Cadbury Beverages, Inc. v. Cott Corporation and Cott Beverages Usa, Inc.

*476 JOSÉ A. CABRANES, Circuit Judge: The question presented is whether there is a likelihood of confusion when two manufacturers use the same mark in the sale of soft drinks, but one uses the mark primarily as a brand label, and the other uses it as a trade name in the distribution of soft drinks to retail stores for sale under their own “private labels.” The United States District Court for the Southern District of New York (Kevin Thomas Duffy, Jvdge) concluded that no reasonable trier of fact could find a likelihood of confusion and therefore granted summary judgment in favor of the defendants. See Cadbury Beverages, Inc. v. Cott Corp., 850 F.Supp. 256 (S.D.N.Y.1994). Since we find that there were genuine issues of material fact inappropriate for resolution on cross-motions for summary judgment, we vacate and remand. I The undisputed facts are as follows. Cadbury Beverages, Inc. (“Cadbury”), the plaintiff, is a Delaware corporation with its principal place of business in Stamford, Connecticut. Cadbury sells soft drinks under the trademark “Cott” to licensed bottlers, who bottle and sell “finished” Cott soft drinks. Cadbury also uses “Cott” as a trade name, as it sells “Cott” soft drinks to retail stores through its Cott U.S.A. Division. Cadbury owns six registrations of the “Cott” mark issued by the United States Patent and Trademark Office. Its right to use five of these registered marks is “incontestable,” because the marks have been registered for five consecutive years. 15 U.S.C. §§ 1065, 1115. 1 Cott Corporation, one of the defendants, is a Canadian company in the soft drink business. Its wholly owned U.S. subsidiary, Cott USA Corp., conducts its business through two wholly owned subsidiaries, Cott Beverages USA, Inc. and Cott Distributions USA, Inc. Cott Beverages USA, Inc. is the second defendant in this case. We refer to either or both defendants as “Cott.” Cott is principally in the so-called “private-label” business, as opposed to the “brand-name” business. This means that Cott sells beverages to retail stores for sale under the retailers’ own labels. Although Cott uses the “Cott” trademark for some branded soft drinks distributed in Canada, it does not sell any branded soft drinks under the “Cott” trademark in the United States. Rather, its sole business in the United States is the sale of private-label beverages that bear the retailer’s chosen brand name — for example, Wal-Mart’s “Sam’s American Choice” label is a Cott beverage, as is A & P’s “Master Choice.” It is no coincidence that the parties to this litigation both use the name “Cott.” In the 1960s, one of the plaintiffs predecessors assigned its rights to the “Cott” mark in Canada to a company that was a predecessor to the defendant Cott Corp. Cott has been engaged in the private-label business in the United States since May 1991. In November 1991, Cadbury complained to Cott about the latter’s use of the “Cott” trademark in its private-label business in the United States. After failing to receive a response it considered satisfactory, Cad-bury filed the instant action in 1992 for statutory trademark infringement under the Lan-ham Act, 15 U.S.C. §§ 1114(1) and 1125(a) (1988), and common law trademark infringement. The complaint alleged that the defendants’ use of the name “Cott” was likely to create the erroneous impression that Cott’s goods are associated with the plaintiff. Cad-bury sought to enjoin the defendants from using “Cott” in their corporate and trade names, and it sought to recover profits, other damages, costs, and attorney’s fees. At the first status conference in the case, and before *477 any discovery was taken, the district court directed the parties to file cross-motions for summary judgment. In its motion, and in a subsequent motion for reconsideration and reargument, Cadbury contended that the defendants’ use of the name “Cott” would leave wholesale purchasing agents (also called “retailers”) — who purchase both brand-name and private-label beverages for sale in retail stores — confused as to the source of the defendants’ products. Cadbury also argued that consumers would be confused as to the source of the private-label beverages if they learned that the products were manufactured by a company called Cott Beverages USA, Inc. Cadbury cited instances in which consumers were or could be exposed to the name “Cott” in connection with the defendants’ product. Cadbury drew the court’s attention to three specific items: (1) a Wal-Mart “shelf hanger” — a tag that hangs over a shelf on which the product sits — bearing the name “Cott Beverages USA, Inc.,” near the beverage “Sam’s Choice Clear and Natural”; (2) the fact that in 1993, four percent of the beverages produced by Cott Beverages USA were sold to consumers in bottles with caps bearing a reference to “Cott Beverages”; and (3) signs at Cott’s plant in Oakfield, New York that say “Cott” and “Cott BEVERAGES.” The defendants did not dispute that “Cott” was part of their corporate names, and that it appeared in the parent company’s filings with the U.S. Securities and Exchange Commission and in its listing on the NASDAQ stock exchange. Cott also agreed that it used the “Cott” name on its letterhead, invoices, and other communications with wholesale purchasing agents. But the defendants contended that the wholesale purchasing agents were sophisticated buyers who would not be confused by these uses. They argued further that the instances of consumer exposure to the “Cott” name in connection with their products were isolated and inadvertent events. In general, the defendants contended, consumers have no way of knowing that the private-label beverages that they purchase are manufactured by Cott. The district court analyzed these facts under the eight factors famously set forth by Judge Friendly in Polaroid Corp. v. Polaroid Electronics Corp., 287 F.2d 492, 495-96 (2d Cir.), cert. denied, 368 U.S. 820, 82 S.Ct. 36, 7 L.Ed.2d 25 (1961). Finding that the defendants were entitled to judgment as a matter of law on the question of the likelihood of confusion created by their use of the mark “Cott,” the court denied the plaintiffs summary judgment motion and granted the defendants’ cross-motion for summary judgment. The plaintiff appeals, asking not only that we vacate the order granting summary judgment in favor of the defendants, but also that we remand with instructions to enter summary judgment in the plaintiffs favor. II To establish a trademark infringement claim under the Lanham Act, a plaintiff must show that the defendant used in commerce, without the plaintiffs consent, a “reproduction, counterfeit, copy, or colorable imitation of a registered mark in connection with the sale, offering for sale, distribution, or advertising of any goods or services or in connection with which such use is likely to cause confusion.” 15 U.S.C. § 1114(l)(a). 2 To prevail under this statute, the plaintiff must show that “it has a valid mark entitled to protection and that the defendant’s use of it is likely to cause confusion.” Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d 1072, 1075 (2d Cir.1993). The defendants do not dispute the validity of Cadbury’s “Cott” trademark. Accordingly, the inquiry turns to whether “numerous ordinary prudent purchasers are likely to be misled or confused as to the source of the product in question because of *478 the entrance in the marketplace of defendant’s mark.” Id. at 1077. In deciding whether the plaintiff has established likelihood of confusion, the law of this Circuit requires that courts consider the eight factors elaborated in Polaroid. 287 F.2d at 495. The factors, explained in greater detail below, are: (1) strength of the plaintiffs mark; (2) the degree of similarity between the two marks; (3) the proximity of the products; (4) the likelihood that the prior owner will “bridge the gap” (in this case, the likelihood that Cadbury will enter Cott’s market under the “Cott” mark); (5) actual confusion; (6) the defendant’s good faith in adopting its mark; (7) the quality of the defendant’s product; and (8) the sophistication of the buyers. Id. In trademark infringement cases, the district court’s findings with regard to each of the Polaroid factors “are entitled to considerable deference,” even on appeals from summary judgment. Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 873 (2d Cir.1986). The district court’s ultimate balancing of the Polaroid factors is always subject to de novo review. Paddington Corp. v. Attiki Importers & Distribs., Inc., 996 F.2d 577, 584-85 (2d Cir.1993). The question before the district court, and the question on appeal, is whether, based on the Polaroid inquiry regarding the likelihood of confusion, either party is entitled to summary judgment. We have held that summary judgment in a trademark action may be appropriate in certain circumstances, where the undisputed evidence would lead only to one conclusion as to whether confusion is likely. See, e.g., Lang v. Retirement Living Publishing Co., Inc., 949 F.2d 576, 582 (2d Cir.1991); Lois Sportswear, 799 F.2d at 876; see also Murphy v. Provident Mutual Life Ins. Co., 923 F.2d 923, 930 (2d Cir.1990), cert. denied, 502 U.S. 814, 112 S.Ct. 65, 116 L.Ed.2d 40 (1991); Universal City Studios, Inc. v. Nintendo Co., Ltd., 746 F.2d 112, 115 (2d Cir.1984). Our task on this appeal, then, is to determine whether any reasonable trier of fact could conclude that confusion is likely and, if so, whether no reasonable trier of fact could conclude that it is not. Ill The district court began its analysis with the proposition that the facts of this case are not in dispute, and that it needed only to determine the “legal conclusions to be drawn from these facts.” 850 F.Supp. at 257. That the parties do not dispute the basic facts in this case does not necessarily mean that these facts alone compel certain findings as to each of the Polaroid factors. If a factual inference must be drawn to arrive at a particular finding on a Polaroid factor, and if a reasonable trier of fact could reach a different conclusion, the district court may not properly resolve that issue on summary judgment. With this in mind, we examine the district court’s decision. The district court focused on two factors as central to its inquiry: proximity of the products and sophistication of purchasers. It concluded that these two factors favored the defendants and “establish[ed] that the defendants’ use of the ‘Cott’ name does not amount to a Lanham Act violation.” Id. at 260. The district court also found or assumed that certain factors (strength of the mark, similarity of the marks, and defendants’ good or bad faith) favored the plaintiff, and that other factors (“bridging the gap,” the lack of actual confusion, and the quality of defendants’ product) favored the defendants. None of these factors, however, truly appears to have influenced the district court’s inquiry into whether there exists a material question of fact as to the likelihood of confusion. In relying almost exclusively upon the proximity-of-the-products and sophistication-of-the-buyers factors for its conclusion that no reasonable trier of fact could find “that an appreciable number of consumers will ... be confused as to the source of [the defendants’ products],” id., the district court appears to have resolved factual questions, rather than finding that no factual questions existed. Granting the appropriate deference to the district court’s findings on the Polaroid factors, it cannot be said as a matter of law that the proximity-of-the-products and sophistication-of-the-buyers inquiries favored the defendants. Because the district court’s findings on these two factors formed the basis of *479 its conclusion that, as a matter of law, eonfusion as to the source of the defendants’ products is not likely, summary judgment was improperly granted in the defendants’ favor, As discussed more fully below, we also con-elude that summary judgment in the plaintiffs favor would not be proper. We examine each Polaroid factor in turn to determine whether the parties’ factual presentations raise a genuine issue of fact. We conclude that two of the factors — the strength of the plaintiffs mark and the similarity of the marks — are properly resolved as a matter of law in the plaintiffs favor. We also conclude that, on the record before us, the lack of actual confusion could properly have been found to favor the defendants, although further discovery could lead to a different result. As to the remaining factors — proximity of the products, sophistication of the buyers, “bridging the gap,” and the quality of the defendants’ product — there exist questions of material fact precluding a conclusion on summary judgment that confusion is or is not likely. 1. Strength of the Mark The “strength” of a mark is its “tendency to identify the goods sold under the mark as emanating from a particular, although possibly anonymous, source.” McGregor-Doniger, Inc. v. Drizzle, Inc., 599 F.2d 1126, 1131 (2d Cir.1979). The district court stated that the “Cott” trademark held by Cadbury “may in fact be strong.” 850 F.Supp. at 259. We agree that it is, for several reasons. First, the mark is fanciful — that is, it is a term without any dictionary meaning. Gruner, 991 F.2d at 1075-76. Unlike a generic term, which can never be a valid trademark, or a descriptive term, which can be protected only if it has acquired “secondary meaning,” 3 see McGregor-Doniger, 599 F.2d at 1131, a fanciful trademark “is entitled to the most protection the Lanham Act can provide.” Lois Sportswear, 799 F.2d at 871. Second, the plaintiff and its predecessors have sold soft drinks under the mark for more than sixty years. See Life Indus. Corp. v. Star Brite Distrib., Inc., 31 F.3d 42, 46 (2d Cir.1994) (fanciful and long-used trade dress considered “strong”). Third, the mark is registered and incontestable. See Gruner, 991 F.2d at 1076-77; Lois Sportswear, 799 F.2d at 871. In urging us to conclude that the plaintiffs •mark is not strong, the defendants draw our attention to the following facts: (1) sales of Cott beverages account for less than one percent of Cadbury’s sales; (2) sales of Cad-bury’s Cott beverage declined approximately nine percent between 1988 and 1992; (3) the defendants’ sales were stronger than Cad-bury’s; and (4) Cadbury’s Cott beverage is only distributed in the northeastern United States. However, when a mark is registered and fanciful, the plaintiff has “met its burden” on the question of strength. Mushroom Makers, Inc. v. R.G. Barry Corp., 580 F.2d 44, 48 (2d Cir.1978), cert. denied, 439 U.S. 1116, 99 S.Ct. 1022, 59 L.Ed.2d 75 (1979). A court’s further consideration of the recognition that the mark actually enjoys in the marketplace is not error, McGregor-Doniger, 599 F.2d at 1132-33, but the facts set forth by the defendants here do not begin to cast doubt on the otherwise firm conclusion that the plaintiff’s fanciful mark is strong. Cadbury’s overall sales do not bear on the commercial success of its Cott brand, and neither a small decline in the plaintiffs’ sales in the course of a sixty-year period nor the success of the defendants’ products works to strip the plaintiff of protection. See id. at 1132 (“the relatively small size of a senior user’s advertising budget or sales volume will not diminish the strength of its valid mark, and the scope of protection accorded to that mark will not be narrowed because of such evidence”). For these reasons, we conclude that as a matter of law “Cott” is a strong mark entitled to broad protection against infringement. See Mobil Oil Corp. v. Pegasus Petroleum Corp., 818 F.2d 254, 257-58 (2d Cir.1987). ’ 2. Similarity of the Marks This factor asks whether the similarity of the marks is likely to cause confusion. *480 W.W.W. Pharmaceutical Co. v. Gillette Co., 984 F.2d 567, 573 (2d Cir.1993). The district court found that the defendants “do employ the identical ‘Cott’ name.” 850 F.Supp. at 259. We agree. Not only is the name the same, but the defendants use the same large, script print that the plaintiff has historically used. The defendants argue that consumers never see their use of “Cott,” and that purchasing agents would not be confused by their use. But in this ease these contentions are more properly addressed under the proximity-of-the-products and sophistication-of-the-purchasers factors. For the purpose of considering the question of the similarity of the marks, the district court correctly determined that as a matter of law these marks are identical. 3. Proximity of the Products and 8. Sophistication of the Purchasers The “proximity-of-the-products” inquiry concerns whether and to what extent the two products compete with each other. See Lang, 949 F.2d at 582. We look to “the nature of the products themselves and the structure of the relevant market.” Vitarroz v. Borden, Inc., 644 F.2d 960, 967 (2d Cir.1981). Among the considerations germane to the structure of the market are the class of customers to whom the goods are sold, the manner in which the products are advertised, and the channels through which the goods are sold. See id. The eighth Polaroid factor, “sophistication of the buyers,” has been called “analogous” to the proximity factor. Id. The sophistication factor “recognizes that the likelihood of confusion between the products at issue depends in part on the sophistication of the relevant purchasers.” Arrow Fastener Co., Inc. v. Stanley Works, 59 F.3d 384, 399 (2d Cir.1995) (citing W.W.W. Pharmaceutical, 984 F.2d at 575). The parties do not dispute that in this case the relevant purchasers are both ordinary consumers of soft drinks who shop in retail stores, and wholesale purchasing agents who purchase soft drinks for sale in retail stores. The district court found that an analysis of the proximity of the products and the sophistication of the buyers together favored the defendants’ argument that there was no likelihood of confusion as to the source of their products. The court found first that consumers “cannot be confused” because the private-label products are “not called ‘Cott.’ ” 850 F.Supp. at 258. That is, inasmuch as the parties’ products are essentially identical, a consumer who knew that the private-label product was manufactured by Cott would likely be confused. But because consumers do not know that fact in the present context, the district court reasoned, there is no risk of confusion. To support this reasoning, the district court relied on a line of cases finding moderate or no proximity between different products that were marketed to different customers. See Windsor, Inc. v. Intravco Travel Ctrs., Inc., 799 F.Supp. 1513, 1523-24 (S.D.N.Y.1992) (packaged sight-seeing tours vs. study-abroad programs); Edison Bros. Stores, 651 F.Supp. at 1557 (women’s clothing and shoes vs. women’s perfume products); Oxford Indus., Inc. v. JBJ Fabrics, Inc., 6 U.S.P.Q.2d 1756, 1761, 1988 WL 9959 (S.D.N.Y.1988) (women’s apparel vs. the service of textile fabric conversion); In re Shipp, 4 U.S.P.Q.2d 1174, 1175-76 (Trademark Trial and App.Bd.1987) (dry cleaning equipment vs. consumer dry cleaning services). These eases, however, are not on point. In this case, the parties sell the same product — soft drinks. Their cans and bottles of soft drinks are sold through the same channels of trade to the same class of customers — from manufacturers or bottlers to wholesale purchasing agents, and then to the general public. Under these circumstances, there is no dispute that members of the general public would be confused if they knew that Cott produced the private-label products. The district court found, however, that the ordinary consumer does not know that the private-label product is produced by Cott. 850 F.Supp. at 258. Such a finding was inappropriate on the defendants’ cross-motion for summary judgment, for purposes of which all inferences must be drawn in favor of the plaintiff; Even without the benefit of discovery, the plaintiff pointed out two instances in which consumers were exposed to *481 the Cott name in connection with the private-label brand — the Wal-Mart shelf hanger and the bottle caps on the private-label beverage. Attempting to characterize the Wal-Mart incident as aberrational, an officer of the defendants stated in an affidavit that “retailers attempt to create an aura that their private-label beverages are produced by [the retailers] rather than coming from another source.” Although this statement seems sensible enough, another retañer could just as easñy use Cott’s name at any time. Indeed, one need not resort to groundless speculation to envisage other situations that would generate confusion among ordinary consumers. Any impurity, contamination, or adulteration that plagued one of the defendants’ products could easñy generate a wave of publicity that would leave Cadbury struggling to explain to the public — through the mass media — the technical niceties of private labeling. The holder of a strong and registered mark should not have to wait for catastrophe to strike to secure legal protection when damage ensuing in the wake of catastrophe could render any after-the-fact protection useless. With respect to the purchasing agents, the district court found that these are sophisticated buyers who are not likely to be confused. 850 F.Supp. at 258. In so doing, the district court appears again to have resolved a disputed issue; the parties submitted conflicting affidavits on this question. Whüe we have recognized that wholesale purchasers (or “retaüers”) are “assumed to be more sophisticated buyers and thus less prone to confusion,” W.W.W. Pharmaceutical, 984 F.2d at 576, that assumption is not necessarily dispositive in these circumstances, where the plaintiffs mark is strong, the defendants’ mark is identical, and the products are sub-stantiañy the same. As Cadbury weU points out, manufacturers of branded products often sefi products under a private label. See Fremont Co. v. ITT Continental Baking Co., 199 U.S.P.Q. 415, 420 (S.D.N.Y.1977) (food industry manufacture products for direct sale also manufacture products for sale by others under private label). Since manufacturers of brand name products often enter the private-label business, it cannot be said as a matter of law that wholesale purchasers — sophisticated or otherwise — are unlikely to be confused by Cott’s use of the “Cott” trade name. Indeed, the plaintiff submitted: (1) an article from a financial magazine featuring a picture of Cott’s CEO holding a soft drink can bearing the “Cott” trademark; (2) the business card of one of defendants’ employees, bearing the name “Cott” in the large, script print that has historicaUy been used on Cadbury’s Cott drinks; (3) listings from a trade reference known as the Beverage Bureau Book— presumably accessible to, and used by, the very wholesale purchasers the parties have in mind — describing Cott’s products to include “Cott soft drinks”; and (4) formula sheets for defendants’ private-label beverages bearing the large, script “Cott” logo. Purchasing agents sophisticated or knowledgeable about the general differences between private-label and branded products could be confused as to what specific company manufactures a particular product, in light of correspondence, invoices, trade listings, and articles displaying the Cott logo or referring to “Cott soft drinks.” These purchasing agents would have to keep in mind that Cadbury Beverages, Inc. uses the Cott mark and Cott trade name to market its Cott beverages with the Cott label, whüe Cott Beverages USA, Inc. uses the Cott mark and Cott trade name to market Cott beverages with private labels. In considering the defendants’ cross-motion for summary judgment, the district court was required to draw all inferences in the plaintiffs favor. It failed to do so, and indeed impermissibly resolved a disputed question of fact in the defendants’ favor. In sum, the district court’s resolution of these two factors in defendants’ favor was improper on a motion for summary judgment. Whüe it is true that consumers could not be confused if they were unaware that Cott produced the private-label beverages, the district court was not permitted to assume such ignorance for purposes of defendants’ summary judgment motion. Simñarly, the district court should not have assumed that purchasing agents at the wholesale level are so sophisticated that no reasonable fact-finder could find confusion likely as to the source of particular private-label products, despite the existence of invoices, correspon *482 dence, and other sources bearing the “Cott” mark. For similar reasons, we decline to conclude that no further factfinding is necessary to resolve these inquiries in plaintiffs favor. Although Cadbury called the court’s attention to incidents in which consumers were or could have been exposed to the Cott name in connection with Cott’s private-label products, it is for a factfinder to assess whether such incidents are, as Cott contends, likely to remain aberrational, or whether some such circumstances could give rise to knowledge on the part of ordinary consumers that Cott produces the products. Similarly, just as the district court is not permitted to assume for purposes of the defendants’ motion for summary judgment that wholesale purchasers are so sophisticated that no possibility of confusion exists, we are not permitted to assume that these purchasers are unsophisticated for purposes of the plaintiffs motion. We find it difficult to imagine that even a sophisticated purchaser would not be confused in the face of defendants’ invoices and correspondence bearing a logo virtually identical to that on Cadbury’s product, but to so conclude would require that we draw inferences impermissible on a motion for summary judgment. As to the proximity-of-the-products and sophistication-of-the-buyers inquiries, a sufficient dispute exists to preclude resolution as a matter of law in favor of the plaintiff or in favor of the defendants. 4.Bridging the Gap This factor concerns the likelihood that Cadbury will enter the private-label business under the “Cott” mark. See Arrow Fastener, 59 F.3d at 397 (quoting Lang, 949 F.2d at 582). The plaintiff argues that there is no “gap to bridge” because the parties already sell the same products in the same channels of trade. The district court instead saw two markets — the brand-name and private-label markets — and it found that Cadbury had expressed no interest in entering the latter. 850 F.Supp. at 259. We are not entirely persuaded by the plaintiffs argument. We find it to be in tension with its position that “many manufacturers of branded products often produce products for sale under private labels.” Appellant’s Br. at 21. The district court did not err in concluding that Cadbury had displayed no present intent to enter the private-label business. Nevertheless, this court has stated that “the absence of a present intent to bridge the gap is not determinative.” McGregor-Doniger, 599 F.2d at 1136. Rather, “the assumptions of the typical consumer ... must be taken into account.” Id. Because it is surely plausible that a manufacturer of branded products such as Cadbury would enter the private-label market, it is also plausible that a wholesale purchasing agent would conclude that Cadbury had already done so — i.e., that Cadbury had already bridged the gap, with its Cott USA Division entering the private-label market under the name Cott Beverages USA, Inc. The district court therefore erred in concluding that, as a matter of law, this factor favored the defendants. Facts remain to be developed as to whether purchasers and consumers would perceive a gap at all, or would assume that any gap has already been bridged. The district court’s resolution of these matters on summary judgment was inappropriate. 5.Actual Confusion While evidence of actual confusion is not necessary to the plaintiffs claim, “its lack may under some circumstances be used against the plaintiff.” Hasbro, Inc. v. Lanard Toys, Ltd., 858 F.2d 70, 78 (2d Cir.1988). Based on Cadbury’s lack of evidence of actual confusion, the district court inferred that consumer confusion was not likely. 850 F.Supp. at 259. That inference was arguably proper on the limited record before the district court, particularly in light of the fact that the record is devoid of any requests by the plaintiff for an opportunity for further discovery. The discovery process may, however, yield instances of actual confusion to buttress the plaintiffs claim. 6.The Defendants’ Good Faith in Adopting Their Mark This factor concerns “whether the defendant adopted its mark with the intention of capitalizing on [the] plaintiffs reputation and *483 goodwill and any confusion between Ms and the senior user’s product.” Lang, 949 F.2d at 588 (quoting Edison Bros. Stores, 651 F.Supp. at 1560). The district court assumed that Cott used the “Cott” mark in the Urnted States “with full knowledge of Cadbury’s rights,” and that this factor therefore favored the plaintiff, but held that the defendants should prevail nevertheless. 850 F.Supp. at 259-60. Full knowledge of a prior use of a protected mark is not necessarily inconsistent with a finding of good faith, particularly where the alleged infringer is unsure as to the scope of the protection. See, e.g., Lang, 949 F.2d at 584; Mushroom Makers, Inc. v. R.G. Barry Corp., 580 F.2d 44, 48 (2d Cir.1978) (per curiam), cert. denied, 439 U.S. 1116, 99 S.Ct. 1022, 59 L.Ed.2d 75 (1979). In this case, however, it is not disputed that Cott proceeded with full knowledge of Cad-bury’s rights to the Cott mark in the Urnted States in the sale of soft drinks. The defendants contend that Cott Beverages USA uses the name “Cott” simply because that is the name of its Canadian parent company. Cott Beverages USA, as a U.S. subsidiary distributing goods in tMs country, must respect trademarks registered here. Particularly in light of the fact that Cadbury’s registered mark is identical — in style as well as in name — to Cott’s adopted logo as used on correspondence and invoices m the Urnted States, Cott’s explanation is not particularly persuasive, and certainly does not put the question of bad faith to rest. See Kiki Undies Corp. v. Promenade Hosiery Mills, Inc. 411 F.2d 1097, 1101 (2d Cir.1969) (where allegedly infringmg mark is identical to registered mark and its use began subsequent to plaintiffs registration, defendant must carry burden of explanation and persuasion). Nonetheless, recognizmg that “[s]ubjeetive issues such as good faith are singularly inappropriate for determination on summary judgment,” American Int'l Group, Inc. v. London Am. Int’l Corp., 664 F.2d 348, 353 (2d Cir.1981), we decline to hold as a matter of law that the defendants have acted in bad faith. 7. The Quality of the Defendants’ Product TMs factor generally considers whether the seMor user’s reputation could be “tar-mshed by [the] inferior merchandise of the juMor user.” Scarves by Vera, Inc. v. Todo Imports Ltd., 544 F.2d 1167, 1172 (2d Cir.1976) (quoted m W.W.W. Pharmaceutical, 984 F.2d at 575). The district court did not fully analyze tMs factor, but simply found that “reasonable consumers will not attribute any inferiority in the defendants’ product to Cadbury, because the general public does not recognize defendants by the name ‘Cott.’” 850 F.Supp. at 259. We agree with the defendants that the plaintiff produced no evidence of inferior quality, and that therefore on the record before us this factor lends no support to the plaintiffs claim. The discovery process may or may not lead to further information that would strengthen that claim. Moreover, as we noted above with regard to proximity of the products and sopMstication of the buyers, genuine issues of material fact exist as to whether in some circumstances consumers’ attention could be drawn to the identity of the producer of the private-label beverages, or whether quality control problems suffered by Cott could redound to the discredit of Cadbury by virtue of any confusion on the part of purchasing agents as to whether Cadbury is associated with Cott. IV Based on our review of the district court’s weighmg of the Polaroid factors, we conclude that neither party is entitled to summary judgment on the record before us. The district court’s findings as to two critical factors in this case — proximity of the products and sopMstication of the buyers — reflect an inappropriate resolution of disputed issues of fact in favor of the defendants. The district court improperly discounted the possibility that an appreciable number of wholesale purchasing agents could be confused, or that ordmary consumers could become aware that a company with the name “Cott” produced certain private-label beverages. The vagaries of production and distribution — as demonstrated by the information in tMs record even before discovery has begun — make the prospect of manufacturer exposure plausible. As to the other Polaroid factors, we conclude that, as a matter of law, the plaintiffs mark *484 is strong and the plaintiffs and the defendants’ marks are virtually identical. Disputed issues remain as to whether a “gap” exists between the plaintiffs and the defendants’ markets, or whether the plaintiff might be perceived to have bridged it; whether the defendants have acted in bad faith; and how the quality of the defendants’ products might affect consumers and purchasers. The only factor that can be weighed against the plaintiff on the record before us is the absence of evidence of actual confusion — and even this factor could turn in the plaintiffs favor following discovery. In these circumstances, summary judgment in favor of the defendant was inappropriate; summary judgment in favor of the plaintiff would be inappropriate as well. V The district court’s order granting summary judgment in favor of the defendant is vacated, and the case is remanded for further proceedings consistent with this opinion. Footnotes [1] 15 U.S.C. § 1065 (1994) provides in pertinent part: "CUhe right of the registrant to use such registered mark in commerce for the goods or services on or in connection with which such registered mark has been in continuous use for five consecutive years subsequent to the date of such registration and is still in use in commerce, shall be incontestable...." 15 U.S.C. § 1115(b) (1994) provides in pertinent part: "To the extent that the right to use the registered mark has become incontestable under section 1065 of this title, the registration shall be conclusive evidence of the validity of the registered mark....” [2] 15 U.S.C. § 1125, under which the plaintiff also sued, prohibits similar conduct, though it is not limited to the uses of registered trademarks. It deems liable for false designation of origin "[a]ny person who, on or in connection with any goods or services, or any container for goods, uses in commerce any word, term, name, symbol, or device, or any combination thereof, or any false designation of origin, false or misleading description of fact, or false or misleading representation of fact, which ... is likely to cause confusion_” 15 U.S.C. § 1125(a)(1)(A) (Supp. V 1993) (formerly designated § 1125(a)(1)). [3] A trademark acquires secondary meaning when " 'the name and the business have become synonymous in the mind of the public, submerging the primary meaning of the term in favor of its meaning as a word identifying that business.' ” Pirone v. MacMillan, Inc., 894 F.2d 579, 583 (2d Cir.1990) (quoting Abraham Zion Corp. v. Lebow, 761 F.2d 93, 104 (2d Cir.1985)).