Defendant The Gap, Inc. d/b/a Old Navy Clothing Company ("Gap"), appeals from a final judgment of the United States District Court for the Southern District of New York, Lewis A. Kaplan, Judge, permanently enjoining Gap from using the term "100%" alone or with other terms as a trademark in connection with the manufacture, promotion, or distribution of its personal care products and bathroom furnishings and implements. The district court found that Gap's use of terms such as "100% BODY CARE" on its products was likely to cause consumers to confuse Gap's products with a moisturizer product manufactured by plaintiff Estee Lauder Inc. ("Lauder") and marketed under Lauder's trademark "100%." On appeal, Gap contends principally that the district court erred in finding Lauder's "100%" mark protectable and in finding that Gap's use of a mark that includes "100%" in conjunction with other terms would create a likelihood of consumer confusion. We agree that the district court erred in its likelihood-of-confusion analysis, and we accordingly reverse.

Lauder is an international cosmetics company that manufactures skin care, makeup, and fragrance products. The company distributes its products through cosmetics specialty stores and "upscale" department stores that, in its view, complement the image of quality it cultivates. Gap is an international retailer of apparel and other products. Gap owns and operates Old Navy Clothing Company ("Old Navy"), whose stores are located primarily in strip malls and shopping centers and cater to the mass middle market. All Old Navy products, and virtually no others, are distributed through Old Navy stores.

In December 1994, Gap developed a line of personal care products to be distributed through Old Navy, intending to market, inter alia, shampoo, soap, and body lotion under the label "100% BODY CARE." In September 1995, Gap began a search into the availability of a trademark containing the term "100%." In the meantime, Lauder was developing a facial moisturizer, and in December 1995 it began its own investigation into the availability of the term "100%" as the trademark for its moisturizer.

Both companies eventually learned that a third company, Les Parfums de Dana, Inc. ("Dana"), was the record owner of the trademark "100% HUNDRED PER CENT," but that Dana had abandoned use of its trademark. Lauder obtained an assignment of the "100%" trademark and registration from Dana, and in December 1995 Lauder filed an intent-to-use ("ITU") application with the United States Patent and Trademark Office seeking to register "100%" as a trademark. In January 1996, Gap filed a similar application to register "100%" alone; it also filed applications to register "100% BODY CARE" and "100% BODY CARE AND DESIGN." Thereafter, it filed numerous additional applications to register "100%" in combination with other terms such as "100% SPORT," "100% BODY CARE FOR KIDS," and "100% HOME SCENT."

In March 1996, Gap contacted Lauder to discuss the two companies' conflicting ITU applications. In light of Lauder's intent to use "100%" as the mark for its moisturizer, Gap offered assurance that for Gap's own products it would not use the term "100%" alone, and it abandoned its ITU application for the mark "100%" alone. Gap subsequently modified its planned "100% BODY CARE" mark to increase emphasis on the words "BODY CARE" and to decrease emphasis on the number "100." The parties failed to reach agreement.

Lauder began to distribute its moisturizer in April 1996, packaged in a 1.7-ounce aquamarine glass bottle, as part of its "blue line" of products that are typically marketed in aquamarine packaging. Lauder's moisturizer has a suggested retail price of $32.50. As the packaging is shown in Appendix A to this opinion, Lauder's "EL" logo is prominently displayed at the top of the bottle, and the company's "ESTEE LAUDER" house mark appears immediately below that. Near the bottom, the following lines appear:

100%

with BioMineral

The back of the bottle contains the notation: "(C) ESTEE LAUDER, DIST."

Gap's "100% BODY CARE" products are packaged principally in plastic bottles that range in size from .15 to 33 fluid ounces. At the top of each bottle appears a generic designation of its contents (e.g., "bubble bath," "shampoo," "body lotion") in large, lower case letters. The suggestion of a fragrance for each product (e.g., "Groovy Grapefruit," "Some Are Pink," "Positively Purple") is printed in an elongated oval just above the center of the bottle. The bottles themselves are translucent, presenting the various colors of their respective contents. The logo

100%

appears towards the bottom of the bottle. As illustrated in Appendix B, the "%" sign is oversized, approximately three times the height of "100." The back of the bottle contains the Old Navy house mark, followed by the notation, "A division of Gap, Inc." Gap planned to release this line of products on a test basis in August 1996. The planned prices for each type and quantity of these products were small fractions of the suggested retail price of Lauder's product.

In June 1996, Lauder commenced the present action seeking preliminary and permanent injunctions against the use of "100%" as part of Gap's marks, alleging principally that such use would lead consumers to believe that Gap's products were associated with Lauder and would thereby violate Lauder's rights under § 43(a) of the Lanham Act (the "Act"), 15 U.S.C. § 1125(a) (1994). Gap disputed Lauder's assertions and counterclaimed for declaratory and injunctive relief against interference by Lauder with Gap's efforts to register various marks incorporating the term "100%." The district court ordered expedited discovery and conducted a two-day bench trial in July 1996.

In a posttrial Opinion dated July 22, 1996, reported at 932 F. Supp. 595, the court ruled that Lauder was entitled to a permanent injunction prohibiting Gap from including the term "100%" in trademarks on Gap's personal care and other products. The court found that Lauder's "100%" mark was suggestive and, therefore, protectable. The court reasoned:

The product here is a skin moisturizer. The designation "100%" has no particular meaning with respect to such a product and certainly does not convey specific product attributes. Indeed, Gap's own personal care products manager admitted at his deposition that "100% on its own is meaningless.".... Rather, "100%" suggests the idea of quality to consumers whose imaginations and thought processes are provoked.

932 F.Supp. at 608. The court also found that Lauder's mark was suggestive even if it were viewed not as "100%" standing alone but as "100% Time Release Moisturizer":

[T]he phrase, particularly given the manner in which it is displayed in Lauder's labelling, packaging, advertising, and promotion, is susceptible of three interpretations: (1) this bottle contains nothing but time release moisturizer, (2) this product moisturizes 100% of the time, and (3) this is 100% (the brand) time release moisturizer.... The third of these interpretations, Gap concedes, is suggestive.

Id. at 609. Finding that "the average prudent consumer would understand the phrase '100% time release moisturizer' in the suggestive sense referred to above," the district court held that Lauder's mark was protectable. Id. at 611.

Having noted Lauder's concession that its mark had not acquired secondary meaning, see, e.g., id. at 606, the court proceeded to address the factors set forth in Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2d Cir.), cert. denied, 368 U.S. 820, 82 S. Ct. 36, 7 L. Ed. 2d 25 (1961), in order to determine whether there was a likelihood of consumer confusion. First the court found that Lauder's mark was of "moderate strength" because it was "inherent[ly] distinctive[ ]," 932 F. Supp. at 612, and that the "prominent use of '100%' by both parties--particularly in view of the fact that each of the parties uses '100%' to convey the same idea in conjunction with closely related personal care products--makes the similarity factor weigh in Lauder's favor," id. at 613.

The court saw some obvious differences between Lauder's product and those of Gap:

Lauder's product is a high technology facial moisturizer sold in tiny elegant bottles containing 1.7 fluid ounces of product. Its suggested retail price is $32.50 per bottle. Gap's line includes a body lotion as well as other personal care products. It typically will be sold in much larger bottles at much lower prices. A consumer who had made an informed decision to buy either product would be unlikely to purchase the other by mistake.

Id. But the court found these differences to be outweighed by the fact that the products are closely related, i.e., that they are "all used for personal care," id. The court also found that, although there was some overlap in clientele, the two companies' products were sold through markedly different channels of distribution:

The difference between the Lauder and Gap products is nowhere greater than with respect to the channels through which they are distributed. Gap's products will be sold only in Gap's Old Navy stores which, with rare exceptions, sell nothing other than merchandise sold under the Old Navy name. Lauder, as noted, sells only through prestige retail stores-upscale department stores and the like. Moreover, the difference in the ambiance between the stores that handle Lauder's product and Old Navy stores is quite dramatic. Whereas Lauder limits distribution to outlets offering upscale images with a quite polished ambiance, Old Navy pursues a very different look and feel. Its stores feature merchandise displayed on industrial shelving and in other imaginative ways that best may be described as funky. Loud modern rock music is played. The floors are uncarpeted. Sales personnel dress informally, and many in the New York store wear headsets. No one knowledgeable about the sort of places in which Lauder sells its upscale products would be likely to think that the very same products would be available in an Old Navy store.

Id. at 614 (footnote omitted). However, the court found these differences outweighed by the fact that "[n]ot all customers are knowledgeable, and even knowledgeable customers could be misled into thinking that Lauder is associated with the '100% BODY CARE' products, either as producer or licensor of an off-price line." Id. The court also found that although there was no suggestion that Gap's products were not inherently of good quality for what they were, there was a "genuine risk that consumers who purchase the Gap product in the belief that it is associated in some way with Lauder will be disappointed and that the disappointment will rub off on Lauder." Id. at 616.

Although the court also found that Lauder had no plans to enter the mass middle market in which Old Navy normally operates, see id. at 615, that there was "no evidence of actual confusion," id., and that in seeking to use "100%" as part of its own marks, "Gap acted in good faith," id. at 616, all of which were factors that did not favor Lauder's claim, the court concluded, weighing all the factors, that Gap's use of "100%" in its marks would create a likelihood of consumer confusion:

Here, Lauder has adopted a mark of middling strength (which is likely to become stronger). The prominence given "100%" by both companies, although in ways that are graphically somewhat different, renders the marks sufficiently similar to cause significant concern. The products stand in close proximity to one another. Lauder is at risk that consumers encountering Gap's products will associate their quite different properties with the source of Lauder's products to Lauder's detriment. While the other Polaroid factors either are neutral or cut in Gap's favor, the Court finds in all the circumstances that an appreciable number of ordinarily prudent consumers are likely to be confused as to the sponsorship, affiliation or connection of the source of Lauder's "100%" product with Gap's "100% BODY CARE." The likelihood of confusion is pronounced in, but not limited to, the post-sale context, in which a consumer may see the Gap product outside the retail setting, wrongly associate the product with Lauder's, and be influenced by the association in a later purchasing decision.

Id. at 617 (footnote omitted).

A final judgment was entered dismissing Gap's counterclaim and permanently enjoining Gap from using the term "100%" alone or with other terms as a trademark in connection with the manufacture, promotion, or distribution of its personal care products and bathroom furnishings and implements. This appeal followed.

In order to prevail on a claim of trademark infringement in violation of the Lanham Act, a plaintiff must show (1) that it has a valid mark that is entitled to protection under the Act, and (2) that use of the defendant's mark infringes, or is likely to infringe, the mark of the plaintiff. See generally Sports Authority, Inc. v. Prime Hospitality Corp., 89 F.3d 955, 960 (2d Cir.1996); Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 390 (2d Cir.1995); Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d 1072, 1075 (2d Cir.1993). The degree to which a mark is entitled to protection under the Act depends on whether the mark is classified as (a) generic, (b) descriptive, (c) suggestive, or (d) fanciful or arbitrary. See, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391; Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1075. "The test for infringement is whether the actor's use of a designation as a trademark ... creates a likelihood of confusion...." Restatement (Third) of Unfair Competition § 21 comment a (1995) ("Restatement ").

There is no contention in the present case that Lauder's mark is generic, arbitrary, or fanciful. It is also undisputed that Lauder's mark had not acquired secondary meaning. Gap contends that the district court erred (1) in finding Lauder's mark suggestive rather than descriptive, and hence protectable in the absence of secondary meaning, and (2) in finding that there was a likelihood of consumer confusion in the absence of secondary meaning. We agree with the latter contention.

A mark is classified as descriptive if it "tells something about a product, its qualities, ingredients or characteristics." Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1076; see, e.g., Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1040 (2d Cir.1992); Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 11 (2d Cir.1976). A descriptive mark generally "may be protected only if it has acquired secondary meaning." Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1076; see, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391. Secondary meaning attaches to a mark when "the consuming public primarily associates the term with a particular source." Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d at 1040; see, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 390 (secondary meaning attaches when "the [term] and the business have become synonymous in the mind of the public" (internal quotation marks omitted)).

A suggestive mark is one that "suggests the product, though it may take imagination to grasp the nature of the product." Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1076; see, e.g., Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d at 1040 (a suggestive mark "requires imagination, thought and perception to reach a conclusion as to the nature of goods" (internal quotation marks omitted)); Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d at 11 (same). A term that is merely self-laudatory, such as "plus" or "super," seeking to convey the impression that a product is excellent or of especially high quality, is generally deemed suggestive. See, e.g., Plus Products v. Plus Discount Foods, Inc., 722 F.2d 999, 1005 (2d Cir.1983) ("PLUS," when applied to goods, "merely implies additional quantity or quality"); In re Ralston Purina Co., 191 U.S.P.Q. 237, 238 (T.T.A.B.1976) ("SUPER," when used to "connote a vague desirable characteristic or quality allegedly connected with [a] product," is suggestive). "[A] suggestive mark is protected without a showing of secondary meaning." Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1076; see, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391.

In determining whether a mark should be classified as descriptive or suggestive, the court must focus on how the mark is used in context, rather than on its use in the abstract. See, e.g., Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d at 1041; Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d at 12. The district court's finding as to a mark's proper classification will not be overturned unless it is clearly erroneous. See, e.g., Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d at 1039-40.

In the present case, the district court's characterization of Lauder's "100%" mark as suggestive, whether read alone or as part of the phrase "100% Time Release Moisturizer," is not clearly erroneous. Standing alone, "100%" does not describe qualities, ingredients, or characteristics. It implies that the product is of the highest quality; as the vice president in charge of Gap's Personal Care Department stated in describing his plan for Gap to use that term, 100% "suggest[s] that a product [i]s the best, [i]s all that a consumer could expect from such a product, and create[s] a positive association with 'giving it your all, or your 100%.' " But in order to have an idea of the nature of the product's qualities, ingredients, or characteristics, a consumer needs additional information.

The phrase "100% Time Release Moisturizer" as a whole is somewhat more descriptive, but the most descriptive parts are "Time Release" and "Moisturizer"; and the term 100% as a modifier of either is more suggestive than descriptive. The phrase "100% Time Release Moisturizer" could be read to indicate the purity of the moisturizing content of Lauder's product, or to imply an enduring effect. Or, as the district court found, it could be read as indicating that the bottle contains nothing but time-release moisturizer or that the product moisturizes continuously until removed or worn off. All of these interpretations require some stretch of the imagination. And of course, as the court found, if the term "100%" is simply viewed as the brand of time-release moisturizer, it plainly is suggestive.

In sum, we see no error in the court's finding that Lauder's use of "100%," either alone or as part of the phrase "100% Time Release Moisturizer," is suggestive rather than descriptive. Thus, despite the absence of secondary meaning, Lauder's mark is entitled to protection. Lauder was not entitled to relief in the present case, however, unless it established a likelihood of confusion.

The issue of likelihood of confusion turns on whether "numerous ordinar[il]y prudent purchasers are likely to be misled or confused as to the source of the product in question because of the entrance in the marketplace of defendant's mark." Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1077; see, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 390-91; Plus Products v. Plus Discount Foods, Inc., 722 F.2d at 1003. Likelihood of confusion means a probability of confusion; "it is not sufficient if confusion is merely 'possible'." 3 J. McCarthy, McCarthy on Trademarks and Unfair Competition § 23:2, at 23-10 to -11 (1996).

"[C]ourts deciding whether a plaintiff has established likelihood of confusion must consider the eight factors elaborated" in Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2d Cir.), cert. denied, 368 U.S. 820, 82 S. Ct. 36, 7 L. Ed. 2d 25 (1961). Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391; see, e.g., Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1077. Those factors are: "1) the strength of the plaintiff's mark; 2) the similarity of plaintiff's and defendant's marks; 3) the competitive proximity of the products; 4) the likelihood that plaintiff will 'bridge the gap' and offer a product like defendant's; 5) actual confusion between products; 6) good faith on the defendant's part; 7) the quality of defendant's product; and 8) the sophistication of buyers." Id.; see, e.g., Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391. As to the first factor,

[t]he distinctiveness or "strength" of a mark measures its capacity to indicate the source of the goods or services with which it is used. The greater the distinctiveness of the mark, the greater the likelihood that prospective purchasers will associate the same or a similar designation found on other goods, services, or businesses with the prior user....

"[S]trength" ... ultimately depends on the degree to which the designation is associated by prospective purchasers with a particular source.

Restatement § 21 comment i.

No one of the Polaroid factors is dispositive, and the list is not exhaustive; "the analysis of the factors is not a mechanical process." Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391 (internal quotation marks omitted); see, e.g., Gruner + Jahr USA Publishing v. Meredith Corp., 991 F.2d at 1077. A trial court's finding as to each individual factor is one of fact, subject to review under the clearly erroneous standard. See, e.g., id.; Arrow Fastener Co. v. Stanley Works, 59 F.3d at 391. The ultimate weighing of the factors, however, is reviewed de novo. See, e.g., id. at 391, 400 (reversing after a de novo review of the court's weighing of the Polaroid factors); Plus Products v. Plus Discount Foods, Inc., 722 F.2d at 1004, 1005 (partial reversal based on conclusion that the district "court's balancing of the Polaroid factors was incorrect").

In the present case, the main Polaroid factors relied on by the district court in concluding that Gap's use of "100%" as part of its marks would create a likelihood of confusion were (1) the strength of Lauder's mark, (2) the similarity of the two companies' marks, and (3) the competitive proximity of the products. We have difficulty principally with the court's findings as to the first of these factors and with its weighing of the third against all other factors. First, the court found that Lauder's mark, though neither at the weakest nor at the strongest end of the range for suggestiveness, was moderately strong. The district court made this finding on the ground that Lauder's mark was "inherent[ly] distinctive[ ]." 932 F. Supp. at 612. These findings are unsupported by any evidence that consumers associated the term "100%" or the phrase "100% Time Release Moisturizer" with Lauder; indeed, they are undercut by Lauder's concession that its mark had not acquired secondary meaning, see id. at 606, 611-12. And while the absence of secondary meaning does not preclude the court from finding that a mark is strong if it is otherwise distinctive, see, e.g., W.W.W. Pharmaceutical Co. v. Gillette Co., 984 F.2d 567, 573 (2d Cir.1993), the court's finding of distinctiveness here is undermined by its own finding that Lauder's use of "100%" in a suggestive sense "is not wholly original," id. at 612. The latter finding is amply supported by the trial evidence that, prior to Lauder's ITU application, there were more than 70 trademark registrations, pending applications for registration or renewal, or publications-for-opposition that incorporated the term "100%," and that some of those marks were for cosmetic products.

Given the principle that a mark's " 'strength' ... ultimately depends on the degree to which the designation is associated by prospective purchasers with a particular source," Restatement § 21 comment i, plus the court's finding that Lauder's use of the term "100%" was not original and Lauder's concession that prospective purchasers did not associate that term with Lauder, we conclude that the court's finding that Lauder's mark is moderately strong is clearly erroneous.

Second, we think the court erred in weighing the similarity of the products against their differences and the disparate channels through which they are distributed. The court found that although the two lines of products are to be sold in mutually exclusive types of stores and "[n]o one knowledgeable about the sort of places in which Lauder sells its upscale products would be likely to think that the very same products would be available in an Old Navy store," confusion "may be" likely because some customers are not knowledgeable, and even knowledgeable customers "could" be misled, as the products are closely related. 932 F. Supp. at 614. The test, however, is not whether confusion is possible; nor is it whether confusion is probable among customers who are not knowledgeable. Rather, the test, correctly stated by the district court elsewhere in its opinion, is whether confusion is probable among numerous customers who are ordinarily prudent. Especially given the concession that consumers do not associate "100%" with Lauder, and the fact that each product is labeled to show which company is its source, we see no support for a finding that an appreciable number of ordinarily prudent consumers would likely think that Lauder, whose product costs $32.50 for less than two ounces in an upscale store, is the source of products that at Old Navy cost a small fraction of that price for 10 or 20 times that quantity.

The predominant element of the district court's conclusion that the use of "100%" by Gap would result in the likelihood of confusion seems to have been that Lauder's moisturizer and some of the products to be marketed by Gap are "closely related products." See, e.g., id. at 612 (lack of originality in Lauder's mark held unimportant in part because of Gap's "closely related products"); id. at 613 (different visual impressions created by the marks held outweighed in part by the fact that the marks are used with "closely related personal care products"); id. at 617 (the two companies' "graphically somewhat different" presentations of their respective marks offset by the fact that "[t]he products stand in close proximity to one another"); id. at 613 (improbability, in light of the price disparity, that a consumer who had made an informed decision to buy either company's product would buy the other by mistake held outweighed by the fact that the products are "closely related"); id. at 614 (fact that knowledgeable consumers would not seek a Lauder product at an Old Navy store, nor an Old Navy product in an upscale store carrying Lauder products, held outweighed by possible confusion from fact that the goods are "closely related"); id. at 615 (lack of any plans by Lauder to enter Gap's market, a factor in favor of Gap, held to be of limited significance because of "closely related products"). Having reviewed de novo the district court's weighing of the factors, we conclude that the close relationship between Lauder's moisturizer and some of Gap's products, and the fact that some consumers shop in both types of stores, do not outweigh the facts that the rendering of "100%" in the two sets of marks is different in appearance; that the appearance of the packaging overall is quite different in graphic design; that on each package the purveyor of that product is expressly identified; that the use of "100%" is not original; that the companies' products are sold in disparate types of stores; that the per-ounce price of Gap's body care products would likely be less than five percent of the per-ounce price of Lauder's product; that Lauder does not plan to enter Gap's market; that Gap did not act in bad faith; and that, as the district court found, no knowledgeable consumer would be likely to buy one product thinking it was the other. In sum, weighing all of the factors, we conclude that Gap's use of "100%" as part of its planned marks would not create a likelihood of confusion among ordinarily prudent consumers.

We have considered all of Lauder's arguments on this appeal and have found in them no basis for upholding the decision below. The judgment of the district court is reversed, and the matter is remanded for entry of an amended judgment dismissing the complaint.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLEAPPENDIX B

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE