Virgin Enterprises Ltd., Plaintiff-appellant, v. Tahir Nawab, Nathan Erlich, Simon Blitz, Daniel Gazal, Gerald Wren, Bob Wroblewski, Corporation Solutions, Llc, Cel-net Communication, Inc., the Cellular Network Communication Group, Inc., D/b/a Cncg, Sd Telecommunications Inc., Virgin Wireless, Inc., Iwaycity, and Virginwireless.com, Defendants-appellees, 335 F.3d 141 (2d Cir. 2003)

Annotate this Case
U.S. Court of Appeals for the Second Circuit - 335 F.3d 141 (2d Cir. 2003)

Argued: August 5, 2002
Decided: July 11, 2003


James W. Dabney (Catherine M. Clayton, Melissa A. Antonecchia, on the brief), Pennie & Edmonds LLP, New York, NY, for Appellant.

Kevin J. Harrington (John C. Mascari, on the brief), Harrington, Ocko & Monk, LLP, White Plains, NY, for Appellees.

Before: LEVAL, CALABRESI, and POOLER, Circuit Judges.

LEVAL, Circuit Judge.


Plaintiff Virgin Enterprises Limited ("VEL" or "plaintiff") appeals from the denial of its motion for a preliminary injunction. This suit, brought under § 32 of the Lanham Act, 15 U.S.C. § 1114(1), alleges that defendants infringed plaintiff's rights in the registered mark VIRGIN by operating retail stores selling wireless telephones and related accessories and services under the trade name VIRGIN WIRELESS. The United States District Court for the Eastern District of New York (Sifton, J.) denied plaintiff's motion for a preliminary injunction, based upon its finding that plaintiff's registration did not cover the retail sale of wireless telephones and related products, and that plaintiff failed to show a likelihood of consumer confusion.

We find that the plaintiff is likely to succeed on the merits and was entitled to a preliminary injunction. We therefore reverse and remand with instructions to enter a preliminary injunction.1 

Plaintiff VEL, a corporation with its principal place of business in London, owns U.S. Registration No. 1,851,817 ("the 817 Registration"), filed on May 5, 1991, and registered on August 30, 1994, for the VIRGIN mark as applied to "retail store services in the fields of ... computers and electronic apparatus" (emphasis added). Plaintiff filed an affidavit of continuing use, pursuant to 15 U.S.C. § 1058(a), on April 27, 2000, which averred that plaintiff had used the mark in connection with retail store services selling computers and electronic apparatus. Plaintiff also owns U.S. Registration No. 1,852,776 ("the 776 Registration"), filed on May 9, 1991, and registered on September 6, 1994, for a stylized version of the VIRGIN mark for use in connection with "retail store services in the fields of ... computers and electronic apparatus," and U.S. Registration No. 1,863,353 ("the 353 Registration"), filed on May 19, 1992, and registered on November 15, 1994, for the VIRGIN MEGASTORE mark. It is undisputed that these three registrations have become incontestable pursuant to 15 U.S.C. § 1065.

VEL, either directly or through corporate affiliates, operates various businesses worldwide under the trade name VIRGIN, including an airline, large-scale record stores called Virgin Megastores, and an internet information service. Plaintiff or its affiliates also market a variety of goods branded with the VIRGIN name, including music recordings, computer games, books, and luggage. Three of plaintiff's megastores are located in the New York area. According to an affidavit submitted to the district court in support of plaintiff's application for preliminary injunction, Virgin Megastores sell a variety of electronic apparatus, including video game systems, portable CD players, disposable cameras, and DVD players. These stores advertise in a variety of media, including radio.

Defendants Simon Blitz and Daniel Gazal are the sole shareholders of defendants Cel-Net Communications, Inc. ("Cel-Net"); The Cellular Network Communications, Inc., doing business as CNCG ("CNCG"); and SD Telecommunications, Inc. ("SD Telecom"). Blitz and Gazal formed Cel-Net in 1993 to sell retail wireless telephones and services in the New York area. Later, they formed CNCG to sell wireless phones and services on the wholesale level. CNCG now sells wireless phones and services to more than 400 independent wireless retailers. In 1998, Cel-Net received permission from New York State regulators to resell telephone services within the state.

Around 1999, Andrew Kastein, a vice-president of CNCG, began to develop a Cel-Net brand of wireless telecommunications products. In early 1999, Cel-Net entered into negotiations with the Sprint PCS network to provide telecommunications services for resale by Cel-Net. In August 1999, Cel-Net retained the law firm Pennie & Edmonds to determine the availability of possible service marks for Cel-Net. Pennie & Edmonds associate Elizabeth Langston researched for Kastein a list of possible service marks; among the marks Cel-Net asked to have researched was VIRGIN. Defendants claim that Langston told Cel-Net officer Simon Corney that VIRGIN was available for use in the telecommunications field. Plaintiff disputed this, offering an affidavit from Langston that she informed defendants that she would not search the VIRGIN mark because her firm represented plaintiff.2 

According to defendants, in December 1999, Cel-Net retained Corporate Solutions, LLC and its principals Nathan Erlich and Tahir Nawab as joint venture partners to help raise capital to launch Cel-Net's wireless telephone service. On December 2, 1999, Erlich and Nawab filed four intent-to-use applications with the U.S. Patent and Trademark Office ("PTO") to register the marks VIRGIN WIRELESS, VIRGIN MOBILE, VIRGIN COMMUNICATIONS, and VIRGIN NET in the field of telecommunications services, class 38. On December 24, 1999, Corporate Solutions incorporated defendant Virgin Wireless, Inc. ("VWI") and licensed to VWI the right to use the marks VIRGIN WIRELESS and VIRGIN MOBILE. Meanwhile, one of plaintiff's affiliates had begun to offer wireless telecommunication services bearing the VIRGIN mark in the United Kingdom. A press release dated November 19, 1999, found on plaintiff's website, stated that its Virgin Mobile wireless services were operable in the United States.

On June 23, 2000, defendant Blitz signed a lease under the name Virgin Wireless for a kiosk location in South Shore Mall in Long Island from which to re-sell AT & T wireless services, telephones, and accessories under the retail name Virgin Wireless. Defendants Cel-Net and VWI later expanded their telecommunications re-sale operations to include two retail stores and four additional retail kiosks in malls in the New York area and in Pennsylvania. All of these stores have been run by VWI under the trade name VIRGIN WIRELESS. VWI also has leases and bank accounts in its name, and has shown evidence of actual retail transactions and newspaper advertisements.

In August 2000, plaintiff licensed Virgin Mobile USA, LLC, to use the VIRGIN mark for wireless telecommunications services in the United States. On August 10, 2000, plaintiff filed an intent-to-use application with the PTO for use of the VIRGIN mark in the United States on telecommunications services and mobile telephones. On October 11, 2001, the PTO suspended this mark's registration in international class 9, which covers wireless telephones, and class 38, which covers telecommunications services, because the VIRGIN mark was already reserved by a prior filing, presumably defendants'. On August 16, 2001, plaintiff filed another intent-to-use application for the mark VIRGIN MOBILE to brand telecommunications services. The PTO issued a non-final action letter for both of plaintiff's pending new registrations on October 31, 2001, which stated that defendant Corporation Solutions' pending applications for similar marks in the same class could give rise to "a likelihood of confusion." The PTO suspended action on plaintiff's application pending the processing of Corporation Solutions' applications.

In October 2001, plaintiff issued a press release announcing that it was offering wireless telecommunications services and mobile telephones in the United States.

Plaintiff became aware of Corporation Solutions' application for registration of the VIRGIN WIRELESS and VIRGIN MOBILE marks by May 2000. In October 2001 and December 2001, defendant VWI filed suits against plaintiff in the federal district courts in Arizona and Delaware, alleging that plaintiff was using VWI's mark. Plaintiff maintains (and the district court found) that it learned in January 2002 that VWI and Cel-Net were operating kiosks under the VIRGIN WIRELESS name and two days later filed the present suit seeking to enjoin defendants from selling mobile phones in VIRGIN-branded retail stores.

On May 2, 2002, the district court considered plaintiff's application for a preliminary injunction. It found that no essential facts were in dispute, and therefore no evidentiary hearing was required. It was uncontested (and the district court accordingly found) that plaintiff sold "electronic apparatus" in its stores, including "various video game systems, portable cassette tape, compact disc, mp3, and mini disc players, portable radios, and disposable cameras," but not including telephones or telephone service, and that the only products the defendants sold in their stores were wireless telephones, telephone accessories, and wireless telephone services.

Noting that a party seeking a preliminary injunction must show the probability of irreparable harm in the absence of relief, and either (1) likelihood of success on the merits or (2) serious questions going to the merits and a balance of hardships tipping decidedly in its favor, the court found that plaintiff had failed to satisfy either standard. Arguing against plaintiff's likelihood of success, the court noted that plaintiff's registrations did not claim use of the VIRGIN mark "in telecommunications services or in the associated retail sale of wireless telephones and accessories." While plaintiff's 817 and 776 Registrations covered the retail sale of "computers and electronic apparatus," they did not extend to telecommunications services and wireless phones.

The court noted that the defendants were the first to use the VIRGIN mark in telecommunications, and the first to attempt to register VIRGIN for telecommunications and retail telephone sales. The court also observed that the dissimilarity in appearance of plaintiff's and defendants' logos and the differences between plaintiff's huge Virgin Megastores and defendants' small retail outlets in malls diminished likelihood of consumer confusion. Finally, because the defendants had expended substantial resources in pursuing their trademark applications and in establishing their retail presence, the court found that plaintiff could not demonstrate that the balance of hardships tipped in its favor.

The court denied the application for preliminary injunction. The crux of the court's decision lay in the facts that plaintiff's prior use and registration of the VIRGIN mark in connection with the sale of consumer electronic equipment did not include the sale of telephones or telephone services, and that defendants were the first to register and use VIRGIN for telephones and wireless telephone service. This appeal followed.

As the court below correctly noted, in order to obtain a preliminary injunction, a party must demonstrate probability of irreparable harm in the absence of injunctive relief, and either a likelihood that it will succeed on the merits of its claim, or a serious question going to the merits and a balance of hardships tipping decidedly in its favor. See Jackson Dairy, Inc. v. H.P. Hood & Sons, Inc., 596 F.2d 70, 72 (2d Cir. 1979) (per curiam); Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1038 (2d Cir. 1992). We review the court's denial of a preliminary injunction for abuse of discretion. Bristol-Myers, 973 F.2d at 1038.

In an action for trademark infringement, where a mark merits protection, a showing that a significant number of consumers are likely to be confused about the source of the goods identified by the allegedly infringing mark is generally sufficient to demonstrate both irreparable harm and a likelihood of success on the merits. Bristol-Myers, 973 F.2d at 1038; W. Publ'g Co. v. Rose Art Indus., Inc., 910 F.2d 57, 59 (2d Cir. 1990). Thus, our inquiry must be whether the district court correctly determined that the plaintiff was not entitled to protection from use of its mark by others in the sale of wireless telephones and related services, and that there was no likelihood that, in the absence of a preliminary injunction, a significant number of consumers would be confused about the sponsorship of defendants' retail stores. For the reasons discussed below, we find that the mark is entitled to protection, and there is a significant likelihood of confusion. We reverse and remand.

A claim of trademark infringement, whether brought under 15 U.S.C. § 1114(1) (for infringement of a registered mark) or 15 U.S.C. § 1125(a) (for infringement of rights in a mark acquired by use), is analyzed under the familiar two-prong test described in Gruner + Jahr USA Publ'g v. Meredith Corp., 991 F.2d 1072 (2d Cir. 1993). See Time, Inc. v. Petersen Publ'g Co. L.L.C., 173 F.3d 113, 117 (2d Cir. 1999) (noting that Gruner test is applicable to claims brought under § 1114(1) and § 1125(a)). The test looks first to whether the plaintiff's mark is entitled to protection, and second to whether defendant's use of the mark is likely to cause consumers confusion as to the origin or sponsorship of the defendant's goods. Gruner, 991 F.2d at 1074. Examining the question as the test dictates, we have no doubt that plaintiff was entitled to a preliminary injunction.

We believe the district court accorded plaintiff too narrow a scope of protection for its famous, arbitrary, and distinctive mark. There could be no dispute that plaintiff prevailed as to the first prong of the test — prior use and ownership. For years, plaintiff had used the VIRGIN mark on huge, famous stores selling, in addition to music recordings, a variety of consumer electronic equipment. At the time the defendants began using VIRGIN, plaintiff owned rights in the mark. The focus of inquiry thus turns to the second prong of the test — whether defendants' use of VIRGIN as a mark for stores selling wireless telephone services and phones was likely to cause confusion. There can be little doubt that such confusion was likely.

The landmark case of Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492 (2d Cir. 1961) (Friendly, J.), outlined a series of nonexclusive factors likely to be pertinent in addressing the issue of likelihood of confusion, which are routinely followed in such cases. See, e.g., Streetwise Maps, Inc. v. Vandam, Inc., 159 F.3d 739, 743-46 (2d Cir. 1998); Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 391-99 (2d Cir. 1995); Mobil Oil Corp. v. Pegasus Petroleum Corp., 818 F.2d 254, 256-60 (2d Cir. 1987).

Six of the Polaroid factors relate directly to the likelihood of consumer confusion. These are the strength of the plaintiff's mark; the similarity of defendants' mark to plaintiff's; the proximity of the products sold under defendants' mark to those sold under plaintiff's; where the products are different, the likelihood that plaintiff will bridge the gap by selling the products being sold by defendants; the existence of actual confusion among consumers; and the sophistication of consumers. Of these six, all but the last (which was found by the district court to be neutral) strongly favor the plaintiff. The remaining two Polaroid factors, defendants' good or bad faith and the quality of defendants' products, are more pertinent to issues other than likelihood of confusion, such as harm to plaintiff's reputation and choice of remedy. We conclude that the Polaroid factors powerfully support plaintiff's position.

Strength of the mark. The strength of a trademark encompasses two different concepts, both of which relate significantly to likelihood of consumer confusion. The first and most important is inherent strength, also called "inherent distinctiveness." This inquiry distinguishes between, on the one hand, inherently distinctive marks — marks that are arbitrary or fanciful in relation to the products (or services) on which they are used — and, on the other hand, marks that are generic, descriptive or suggestive as to those goods. The former are the strong marks. Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976). The second sense of the concept of strength of a mark is "acquired distinctiveness," i.e., fame, or the extent to which prominent use of the mark in commerce has resulted in a high degree of consumer recognition. See TCPIP Holding Co. v. Haar Communications Inc., 244 F.3d 88, 100 (2d Cir. 2001) (describing these two concepts of strength); Streetwise Maps, 159 F.3d at 743.

Considering first inherent distinctiveness, the law accords broad, muscular protection to marks that are arbitrary or fanciful in relation to the products on which they are used, and lesser protection, or no protection at all, to marks consisting of words that identify or describe the goods or their attributes. The reasons for the distinction arise from two aspects of market efficiency. The paramount objective of the trademark law is to avoid confusion in the marketplace. The purpose for which the trademark law accords merchants the exclusive right to the use of a name or symbol in their area or commerce is identification, so that the merchants can establish goodwill for their goods based on past satisfactory performance, and the consuming public can rely on a mark as a guarantee that the goods or services so marked come from the merchant who has been found to be satisfactory in the past. See Estee Lauder Inc. v. The Gap, Inc., 108 F.3d 1503, 1510 (2d Cir. 1997) (quoting Restatement (Third) of Unfair Competition § 21 comment i (1995)); Power Test Petroleum Distribs., Inc. v. Calcu Gas, Inc., 754 F.2d 91, 97 (2d Cir. 1985); McGregor-Doniger Inc. v. Drizzle Inc., 599 F.2d 1126, 1131 (2d Cir. 1979). At the same time, efficiency and the public interest require that every merchant trading in a class of goods be permitted to refer to the goods by their name, and to make claims about their quality. Thus, a merchant who sells pencils under the trademark Pencil or Clear Mark, for example, and seeks to exclude other sellers of pencils from using those words in their trade, is seeking an advantage the trademark law does not intend to offer. To grant such exclusivity would deprive the consuming public of the useful market information it receives where every seller of pencils is free to call them pencils. Abercrombie, 537 F.2d at 9; CES Publ'g Corp. v. St. Regis Publ'ns, Inc., 531 F.2d 11, 13 (2d Cir. 1975). The trademark right does not protect the exclusive right to an advertising message — only the exclusive right to an identifier, to protect against confusion in the marketplace. Thus, as a matter of policy, the trademark law accords broader protection to marks that serve exclusively as identifiers and lesser protection where a grant of exclusiveness would tend to diminish the access of others to the full range of discourse relating to their goods. See TCPIP, 244 F.3d at 100; Nabisco, Inc. v. PF Brands, Inc., 191 F.3d 208, 215 (2d Cir. 1999); Otokoyama Co. Ltd. v. Wine of Japan Import, Inc., 175 F.3d 266, 270 (2d Cir. 1999).

The second aspect of efficiency that justifies according broader protection to marks that are inherently distinctive relates directly to the likelihood of confusion. If a mark is arbitrary or fanciful, and makes no reference to the nature of the goods it designates, consumers who see the mark on different objects offered in the marketplace will be likely to assume, because of the arbitrariness of the choice of mark, that they all come from the same source. For example, if consumers become familiar with a toothpaste sold under an unusual, arbitrary brand name, such as ZzaaqQ, and later see that same inherently distinctive brand name appearing on a different product, they are likely to assume, notwithstanding the product difference, that the second product comes from the same producer as the first. The more unusual, arbitrary, and fanciful a trade name, the more unlikely it is that two independent entities would have chosen it. In contrast, every seller of foods has an interest in calling its product "delicious." Consumers who see the word delicious used on two or more different food products are less likely to draw the inference that they must all come from the same producer. Cf. Streetwise Maps, 159 F.3d at 744 (noting that several map producers use "street" in product names; thus plaintiff's mark using "street" was not particularly distinctive); W. Publ'g, 910 F.2d at 61 (noting numerous registrations of marks using word "golden"). In short, the more distinctive the mark, the greater the likelihood that the public, seeing it used a second time, will assume that the second use comes from the same source as the first. The goal of avoiding consumer confusion thus dictates that the inherently distinctive, arbitrary, or fanciful marks, i.e., strong marks, receive broader protection than weak marks, those that are descriptive or suggestive of the products on which they are used. See Abercrombie, 537 F.2d at 9-11; TCPIP, 244 F.3d at 100-01.

The second sense of trademark strength, fame, or "acquired distinctiveness," also bears on consumer confusion. See TCPIP, 244 F.3d at 100-01; Streetwise Maps, 159 F.3d at 744. If a mark has been long, prominently and notoriously used in commerce, there is a high likelihood that consumers will recognize it from its prior use. Widespread consumer recognition of a mark previously used in commerce increases the likelihood that consumers will assume it identifies the previously familiar user, and therefore increases the likelihood of consumer confusion if the new user is in fact not related to the first. See Nabisco, 191 F.3d at 216-17. A mark's fame also gives unscrupulous traders an incentive to seek to create consumer confusion by associating themselves in consumers' minds with a famous mark. The added likelihood of consumer confusion resulting from a second user's use of a famous mark gives reason for according such a famous mark a broader scope of protection, at least when it is also inherently distinctive. See McGregor, 599 F.2d at 1132 (noting that secondary meaning may further enlarge the scope of protection accorded to inherently distinctive marks).

Plaintiff's VIRGIN mark undoubtedly scored high on both concepts of strength. In relation to the sale of consumer electronic equipment, the VIRGIN mark is inherently distinctive, in that it is arbitrary and fanciful; the word "virgin" has no intrinsic relationship whatsoever to selling such equipment. Because there is no intrinsic reason for a merchant to use the word "virgin" in the sale of consumer electronic equipment, a consumer seeing VIRGIN used in two different stores selling such equipment will likely assume that the stores are related.

Plaintiff's VIRGIN mark was also famous. The mark had been employed with world-wide recognition as the mark of an airline and as the mark for megastores selling music recordings and consumer electronic equipment. The fame of the mark increased the likelihood that consumers seeing defendants' shops selling telephones under the mark VIRGIN would assume incorrectly that defendants' shops were a part of plaintiff's organization. See Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 873 (2d Cir. 1986).

There can be no doubt that plaintiff's VIRGIN mark, as used on consumer electronic equipment, is a strong mark, as the district court found. It is entitled as such to a broad scope of protection, precisely because the use of the mark by others in connection with stores selling reasonably closely related merchandise would inevitably have a high likelihood of causing consumer confusion.

Similarity of marks. When the secondary user's mark is not identical but merely similar to the plaintiff's mark, it is important to assess the degree of similarity between them in assessing the likelihood that consumers will be confused. See McGregor, 599 F.2d at 1133. Plaintiff's and defendants' marks were not merely similar; they were identical to the extent that both consisted of the same word, "virgin."

The district court believed this factor did not favor plaintiff because it found some differences in appearance. Defendants' logo used a difference typeface and different colors from plaintiff's. While those are indeed differences, they are quite minor in relation to the fact that the name being used as a trademark was the same in each case.

Advertisement and consumer experience of a mark do not necessarily transmit all of the mark's features. Plaintiff, for example, advertised its Virgin Megastores on the radio. A consumer who heard those advertisements and then saw the defendants' installation using the name VIRGIN would have no way of knowing that the two trademarks looked different. See Sports Auth., Inc. v. Prime Hospitality Corp., 89 F.3d 955, 962 (2d Cir. 1996). A consumer who had visited one of plaintiff's Virgin Megastores and remembered the name would not necessarily remember the typeface and color of plaintiff's mark. The reputation of a mark also spreads by word of mouth among consumers. One consumer who hears from others about their experience with Virgin stores and then encounters defendants' Virgin store will have no way knowing of the differences in typeface. See Hills Bros. Coffee, Inc. v. Hills Supermarkets, Inc., 428 F.2d 379, 381 (2d Cir. 1970) (per curiam).

In view of the fact that defendants used the same name as plaintiff, we conclude the defendants' mark was sufficiently similar to plaintiff's to increase the likelihood of confusion. This factor favored the plaintiff as a matter of law. We conclude that the district court erred in concluding otherwise on the basis of comparatively trivial and often irrelevant differences.

Proximity of the products and likelihood of bridging the gap. The next factor is the proximity of the products being sold by plaintiff and defendant under identical (or similar) marks. See Arrow Fastener, 59 F.3d at 396. This factor has an obvious bearing on the likelihood of confusion. When the two users of a mark are operating in completely different areas of commerce, consumers are less likely to assume that their similarly branded products come from the same source. In contrast, the closer the secondary user's goods are to those the consumer has seen marketed under the prior user's brand, the more likely that the consumer will mistakenly assume a common source. See Cadbury Beverages, Inc. v. Cott Corp., 73 F.3d 474, 480-81 (2d Cir. 1996).

While plaintiff had not sold telephones or telephone service prior to defendant's registration evincing intent to sell those items, plaintiff had sold quite similar items of consumer electronic equipment. These included computer video game systems, portable cassette-tape players, compact disc players, MP3 players, mini-disc players, and disposable cameras. Like telephones, many of these are small consumer electronic gadgets making use of computerized audio communication. They are sold in the same channels of commerce. Consumers would have a high expectation of finding telephones, portable CD players, and computerized video game systems in the same stores. We think the proximity in commerce of telephones to CD players substantially advanced the risk that consumer confusion would occur when both were sold by different merchants under the same trade name, VIRGIN.

Our classic Polaroid test further protects a trademark owner by examining the likelihood that, even if the plaintiff's products were not so close to the defendants' when the defendant began to market them, there was already a likelihood that plaintiff would in the reasonably near future begin selling those products. See Cadbury Beverages, 73 F.3d at 482. VEL's claim of proximity was further strengthened in this regard because, as the district court expressly found, "plans had been formulated [for VEL] to enter [the market for telecommunications products and services] shortly in the future." VEL had already begun marketing telephone service in England which would operate in the United States, and, as the district court found, had made plans to sell telephones and wireless telephone service under the VIRGIN name from its retail stores.

The district court, nonetheless, found in favor of the defendants with respect to the proximity of products and services. We would ordinarily give considerable deference to a factual finding on this issue. Here, however, we cannot do so because it appears the district court applied the wrong test. The court did not assess the proximity of defendants' VIRGIN-branded retail stores selling telephone products to plaintiff's VIRGIN-branded retail stores selling other consumer electronic products. It simply concluded that, because defendants were selling exclusively telephone products and services, and plaintiff's electronic products did not include telephones or related services, the defendants must prevail as to the proximity factor.

This represents a considerable misunderstanding of the Polaroid test. The famous list of factors of likely pertinence in assessing likelihood of confusion in Polaroid was specially designed for a case like this one, in which the secondary user is not in direct competition with the prior user, but is selling a somewhat different product or service. In Polaroid, the plaintiff sold optical and camera equipment, while the defendant sold electronic apparatus. The test the court discussed was expressly addressed to the problem "how far a valid trademark shall be protected with respect to goods other than those to which its owner has applied it." 287 F.2d at 495 (emphasis added); see also Arrow Fastener, 59 F.3d at 396 (noting that products need not actually compete with each other). The very fact that the test includes the "proximity" between the defendant's products and the plaintiff's and the likelihood that the plaintiff will "bridge the gap" makes clear that the trademark owner does not lose, as the district court concluded, merely because it has not previously sold the precise good or service sold by the secondary user.

In our view, had the district court employed the proper test of proximity, it could not have failed to find a high degree of proximity as between plaintiff VEL's prior sales of consumer electronic audio equipment and defendants' subsequent sales of telephones and telephone services, which proximity would certainly contribute to likelihood of consumer confusion. And plaintiff was all the more entitled to a finding in its favor in respect of these matters by virtue of the fact, which the district court did find, that at the time defendants began using the VIRGIN mark in the retail sale of telephones and telephone services, plaintiff already had plans to bridge the gap by expanding its sales of consumer electronic equipment to include sales of those very goods and services in the near future. Consumer confusion was more than likely; it was virtually inevitable.

Actual confusion. It is self-evident that the existence of actual consumer confusion indicates a likelihood of consumer confusion. Nabisco, 191 F.3d at 228. We have therefore deemed evidence of actual confusion "particularly relevant" to the inquiry. Streetwise Maps, 159 F.3d at 745.

Plaintiff submitted to the district court an affidavit of a former employee of defendant Cel-Net, who worked at a mall kiosk branded as Virgin Wireless, which stated that individuals used to ask him if the kiosk was affiliated with plaintiff's VIRGIN stores. The district court correctly concluded that this evidence weighed in plaintiff's favor.

Sophistication of consumers. The degree of sophistication of consumers can have an important bearing on likelihood of confusion. Where the purchasers of a products are highly trained professionals, they know the market and are less likely than untrained consumers to be misled or confused by the similarity of different marks. The district court recognized that " [r]etail customers, such as the ones catered to by both the defendants and [plaintiff], are not expected to exercise the same degree of care as professional buyers, who are expected to have greater powers of discrimination." On the other hand, it observed that purchasers of cellular telephones and the service plans were likely to give greater care than self-service customers in a supermarket. Noting that neither side had submitted evidence on the sophistication of consumers, the court made no finding favoring either side. We agree that the sophistication factor is neutral in this case.

Bad faith and the quality of the defendants' services or products. Two factors remain of the conventional Polaroid test: the existence of bad faith on the part of the secondary user and the quality of the secondary user's products or services. Polaroid, 287 F.2d at 495. Neither factor is of high relevance to the issue of likelihood of confusion. A finding that a party acted in bad faith can affect the court's choice of remedy or can tip the balance where questions are close. It does not bear directly on whether consumers are likely to be confused. See TCPIP, 244 F.3d at 102. The district court noted some evidence of bad faith on the defendants' part, but because the evidence on the issue was scant and equivocal, the court concluded that such a finding "at this stage [would be] speculative." The court therefore found that this factor favored neither party.

The issue of the quality of the secondary user's product goes more to the harm that confusion can cause the plaintiff's mark and reputation than to the likelihood of confusion. See Arrow Fastener, 59 F.3d at 398 (noting that first user's reputation may be harmed if secondary user's goods are of poor quality). In any event, the district court found this factor to be "neutral" with respect to likelihood of confusion.

* * * * * *

In summary we conclude that of the six Polaroid factors that pertain directly to the likelihood of consumer confusion, all but one favor the plaintiff, and that one — sophistication of consumers — is neutral. The plaintiff is strongly favored by the strength of its mark, both inherent and acquired; the similarity of the marks; the proximity of the products and services; the likelihood that plaintiff would bridge the gap; and the existence of actual confusion. None of the factors favors the defendant. The remaining factors were found to be neutral. Although we do not suggest that likelihood of confusion may be properly determined simply by the number of factors in one party's favor, the overall assessment in this case in our view admits only of a finding in plaintiff's favor that defendants' sale of telephones and telephone-related services under the VIRGIN mark was likely to cause substantial consumer confusion.

One issue remains. Defendants argue that plaintiff should be barred by laches from seeking injunctive relief. They contend that because of plaintiff's delay after learning of the defendants' applications to register the VIRGIN marks, they expended considerable sums and developed goodwill in their use of the VIRGIN marks before plaintiff brought suit. Because the district court ruled in the defendants' favor it made no express finding on the issue of laches. But the district court explicitly found that plaintiff first learned of defendants' use of the name VIRGIN in commerce only two days before plaintiff instituted this suit. Given that finding, plaintiff could not be chargeable with laches.

We conclude that, as a matter of law, plaintiff demonstrated irreparable harm and likelihood of success on the merits and was entitled to a preliminary injunction.

REVERSED and REMANDED.

 1

Immediately following oral argument, we issued a summary order, opinion to follow, directing the district court to enter a preliminary injunction

 2

Because of Pennie & Edmonds's involvement in searching marks for defendants, and because of the factual dispute about whether plaintiff's counsel searched the VIRGIN mark for defendants, defendants sought in the district court to have Pennie & Edmonds disqualified from representing plaintiff. The district court denied this motion on March 13, 2002