was, strictly speaking, a roadmap for the remand—but that list has governed the Second Circuit ever since.

The eight Polaroid factors, as the Second Circuit catalogs them, are:

  1. the strength of the senior user's mark;
  2. the similarity of the two marks;
  3. the competitive proximity of the products or services in the marketplace;
  4. the likelihood that the senior user will bridge the gap between the parties' goods;
  5. evidence of actual confusion;
  6. the junior user's intent (good faith or bad faith) in adopting its mark;
  7. the quality of the junior user's goods; and
  8. the sophistication of the relevant consumers.

(Polaroid, 287 F.2d at 495.) Courts list them in slightly different orders and occasionally fold in related considerations, so the exact numbering matters less than the substance. Two structural points shape the entire summary judgment analysis.

First, the factors are a guide, not a scorecard. A court does not tally how many factors favor each side and crown the higher number. It weighs them holistically, assigning each whatever weight the facts warrant, always in service of the ultimate question: are ordinary prudent purchasers likely to be confused? The Second Circuit has been emphatic that the inquiry is not "a mechanical process where the party with the most factors wins"; the court must "focus on the ultimate question of whether consumers are likely to be confused." Guthrie Healthcare Sys. v. ContextMedia, Inc., 826 F.3d 27, 37 (2d Cir. 2016); Playtex Prods., Inc. v. Georgia-Pacific Corp., 390 F.3d 158, 162 (2d Cir. 2004). The factors "serve only as a guide and each case presents its own unique circumstances." Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 872 (2d Cir. 1986). And while the list is non-exhaustive and "generally no one factor is dispositive," a court must still address each relevant factor and explain any it deems inapplicable. Int'l Info. Sys. Sec. Certification Consortium, Inc. v. Sec. Univ., LLC, 823 F.3d 153, 160 (2d Cir. 2016).

Second—and this is the single most important thing to understand about Polaroid on summary judgment—because the inquiry is holistic, a defendant can win even if some factors favor the plaintiff, provided the factors that matter most in the particular case point decisively toward no confusion. The Second Circuit treats three factors as often the most important: the strength of the mark, the similarity of the marks, and the competitive proximity of the goods. Mobil Oil Corp. v. Pegasus Petroleum Corp., 818 F.2d 254, 258 (2d Cir. 1987). When those three line up for the defendant, the rest of the balance tends to follow—and the case becomes a candidate for judgment as a matter of law.

The summary judgment standard, mapped onto a fact-intensive test

Summary judgment is governed by Federal Rule of Civil Procedure 56. A court grants it when "there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." Fed. R. Civ. P. 56(a). The Supreme Court's trilogy supplies the operating instructions: a dispute is "genuine" only if a reasonable jury could return a verdict for the non-movant, Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986); the movant bears the initial burden of showing the absence of a genuine dispute, Celotex Corp. v. Catrett, 477 U.S. 317, 322–23 (1986); and the court views the evidence in the light most favorable to the non-movant, drawing all reasonable inferences in its favor, without weighing evidence or making credibility findings, Anderson, 477 U.S. at 255.

Critically, the non-movant cannot survive on assertion alone. It must come forward with evidence "showing more than just a metaphysical doubt" and "may not rely on mere conclusory, self-serving allegations to defeat summary judgment." A "genuine dispute over an immaterial fact" does not block the motion, and neither does a "non-genuine dispute where no reasonable jury could find for the non-moving party." Anderson, 477 U.S. at 247–52. Those principles—familiar from every corner of federal practice—collide, in the trademark context, with a test that is openly described as fact-laden. The collision is exactly what makes this area subtle.

The Second Circuit reconciles the two in a way worth stating with care, because getting it wrong is how district courts get reversed. The court's framing, drawn from Cadbury Beverages, Inc. v. Cott Corp., 73 F.3d 474 (2d Cir. 1996), and reaffirmed in Guthrie and Malletier v. Burlington Coat Factory Warehouse Corp., 426 F.3d 532, 537 (2d Cir. 2005), is that the analysis is a mixed question: the court's findings on each individual Polaroid factor are findings of fact, but "the ultimate determination of whether confusion is likely[,] based on a weighing of the factors[,] is a question of law." Guthrie, 826 F.3d at 37–38.

That distinction does all the work. It means the court does not resolve genuine disputes about the underlying facts that feed each factor—what the marks look like in the marketplace, what the products are and to whom they are sold, whether real consumers were actually confused. But once those underlying facts are established or undisputed, the court may perform the ultimate weighing—whether confusion is likely—as a matter of law. Cadbury is the warning every plaintiff cites: there, the Second Circuit reversed a grant of summary judgment, stressing that the confusion inquiry is fact-laden and that a district court may not short-circuit it by resolving disputed factual questions or by weighing the factors as if it were the jury. A movant therefore wins not by asking the court to weigh contested evidence, but by showing that even on the plaintiff's best version of the facts, no reasonable jury could find confusion likely.

This is also why some factors are far more "summary-judgment-friendly" than others. Factors whose inputs are objective and fixed—what the marks are, what the goods are, who buys them—can often be resolved on the papers because there is nothing genuinely contested for a jury to find. Factors whose inputs are subjective or evidentiary—what was in the defendant's mind, whether real consumers were confused—are more likely to require trial if the record contains conflicting evidence, though they too can be resolved on summary judgment when one side has no evidence at all. With that framework in place, we can walk through the factors in roughly the order they tend to matter on these motions. (For how these same questions are reviewed after trial, see our companion guide, Second Circuit Appellate Standards in Trademark Cases.)

Factor 2: Similarity of the marks—often the whole ballgame

We begin with similarity, because on a defense motion it is frequently the factor that decides everything. The reason is structural: the marks are what they are. A court can place the two before it, consider them as they actually appear in the marketplace—packaging, house marks, color, font, presentation, retail setting—and assess whether they create a confusingly similar overall impression. There is usually no disputed historical fact for a jury to resolve; the comparison is something the court can perform directly.

The Second Circuit has said explicitly that this factor can be decisive. "When marks are so dissimilar that confusion is not possible, a court may find in favor of a defendant on the similarity factor alone." Car-Freshner Corp. v. Am. Covers, LLC, 980 F.3d 314, 330 (2d Cir. 2020); Nabisco, Inc. v. Warner-Lambert Co., 220 F.3d 43, 46 (2d Cir. 2000). That is a striking concession from a court that otherwise insists no single factor controls—and it tells you where to aim a defense motion.

How does the court measure similarity? Three rules drive the analysis. First, the comparison is of sight, sound, and meaning—the marks' visual appearance, their pronunciation, and their connotation. Second, the court assesses the "overall impression of the marks as they appear in the context of the marketplace," considering "all of the factors that consumers are likely to perceive and remember." Nabisco, 220 F.3d at 47; Lang v. Retirement Living Publ'g Co., 949 F.2d 576, 581 (2d Cir. 1991). Differences in "typeface, font, and general presentation may render marks dissimilar." Third—and counterintuitively for many newcomers—a clean side-by-side comparison is improper, "because consumers typically encounter marks sequentially, not at the same time." Louis Vuitton Malletier v. Dooney & Bourke, Inc., 454 F.3d 108, 117 (2d Cir. 2006). The test imagines a consumer carrying an imperfect memory of one mark into an encounter with the other, not a lab technician comparing two specimens under glass.

Star Industries, Inc. v. Bacardi & Co., 412 F.3d 373 (2d Cir. 2005), is the template defense decision and repays study. Star owned a stylized letter "O" design used for an orange-flavored vodka; it sued over Bacardi's use of an "O" in connection with an orange-flavored rum. The Second Circuit affirmed summary judgment for Bacardi, concluding that no reasonable jury could find a likelihood of confusion. Similarity did real work: the court examined the marks in their actual settings—each appearing alongside the producer's prominent house brand and distinct trade dress—and found the overall impressions different enough that confusion was implausible. The presence of the house marks mattered, because "use of a house mark with a challenged mark generally reduces the likelihood of confusion." Nabisco, 220 F.3d at 46–47. Star Industries is the recurring profile of a defense win: a weak plaintiff's mark, marks that differ in overall commercial impression, and surrounding context that further distinguishes them.

Streetwise Maps, Inc. v. VanDam, Inc., 159 F.3d 739 (2d Cir. 1998), points the same direction. There the court affirmed summary judgment for the defendant where the competing map products and their marks were sufficiently different that confusion was unlikely, looking at the marks in their entireties and in their actual commercial dress rather than dissecting them into shared fragments. The lesson is that similarity is judged by overall impression, in context—not by hunting for a common syllable or a shared descriptive word.

But similarity is not a free pass for defendants, and two doctrines run the other way. One may not "appropriate the entire mark of another and add descriptive or generic elements to avoid a likelihood of confusion." NYC Triathlon, LLC v. NYC Triathlon Club, Inc., 704 F. Supp. 2d 305, 317 (S.D.N.Y. 2010). And a house mark does not always save the junior user; in some circumstances it can aggravate confusion—for instance, by suggesting a sponsorship or licensing relationship. Playtex, 390 F.3d at 165. The practical takeaway is that context is everything. A movant should present the marks as consumers actually encounter them—full packaging, house brands, retail setting. A non-movant resists by emphasizing the shared dominant element, similar sound or meaning, and any settings (a bare word-mark use online, a point-of-sale display, a phone order) in which the distinguishing context falls away. When the marks really are close, similarity becomes a jury question; when they are plainly different in overall impression, it can carry a summary judgment motion by itself.

Factor 1: Strength of the mark—decidable, but layered

Strength measures how distinctive and recognizable the plaintiff's mark is, and thus how broad a scope of protection it deserves. It has two components, and keeping them separate is the key to predicting how the factor behaves under Rule 56.

Conceptual (or inherent) strength turns on where the mark falls on the classic distinctiveness spectrum—generic, descriptive, suggestive, arbitrary, or fanciful. Arbitrary and fanciful marks (think KODAK for film, or APPLE for computers) "generally are strong and receive a broad scope of protection." Virgin Enters. Ltd. v. Nawab, 335 F.3d 141, 147–48 (2d Cir. 2003). Descriptive marks, protectable only on a showing of secondary meaning, "generally are considered weak and receive a narrow scope of protection." TCPIP Holding Co. v. Haar Commc'ns, Inc., 244 F.3d 88, 100 (2d Cir. 2001); Star, 412 F.3d at 385. This placement is often a question courts can resolve as a matter of law. (For the mechanics of how descriptive marks gain strength, see our discussion in Trademark Overview: Substantive Standards for Protection.)

Commercial strength turns on marketplace recognition: advertising expenditures, consumer studies linking the mark to its source, unsolicited media coverage, sales success, attempts by others to plagiarize the mark, and the length and exclusivity of use. Car-Freshner, 980 F.3d at 329. This dimension is more evidentiary and occasionally fact-bound. Crucially, the two dimensions can diverge: "even arbitrary marks can be considered relatively weak if they lack commercial strength," SLY Magazine, LLC v. Weider Publ'ns L.L.C., 529 F. Supp. 2d 425, 437 (S.D.N.Y. 2007), and conversely "even descriptive marks can be strong on a showing of high consumer recognition," Sports Auth., Inc. v. Prime Hospitality Corp., 89 F.3d 955, 961 (2d Cir. 1996).

Time, Inc. v. Petersen Publishing Co., 173 F.3d 113 (2d Cir. 1999), shows strength setting the stage for a no-confusion ruling. Petersen published a magazine called Teen; Time launched Teen People and sought a declaration of non-infringement. The Second Circuit affirmed summary judgment for Time, and the descriptiveness of "Teen"—a term that merely names the audience—was central to the analysis. A conceptually weak mark is entitled to a narrow scope of protection, so even similar uses are less likely to cause confusion. When a plaintiff's mark is descriptive and commercially modest, strength tilts toward the defendant and makes the overall balance much easier to resolve without trial.

Two further wrinkles are worth flagging. First, federal registration carries a presumption of validity and distinctiveness, and an incontestable registration creates a conclusive presumption of distinctiveness as to the registered goods—though even that does not foreclose a defendant from attacking the mark's overall strength, including its commercial weakness. Savin Corp. v. Savin Grp., 391 F.3d 439, 457 (2d Cir. 2004); Lois Sportswear, 799 F.2d at 871. (We unpack those presumptions in Benefits of Federal Trademark Registration, and the related option of attacking a registration head-on in Trademark Cancellation in Federal Litigation.) Second, extensive third-party use of similar marks weakens a plaintiff's mark—but only if the defendant proves actual use and consumer exposure, not a pile of dormant registrations. "[M]ere evidence of trademark registrations is not evidence of actual use or consumer recognition." Scarves by Vera, Inc. v. Todo Imports Ltd., 544 F.2d 1167, 1173–74 (2d Cir. 1976). That evidentiary requirement is where a commercial-strength fight can become genuinely contested—and occasionally heads to a jury.

In practice, strength rarely decides a motion by itself, but a finding that the mark is weak is a powerful accelerant for the rest of the analysis. As the Second Circuit puts it, "the stronger the senior user's mark and the greater the similarity between the marks, the less competitive proximity is required" for confusion. Katz v. Modiri, 283 F. Supp. 2d 883, 896 (S.D.N.Y. 2003). The contrapositive is the defense's friend: a weak mark needs a closer copy and closer proximity before confusion becomes likely.

Factor 3: Competitive proximity—frequently resolved as a matter of law

Competitive proximity asks how close the parties' products or services are in the marketplace—whether they compete, serve the same consumers, occupy the same trade channels, and satisfy the same needs. Like the marks themselves, the products are generally fixed and identifiable, which makes this factor highly amenable to summary judgment. A court can often determine, without any genuine factual dispute, whether the parties sell competing goods to overlapping customers through similar channels, or whether they operate in genuinely separate markets.

The Second Circuit's standard list of proximity considerations includes "the nature of the goods or services and the relevant market, including the trade channels and advertisement practices," and whether the goods have similar purposes, fall in the same general class, are used together, share geographic distribution, occupy the same market position, and appeal to the same audience. Woodstock Ventures LC v. Woodstock Roots, LLC, 387 F. Supp. 3d 306, 317 (S.D.N.Y. 2019); W.W.W. Pharm. Co. v. Gillette Co., 984 F.2d 567, 573 (2d Cir. 1993). Two refinements matter on a motion. First, direct competition is not required: "competition is not necessary for confusion to be likely." Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 396 (2d Cir. 1995). The relevant question is whether consumers will think the plaintiff sponsors or is connected to the defendant's goods. That is the entire reason Polaroid exists—Judge Friendly devised the test precisely to handle non-competing goods (cameras and optics versus electronics). Second, proximity interacts with strength and similarity: a very strong, very similar mark needs less proximity to generate confusion, and vice versa.

When the products are remote—different industries, different customers, different channels—proximity points strongly away from confusion and supports summary judgment for the defendant. Consumers encountering similar marks on plainly different goods are unlikely to assume a common source. Proximity is rarely the sole basis for a ruling, but together with similarity and the treatment of actual confusion, it is one of the two or three factors that most often does the heavy lifting in a defense win. A movant establishes the separate markets, customers, price points, and trade channels through undisputed evidence; a non-movant resists by showing overlap—shared retailers, a shared customer base, adjacency on the shelf or on a search-results page, or a market trend toward convergence. Where the overlap is genuinely contested, proximity becomes a jury issue; where the markets are plainly distinct, the factor is resolved as a matter of law and frequently drives the result.

Factor 4: Bridging the gap—the quiet companion to proximity

Bridging the gap is proximity's natural companion. It asks whether the senior user is likely to expand into the junior user's market, or whether consumers would expect it to—because if so, present non-competition may give way to future competition and confusion. The factor protects the senior user's interest in preserving its natural zone of expansion. Morningside Grp. Ltd. v. Morningside Capital Grp., L.L.C., 182 F.3d 133, 141 (2d Cir. 1999). (We explore a cousin of this idea—the territorial "zone of natural expansion"—in The Geographic Scope of Common Law Trademark Rights, in a different but related setting.)

Bridging the gap is quite summary-judgment-friendly, and it usually favors the defendant when the products are remote, because of a doctrine non-movants routinely overlook. A senior user's subjective intent to expand "is irrelevant to the likelihood of confusion analysis unless it is known to prospective consumers." Lang, 949 F.2d at 582. In other words, a plaintiff cannot manufacture a fact dispute merely by submitting an affidavit swearing it might someday enter the defendant's market. The factor favors the senior user only when "the average consumer expects the senior user to enter the junior user's market," Sports Auth., 89 F.3d at 963, or when the plaintiff comes forward with publicly known, concrete expansion plans, executed contracts, or evidence that crossover in this industry is common and expected. Absent that, courts routinely resolve the factor in the defendant's favor on the papers—and when the products are already proximate, the factor drops out as largely beside the point.

Factor 5: Actual confusion—powerful when present, telling when absent

Actual confusion carries the most direct intuitive force: if real consumers have already been confused, that is strong evidence confusion is likely. "Evidence that confusion has already occurred is convincing evidence that it is likely to occur in the future." Guthrie, 826 F.3d at 44; Morningside Grp., 182 F.3d at 141. But the Second Circuit pairs that principle with an equally important caveat: actual confusion "is extremely difficult to prove and evidence of actual confusion is not necessary to prevail." Guthrie, 826 F.3d at 45. The legal standard is likelihood, not actuality. A plaintiff with no proof of actual confusion can still win; a defendant cannot prevail merely by pointing to its absence in a vacuum.

On summary judgment, this factor cuts in two directions, and both are decisive in their own circumstances.

When confusion evidence is substantial, it can defeat a defense motion outright. Credible reports of misdirected customers, lost orders, or a well-designed consumer survey showing meaningful confusion create a triable issue, because a reasonable jury could rely on that evidence to find confusion likely. Survey evidence in particular can be the difference-maker; "actual confusion can be proven by survey." Mobil Oil, 818 F.2d at 259. But surveys are double-edged swords—a methodologically flawed survey can be excluded under Daubert, and the absence of a survey, while not fatal, invites the inference that the plaintiff looked for confusion and could not find it. Survey design and admissibility is a discipline unto itself, treated in our companion guides, Consumer Survey Expert Methodology in Trademark Cases and Daubert Challenges to Consumer Survey Experts. A plaintiff who plans to oppose summary judgment with a survey must build it to survive both a Daubert motion and skeptical cross-examination.

When confusion evidence is thin, courts do not let it reach a jury. "Isolated instances of confusion constitute de minimis evidence of actual confusion, which can be insufficient to withstand summary judgment." Nora Beverages, Inc. v. Perrier Grp. of Am., Inc., 269 F.3d 114, 124 (2d Cir. 2001). In Nora Beverages, the Second Circuit affirmed summary judgment for the defendant in part because a mere handful of stray, ambiguous instances could not support an inference that ordinary consumers were likely to be confused. The court has held that even a single instance of possible confusion can leave the factor favoring the defendant. Car-Freshner, 980 F.3d at 332. And there is a subtle trap for plaintiffs: a consumer inquiry about whether two companies are affiliated "is not actual confusion and arguably shows a lack of confusion," because a person who has to ask has not been deceived. Nora Beverages, 269 F.3d at 124. (Though the line is not absolute—evidence that a defendant fielded inquiries about affiliation has sometimes been read in the plaintiff's favor. Virgin Enters., 335 F.3d at 151.)

This factor also frames one of the most powerful arguments in the defense arsenal: the inference from coexistence over time. The Second Circuit recognizes that "[a] lack of any evidence of actual confusion can show that confusion is unlikely if the parties have coexisted in the marketplace for an extended period of time." Starbucks Corp. v. Wolfe's Borough Coffee, Inc., 588 F.3d 97, 117 (2d Cir. 2009). The logic is straightforward: if two marks have circulated side by side for years in overlapping markets, giving confusion every opportunity to surface, and almost none has, a court can reasonably conclude that consumers are not in fact being confused. Combined with dissimilar marks, remote products, and a weak mark, a long, confusion-free track record can be the fact that pushes a case across the line to judgment as a matter of law.

Two cautions temper the inference, both grounded in the case law. First, the coexistence must be genuine and meaningful. "Actual confusion evidence is not expected when the parties have coexisted for only a short period of time and there has been little opportunity for confusion to occur." Guthrie, 826 F.3d at 44; TCPIP Holding, 244 F.3d at 102. A brief overlap, or coexistence in non-overlapping markets, blunts the inference because confusion never had a real chance to appear. Second, because likelihood—not actuality—is the standard, a defendant must situate the absence of confusion within a record where confusion would have been expected if the marks were truly close. The inference is a powerful supporting beam, not a load-bearing wall on its own.

Factor 6: Good faith and bad faith—a demanding evidentiary showing

Factor six examines the junior user's intent in adopting its mark—specifically, whether it acted in bad faith, meaning with the intent to exploit the senior user's goodwill and sow confusion. The relevant question is "whether the defendant attempted to exploit the senior user's goodwill by adopting a mark to cause confusion." Tiffany & Co. v. Costco Wholesale Corp., 971 F.3d 74, 88 (2d Cir. 2020). This factor draws heavy briefing on summary judgment, and understanding what bad faith actually requires is essential, because plaintiffs routinely overestimate it.

The crucial point: mere knowledge of the senior user's mark is not bad faith. "A junior user's prior knowledge of the senior user's trademark does not necessarily create an inference of bad faith." Playtex, 390 F.3d at 166; LVL XIII Brands, Inc. v. Louis Vuitton Malletier S.A., 209 F. Supp. 3d 612, 685 (S.D.N.Y. 2016). This matters decisively on a motion, because plaintiffs often have nothing more than evidence that the defendant knew of them—which, standing alone, does not create a genuine dispute on bad faith. Where the plaintiff's "bad faith" showing reduces to knowledge plus suspicion, courts resolve the factor in the defendant's favor or treat it as neutral. It is also worth recalling that good faith is not a defense at all—"[w]rongful intent is not a requirement for a trademark infringement claim, and good faith is not a defense," Sunward Elecs., Inc. v. McDonald, 362 F.3d 17, 25 (2d Cir. 2004)—so the factor cuts only one way as a practical matter: it can hurt a defendant, but its absence cannot, by itself, win the case for a plaintiff.

Defendants can also affirmatively build a good-faith record, which both neutralizes the factor and reinforces the overall no-confusion story. "A request for a trademark search and reliance on advice of counsel generally support a finding of good faith." Lang, 949 F.2d at 583. A defendant who ran a clearance search, obtained an opinion of counsel, adopted its mark for independent reasons, and deployed its own distinct house brand and trade dress has assembled exactly the kind of record that defeats an inference of intent to deceive. (We treat the protective power of clearance searches and counsel opinions in depth in The Shield of Good Faith.) Notably, the converse is not true: a failure to conduct a search "does not prove bad faith." Savin, 391 F.3d at 460.

Genuine bad-faith disputes do arise, and when they do they can defeat a defense motion, because intent is a quintessentially factual question for the jury. Evidence such as deliberate copying of distinctive features, internal communications revealing an intent to trade on the plaintiff's name, an "absence of a credible explanation" for the mark's selection in light of prior knowledge, or a suspicious adoption following failed dealings can each create a triable issue. NYC Triathlon, 704 F. Supp. 2d at 319. And intentional copying carries a doctrinal kicker: "[a] junior user's intentional copying of a senior user's mark creates a presumption that confusion is likely." Mobil Oil, 818 F.2d at 258. Once bad faith is genuinely in play, it can color an entire case and bleed into remedies—Patsy's Brand, Inc. v. I.O.B. Realty, Inc., 317 F.3d 209 (2d Cir. 2003), is a cautionary illustration of how bad-faith conduct, including misconduct in the litigation itself, can dominate an outcome. But the threshold question on summary judgment is always evidentiary: does the plaintiff have actual evidence of intent to confuse, or only evidence of awareness? If only the latter, the factor will not save the plaintiff's case—and may quietly support the defendant.

Factor 7: Quality of the junior user's goods—usually a minor note

The quality factor is unusual—it is "unique to the Second Circuit"—and it is the least probative of the eight. Virgin Enters., 335 F.3d at 151. It can cut in two directions, which is precisely why courts give it little independent weight. A "marked difference between the quality of the parties' products makes confusion less likely," while "where the quality of the parties' products is similar, confusion is more likely." Savin, 391 F.3d at 461. The intuition behind the first half is that obviously inferior knockoffs telegraph their difference; behind the second half, that comparable goods are more interchangeable and easier to mistake.

There is a deeper conceptual muddle the Second Circuit has flagged about its own factor. Courts often read quality to ask whether the plaintiff's reputation would be jeopardized by association with an inferior product—but that is arguably a question about the harm from confusion, not the likelihood of it. Savin, 391 F.3d at 460–61. Because of this ambiguity, quality rarely tips a motion. On summary judgment it is typically addressed in a paragraph, found neutral, and folded into the larger balance. Neither side should rest weight on it; the action is in similarity, proximity, strength, actual confusion, and intent.

Factor 8: Consumer sophistication—decidable where the buyers are discerning

The final factor asks how sophisticated and careful the relevant consumers are. The premise is intuitive: "[t]he more attention consumers give, the less likely they are to be confused by similar marks." Tiffany, 971 F.3d at 90; Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1046 (2d Cir. 1992). The inquiry focuses on "the attention typically given by a purchaser of the goods at issue under normal market conditions." Sports Auth., 89 F.3d at 965.

Price is the usual proxy. "Where the cost of a product is high, courts generally assume that purchasers are more discriminating," and "consumers of low priced items generally are presumed to be less discriminating." Savin, 391 F.3d at 461; Patsy's Brand, 317 F.3d at 219. Sophistication is reasonably summary-judgment-friendly when the nature of the product and its buyers is clear from the record: where the goods are costly, technical, or sold to professional purchasers who investigate before buying, courts can conclude as a matter of law that the relevant consumers are sophisticated and unlikely to be confused. Where the goods are inexpensive impulse buys, the factor favors the plaintiff. It becomes contested only where the actual buyer population or purchasing conditions are genuinely disputed.

One important limiting principle keeps this factor from rescuing a doomed defense: "[w]hen the parties' goods and services are highly similar, consumer sophistication cannot be relied on to prevent confusion." Morningside Grp., 182 F.3d at 143. Even careful buyers can be confused by nearly identical marks on nearly identical goods. So sophistication is a meaningful factor when it combines with dissimilar marks or remote products, and a weak reed when the marks and goods are close.

Which factors most often drive summary judgment

Stepping back, a clear pattern emerges, and it answers the question that matters most to a practitioner sizing up a motion: which factors actually carry a no-confusion ruling at summary judgment?

The factors whose inputs are objective and fixed—similarity (factor 2), competitive proximity (factor 3), bridging the gap (factor 4), and consumer sophistication (factor 8)—are the most amenable to resolution as a matter of law, because they turn on what the marks are, what the products are, and who buys them, rather than on contested testimony about states of mind. Strength (factor 1) is also frequently decidable, at least on its conceptual dimension, and a finding of weakness lowers the bar for every other factor to favor the defendant. Actual confusion (factor 5) is the double-edged one: substantial evidence (a clean survey, credible misdirection) can defeat a defense motion, but its absence over a long, genuine period of coexistence is one of the most powerful engines of summary judgment for defendants. Intent (factor 6) supports a defendant whenever the plaintiff's evidence amounts only to knowledge of the senior mark—the common situation. Quality (factor 7) is usually a minor, neutral note.

The recurring profile of a successful defense motion therefore looks like this: a conceptually weak or descriptive plaintiff's mark; marks that differ in overall commercial impression when viewed in context (house brands, packaging, trade dress); products in genuinely separate markets with no realistic likelihood of bridging; little or no probative evidence of actual confusion, often reinforced by a substantial period of confusion-free coexistence; no evidence of bad faith beyond mere awareness; and sophisticated purchasers. Star Industries, Streetwise Maps, Time v. Petersen, and Nora Beverages are all variations on this profile—cases in which the Second Circuit was comfortable holding that no reasonable jury could find confusion likely. The factors that did the work were similarity, proximity, and strength, with the treatment of actual confusion, intent, and sophistication reinforcing the result.

What the pattern does not mean is that summary judgment is easy or automatic. Cadbury remains the counterweight: where the marks are genuinely close, the products overlap, and the record contains real evidence of confusion or bad faith, the confusion inquiry belongs to the jury, and a district court that resolves those disputes on the papers risks reversal. The honest summary is that summary judgment for a defendant is achievable and regularly achieved when the key objective factors line up decisively, but it remains the exception—reserved for cases where the plaintiff's best evidence still cannot get confusion to a jury. (When it cannot be resolved on the papers, the question of who decides—judge or jury—takes center stage; we cover that in Bench Trial vs. Jury Trial Issues in Trademark Litigation.)

A worked hypothetical

Consider a concrete, invented dispute that ties the factors together. The following is a hypothetical for illustration only; it does not describe any real company, mark, or case.

"Meridian" is a well-regarded maker of high-end home audio equipment—amplifiers and speakers that sell for thousands of dollars to audiophiles and custom-installation professionals. It holds an incontestable federal registration for MERIDIAN covering audio hardware. A software company, also using "Meridian," sells subscription accounting software to small businesses. Meridian Audio sues Meridian Software for trademark infringement in the Eastern District of New York, and the software company moves for summary judgment of no likelihood of confusion.

Run the factors.

Strength. "Meridian" is a real word used arbitrarily for both products—it describes neither audio gear nor accounting software—so it is conceptually strong, and the incontestable registration adds a presumption of distinctiveness. But its commercial strength is confined to the audio market. Roughly neutral, modestly favoring the plaintiff.

Similarity. The word marks are identical, which favors the plaintiff—but the court will assess overall impression in context, and each company uses the word with its own distinct logo, color scheme, and product presentation, somewhat softening the identity. Tilts to the plaintiff, but not as far as a naked side-by-side would suggest.

Proximity. The products could hardly be more remote—luxury audio hardware versus business accounting software, sold to different buyers through entirely different channels, with no overlap in trade shows, retailers, or advertising. Strongly favors the defendant.

Bridging the gap. The plaintiff has no publicly known plans to enter the accounting-software business, and no consumer would expect an audiophile speaker company to do so. Under Lang, an unannounced subjective intent to expand is irrelevant. Strongly favors the defendant.

Actual confusion. The two have coexisted for six years in overlapping geographic markets, and the plaintiff can point to only one ambiguous email asking whether the companies are "related." Under Nora Beverages, that is de minimis—and arguably an inquiry, not confusion at all—while the long, confusion-free coexistence supports the Starbucks inference. Favors the defendant.

Intent. The software company chose "Meridian" because it evokes a high point or peak, ran a clearance search, relied on counsel, and had no awareness of the audio company. Even if it had known, mere knowledge is not bad faith under Playtex. Favors the defendant or neutral.

Quality. Both products are well regarded. Neutral.

Sophistication. Buyers of multi-thousand-dollar audio systems and buyers of business accounting software are both deliberate, researched purchasers. Favors the defendant.

Now weigh holistically rather than by counting. Identity of the word marks favors the plaintiff and is not trivial. But the marks appear in radically different contexts, the products occupy entirely separate markets with no realistic bridging, the buyers are sophisticated, there is no probative evidence of actual confusion despite years of coexistence, and there is no evidence of bad faith. On this record, a court could reasonably conclude that no reasonable jury could find a likelihood of confusion, even though one strong factor—mark identity—favors the plaintiff. This is the Star Industries lesson in action: a defendant can win summary judgment despite identical word marks, where proximity, sophistication, the absence of confusion, and good faith point decisively the other way.

Change a few facts, though, and the motion collapses. Suppose Meridian Software sold audio-production software to recording studios, advertised at the same trade shows as the hardware company, and a dozen documented instances showed dealers and customers conflating the two. Now proximity, bridging, and actual confusion all favor the plaintiff, Morningside's rule that sophistication cannot save highly similar offerings kicks in, real factual disputes exist, and the case belongs to a jury under Cadbury. The factors did not change. The facts did—and that is precisely the point.

Practical takeaways

For the defendant moving for summary judgment, build the motion around the objective factors the court can resolve without trial. Present the marks as consumers actually see them—full commercial context, house brands, trade dress—to argue dissimilarity in overall impression, and remember that the court is forbidden from a clean side-by-side comparison, so frame the encounter as a consumer would experience it. Establish the separate markets, customers, and channels to win proximity and bridging. Document the nature and sophistication of the buyers. And—often most powerfully—develop the record of confusion-free coexistence, marshaling the absence of probative actual-confusion evidence over a meaningful period in overlapping markets under Starbucks and Nora Beverages. Anticipate the bad-faith argument by showing it reduces to mere knowledge, and reinforce good faith with evidence of independent adoption, a clearance search, advice of counsel, and distinct branding. Concede the factors you cannot win (such as identity of the word marks) and explain why they do not change the holistic result. The goal is to demonstrate that even on the plaintiff's best version of the facts, confusion is not likely as a matter of law.

For the plaintiff opposing the motion, the path runs through Cadbury: identify the genuine factual disputes that make the question a jury issue, and resist the temptation to argue every factor with equal force. Marshal real evidence of actual confusion—credible misdirection, lost sales, a methodologically sound survey that will survive a Daubert challenge—because thin or anecdotal confusion will be dismissed as de minimis. Show genuine market overlap or a publicly known, concrete likelihood of bridging, with evidence rather than aspiration. If bad faith is in play, produce evidence of intent to exploit goodwill—copying, the absence of a credible explanation, a suspicious adoption—not merely awareness. And emphasize the contexts in which the distinguishing trade dress falls away and the marks are encountered in close, confusing proximity (phone orders, search results, point-of-sale displays). The plaintiff's task is not to win the factors outright on paper but to show that reasonable jurors could find confusion likely.

For both sides, three principles control. First, the factors are weighed holistically, not counted, so the number of factors in your favor matters far less than which factors and how strongly. Second, summary judgment is available but exceptional—achievable when the key objective factors align decisively, and off the table when genuine disputes infect the factors that matter. Third, the absence-of-actual-confusion inference from long, genuine coexistence is one of the most powerful tools in the defense arsenal and one of the most important records for a plaintiff to rebut. Master those three principles, and the Second Circuit's summary judgment case law stops looking like a thicket of inconsistent outcomes and starts looking like a coherent application of a single question: on this record, could a reasonable jury find that ordinary prudent purchasers are likely to be confused?

If you litigate these motions, two adjacent topics deserve a place on your reading list. The remedies that follow a finding of infringement—and the fee-shifting that can follow a finding of bad faith—are covered in Lanham Act Attorneys' Fees Under 15 U.S.C. § 1117(a) and Damages Apportionment in Trademark Cases. And because the confusion analysis is reviewed differently on appeal than the underlying fact findings, the standards of review can decide whether a summary judgment survives—see Second Circuit Appellate Standards in Trademark Cases.

Frequently asked questions

Is likelihood of confusion a question of fact or a question of law? Both, in a specific sense. In the Second Circuit, the court's findings on each individual Polaroid factor are findings of fact, but the ultimate weighing of those factors into a conclusion about whether confusion is likely is a question of law. Guthrie, 826 F.3d at 37–38. That mixed character is exactly what allows a court to grant summary judgment: it does not resolve genuine disputes about the underlying facts, but once those facts are undisputed, it may perform the ultimate balance as a matter of law.

Can a defendant win summary judgment even though some Polaroid factors favor the plaintiff? Yes—and this is the most important practical lesson. Because the analysis is holistic rather than a tally, a defendant can prevail even when one or more factors (often the similarity of the marks) favor the plaintiff, so long as the factors that matter most in the case point decisively toward no confusion. Star Industries affirmed summary judgment for the defendant despite real similarity in the parties' marks.

How long must two marks coexist before the absence of actual confusion becomes meaningful? There is no bright-line number, but the coexistence must be long enough, and overlapping enough, that confusion would have surfaced if it were going to. A short overlap or coexistence in separate markets gives little weight to the absence of confusion, because there was never a real opportunity for it to occur. Guthrie, 826 F.3d at 44; TCPIP Holding, 244 F.3d at 102. Where the parties have sold side by side in the same markets for years with no meaningful confusion, the inference becomes strong. Starbucks, 588 F.3d at 117.

Does proof that the defendant knew about the plaintiff's mark establish bad faith? No. Prior knowledge of the senior user's mark does not, by itself, create an inference of bad faith. Playtex, 390 F.3d at 166. Bad faith requires evidence that the defendant adopted its mark intending to exploit the plaintiff's goodwill and cause confusion—deliberate copying, the absence of a credible explanation for the choice, or similar proof. Knowledge plus suspicion is not enough to create a genuine dispute.

Do I need a consumer survey to survive summary judgment? Not strictly. Actual confusion can be shown without a survey, and likelihood of confusion can be found even with no proof of actual confusion at all. Sports Auth., 89 F.3d at 964. But surveys are among the most persuasive ways to manufacture a genuine dispute, and their absence invites the inference that none could be found. If you use one, it must be built to survive a Daubert challenge—see Consumer Survey Expert Methodology in Trademark Cases.

Does the Eastern District of New York apply a different standard than the rest of the Second Circuit? No. The EDNY, like every district court in the Circuit, applies Polaroid and the Circuit's gloss on it. What varies among districts is the local rules and procedural mechanics—statements of material fact, page limits, and the like—not the substantive confusion standard.

If I lose my summary judgment motion, what happens to the case? Denial of summary judgment generally means the confusion question proceeds to trial, where the Polaroid factors are resolved by the factfinder. Whether that factfinder is a judge or a jury depends on how the case is postured and pleaded; the consequences of that choice—and how to make it—are discussed in Bench Trial vs. Jury Trial Issues in Trademark Litigation.

Conclusion and next steps

The Polaroid factors have governed trademark confusion in the Second Circuit since a camera company sued an electronics maker in 1961, and for all the case law that has accreted since, the framework remains workable precisely because it is honest about its own limits: it is a guide to a single question, not an algorithm. On summary judgment, that honesty translates into a usable rule of thumb. The objective factors—similarity, proximity, bridging the gap, sophistication, and the conceptual side of strength—are the ones a court can decide on a paper record, and a defense motion lives or dies on whether they align. The subjective and evidentiary factors—actual confusion and intent—either supply the genuine dispute that sends a case to trial or, in their telling absence, reinforce a judgment as a matter of law.

If you are evaluating a motion, start by sorting the factors into those buckets and asking, factor by factor, whether the record contains a genuine dispute or merely an asserted one. Then weigh holistically, not numerically. A defendant who does that honestly will know whether it has a Star Industries case or a Cadbury case before it spends a dollar on briefing. A plaintiff who does the same will know exactly which factual disputes to develop in discovery and document in opposition. Either way, because the application of these factors is intensely fact-specific, the prudent next step is to consult experienced trademark litigation counsel about the particular marks, goods, and record in your case.

Related articles

This article is provided for general informational purposes and does not constitute legal advice. The application of the Polaroid factors is intensely fact-specific, and outcomes vary by case and procedural posture. Consult qualified trademark litigation counsel about any particular matter.