717 F.2d 352
7th Cir.
1983

Lead Opinion

CUMMINGS, Chief Judge.

Kraft, Inc. (“Kraft”) appeals from a district court decision and order granting to Henri’s Food Products Co., Inc. (“Henri’s”) a declaratory judgment that its trademark Yogowhip used on spoonable salad dressing does not infringe Kraft’s registered trademark Miracle Whip or the word Whip, and denying Kraft’s trademark infringement counterclaim. The district court also denied Kraft’s counterclaim for injunctive relief against the mislabeling by Henri’s of a yogurt product as salad dressing, and refused to declare void and order cancellation of Henri’s trademark application for registration of a yogurt product as salad dressing.

In 1933 Kraft began marketing Miraole Whip Salad Dressing, a spoonable dressing. In the same year, Kraft was granted United States Trademark Registration 308,260 for the trademark Miraole Whip for salad dressing. In that application for registra*354tion, and in several later filings Kraft disclaimed any rights in the word Whip alone apart from the composite mark Miraole Whip as shown. The trademark Miraole Whip is incontestable under 15 U.S.C. § 1065. Not until 1978, after the commencement of this action, did Kraft use the word Whip alone on salad dressing.

Kraft began using “Miracle Whip” in the early 1930’s in connection with a new machine and method for the continuous production of mayonnaise. The machine was referred to as the Miracle Whip Machine, which performed “miracles in mayonnaise blending” (App. 3). In 1933 Kraft introduced Miracle Whip Salad Dressing, created in its Miracle Whip Machine. Advertisements indicated that Miraole Whip Salad Dressing was creamy, silky smooth and fluffy light.

In 1973 and 1974 Henri’s developed a successful line of pourable dressings made with yogurt. In 1976 Henri’s decided to enter the spoonable salad dressing and the mayonnaise markets with two new products made with yogurt: YOGONAISE and YOGO-WHIP. These products are sold primarily in the midwest. Henri’s attempted to sell its Yogowhip and Yogonaise products with the same marketing approach used for its pourable dressings made with yogurt: the yogurt dressings taste good but are lower in fat than the leading competitors in the marketplace. Since 1958 Henri’s has had a registered trademark for the logo, referred to as the “Frenchman,” which has been used or affixed in labeling and advertising for Henri’s Yogowhip product. Since 1968 Henri’s has had a registered trademark for the word “Henri’s” which has also been used or affixed in labeling and advertising for Henri’s Yogowhip product. In July of 1976 Henri’s filed an application to register Yogowhip for salad dressing; that application has been stayed pending resolution of this case. In August 1981, just before trial, Henri’s filed an application to register Yo-gowhip as a reduced calorie dressing for salads.

The central issue in this appeal is whether Henri’s Yogowhip infringes on Kraft’s Miracle Whip trademark. The test for trademark infringement under the Lanham Act, 15 U.S.C. § 1114(1), is “likelihood of confusion, deception or mistake on the part of the consuming public.” James Burrough Ltd. v. Sign of Beefeater, Inc., 540 F.2d 266, 274 (7th Cir.1976). The key question therefore is whether it is likely that the consuming public will be confused into believing that Henri’s Yogowhip comes from the same source as Miracle Whip Salad Dressing. Union Carbide Corp. v. Ever-Ready Inc., 531 F.2d 366, 381 (7th Cir.1976), certiorari denied, 429 U.S. 830, 97 S.Ct. 91, 50 L.Ed.2d 94. After reviewing the record, we cannot conclude that the district court clearly erred in finding no likelihood of confusion.

Infringement

In determining whether likelihood of confusion exists, this Circuit considers such factors as distinctiveness of the trademark in issue, similarity of the marks, similarity of the products, similarity in the channels of distribution, identity of advertising media utilized, the intent of the alleged infringer, and evidence of actual confusion. Union Carbide, supra, 531 F.2d at 381-382. The district court’s findings as to likelihood of confusion are findings of fact, and are subject to the clearly erroneous standard of review, Processed Plastic Co. v. Warner Communications, Inc., 675 F.2d 852, 857 (7th Cir.1982), though to the extent the determination is predicated on the similarity of the marks themselves, this Court is in as good a position as the trial judge to determine likelihood of confusion. Union Carbide, supra, 531 F.2d at 383. Kraft argues, and Henri’s does not really dispute, that several of the above factors weigh on the infringement side of the balance. The district court found, and Kraft emphasizes, that Miracle Whip is an “extremely strong trademark” (App. 14). Both Miraole Whip and Yogowhip are sold through the same channels of distribution and are found in the same section of the supermarket. The products are advertised in the same media. The products themselves are similar in that both are spoonable salad dressings, though *355Yogowhip contains yogurt and is a low fat dressing. On the more important factors— similarity of the marks, evidence of confusion, and intent of the alleged infringer— Kraft and Henri’s part company.

A. Similarity of the marks

An examination of MiRacle Whip and Yogowhip for similarity of “sound, sight and meaning,” Plough, Inc. v. Kreis Laboratories, 314 F.2d 635, 638 (9th Cir.1963); 1 J. Gilson, Trademark Protection and Practice § 5.02 (1979), reveals the following: Miracle Whip is two words and each word appears on a separate line. Yogowhip is one word and naturally appears on the same line. This distinction was given weight by the court in Plough, supra, 314 F.2d at 638, in comparing Coca Tan and Coca Tint to Coppertone. Miracle Whip has four syllables and Yogowhip has only three. The word Miracle is fanciful, revealing nothing about the product. The prefix Yogo on the other hand is descriptive, indicating a yogurt product. Miracle Whip and Yogowhip have the final syllable, Whip, in common, though as noted above it stands as a separate word in Miracle Whip.

Looking at the marks as a whole, Spice Islands, Inc. v. Frank Tea and Spice Co., 505 F.2d 1293, 1295 (Cust. & Pat.App.1974), we agree with the district court that the differences outweigh the similarities. This is not to say that two trademarks having a final syllable in common can never be sufficiently similar to lead to a likelihood of confusion (Kraft Br. 29). To the contrary, this Circuit in G.D. Searle & Co. v. Chas. Pfizer & Co., 265 F.2d 385, 387 (1959), held that Bonamine infringed Dramamine. In comparing the marks, the Court found that they

contain the same number of syllables; they have the same stress pattern, with primary accent on the first syllable and secondary accent on the third; the last two syllables of Dramamine and Bona-mine are identical. The initial sounds of Dramamine and Bonamine (“d” and “b”) are both what are known as “voiced plosives” and are acoustically similar; the consonants “m” and “n” are nasal sounds and are acoustically similar.

See also id. at 388-389 and cases cited therein. However, the Court also noted that reference to decisions as to other trademarks is not of great help because each case is dependent on its own facts, id. at 389. A comparison of the marks in this ease shows far less similarity than in G.D. Searle, supra. The marks must be so similar that it is likely or probable, not just possible, that consumers, when presented with the marks singly, rather than side-by-side, would confuse Henri’s Yogowhip with the source of Miracle Whip. See James Burrough, supra, 540 F.2d at 275. Based on our examination of the marks, we think that a purchaser, even a supermarket shopper who will make a selection quickly and without careful examination, is likely to differentiate between the products.

This finding is further supported by the differences in the products’ labels. Despite Kraft’s objection to the district court’s analysis of the labels (Br. 32-34), there is authority for the proposition that a comparison of the labels rather than simply the trademarks is appropriate. See McGregor-Doniger Inc. v. Drizzle Inc., 599 F.2d 1126, 1133 (2d Cir.1979); Sun-Maid Raisin Growers v. Sunaid Food Prods., Inc., 356 F.2d 467, 469 (5th Cir.1966); John Morrell & Co. v. Reliable Packing Co., 295 F.2d 314, 317 (7th Cir.1961); Ye Olde Tavern Cheese Prods., Inc. v. Planters Peanuts Div., 261 F.Supp. 200, 206 (N.D.Ill.1966), affirmed, 394 F.2d 833 (7th Cir.1967). When a prospective purchaser goes to the supermarket to buy salad dressing, it is the label that the purchaser sees. Sun-Maid, supra, 356 F.2d at 469. Kraft is of course correct in stating that in various fact situations the presence of a newcomer’s housemark may enhance rather than lessen the likelihood of confusion. See W.E. Bassett Co. v. Revlon, Inc., 435 F.2d 656, 662 (2d Cir.1970); A.T. Cross Co. v. Jonathan Bradley Pens, Inc., 470 F.2d 689, 692 (2d Cir.1972). Yet even Kraft agrees that a prominent housemark may also tend to lessen confusion. (Br.33). Here, though the “Henri’s” trademark is in smaller type than the word Yogowhip, it *356appears in thick, bold, black print under the “Frenchman” trademark on the Yogowhip label. In addition, the district court found the Yogowhip label distinct from the Miracle Whip label because of the green colored border, as opposed to the blue, scalloped Miraole Whip border, and the addition on the Yogowhip label of the phrases “made with non-fat yogurt” and “70% fat free” (App. 14). While a comparison of labels is not conclusive, the district court did not err in considering them, and the differences between the marks and the labels are factors weighing against infringement.

Kraft argues that the district court erroneously dissected the marks when comparing them and unjustifiably found the word Miracle to be the dominant portion of the trademark Miracle Whip. However, this Circuit has recognized the rule that if one word or feature of a composite trademark is the salient portion of the mark, it may be given greater weight than the surrounding elements. Union Carbide, supra, 531 F.2d at 383; Independent Nail & Packing Co., Inc. v. Stronghold Screw Prods., Inc., 205 F.2d 921, 924 (7th Cir.1953). See also 1 J. Gilson, supra, § 5.03. When one portion of a composite mark is a descriptive or generic word, that feature of the mark may be of less significance in designating a source of origin. See American Aloe Corp. v. Aloe Creme Laboratories, Inc., 420 F.2d 1248, 1256 (7th Cir.1970) (Tan Life does not infringe Fashion Tan or After Tan); Fleetwood Co. v. Hazel Bishop, Inc., 352 F.2d 841, 844 (7th Cir.1965) (Tintstik does not infringe Tintz Stick). The district court held that Whip was a merely descriptive term that had not achieved secondary meaning as a source indicator (App. 8-13), and apparently relied on that holding to conclude that Miracle was the dominant or salient portion of the mark (App. 14-15). Kraft essentially objects to the characterization of Whip as the weak portion of the mark in the face of evidence of secondary meaning.

It is not necessary to decide whether the district court was clearly erroneous in its finding of no secondary meaning because this Court has itself re-examined the trademarks de novo to determine the extent of similarity.1 A finding of secondary meaning in the word Whip would have no impact on our comparison of the marks, regardless of whether or not it infected the district court’s comparison. After comparing the trademarks and the labels, we conclude that their dissimilarity must be counted as weighing against infringement.

B. Evidence of confusion

Kraft presented no evidence of actual confusion. Kraft did, however, rely on surveys taken by Henri’s to show likelihood of confusion. Henri’s through its counsel commissioned Market Facts, Inc. to conduct two confusion surveys and two secondary meaning surveys. The first confusion survey, the Bendikas survey, was conducted for Henri’s attorneys in order to help plan the overall strategy for the case. Mr. Bendikas testified that the preliminary survey was designed to be biased in favor of Kraft (Tr. 1242-1245), and Kraft’s survey expert agreed that the first survey was biased against Henri’s (Tr. 1261). The survey was biased in the following way: respondents were asked inter alia

3. When thinking about food products, what does the word “whip” mean to you? Miracle Whip 13%
4. What kinds of food products do you think of in connection with the word “whip?” Miracle Whip 16%
5. When thinking about mayonnaise-type spoonable salad dressings, what does the word “whip” mean to you? Miracle Whip 38%

Prior to asking the sixth question on the Bendikas survey, respondents were shown a jar with a label identifying Henri’s Yogowhip for about five seconds and asked “who *357do you believe makes this product?” According to Mr. Bendikas (Tr. 1248), if he had “wanted to measure the confusion purely, [Question Six] should not be preceded by a sequence of questions that would give the respondent a chance to take the exam over and over again up to that point.” The only way to elicit objective responses to Question Six “would be not to contaminate the respondent’s mind or witness to give answers by preceding questions” (Tr. 1245). Cf. Sears, Roebuck & Co. v. All States Life Insur. Co., 246 F.2d 161, 171-172 (5th Cir.1957). Despite this bias toward Kraft, only 11% of the respondents answered Kraft or Miracle Whip to Question Six.

The second confusion survey, the Hardin survey, was conducted in 1980 by face-to-face interviews with 1,068 female respondents at shopping malls in Milwaukee, Minneapolis and Grand Rapids. These areas were chosen because both Kraft’s Miracle Whip Salad Dressing and Henri’s Yogowhip have been sold in these markets. Respondents were shown a jar of Henri’s Yogow-hip and asked, “Who puts out this product?” 72.4% of the respondents identified Henri’s as the company that puts out the Yogowhip product. 18.3% said they did not know who puts out the product. 7% thought that Kraft, Miracle Whip or Kraft Miracle Whip put out the product. The remaining 2.3% named other brands, including Hellman’s. The 93% of the respondents who did not answer Kraft or Miracle Whip to the first question were then asked, “What other products, if any, are put out by the company that puts out this product?” An additional .6% of the respondents named Miracle Whip as such a product. Thus 7.6% of the respondents believed that Henri’s Yo-gowhip was produced by Kraft or the same company that produced Miracle Whip Salad Dressing.

Henri’s introduced the Hardin survey to show that there was no likelihood of confusion; Kraft introduced the Bendikas survey to show that there was likelihood of confusion. Kraft also caused its own survey to be done, but did not introduce it into evidence. The district court did not mention the Bendikas survey, evidently giving it no weight, but concluded that the Hardin survey supported a finding that the trademark Yogowhip does not cause confusion in consumers’ minds.

After reviewing the survey evidence, we conclude that the district court was correct in discrediting the Bendikas survey. The undisputed testimony indicates that the preliminary survey was biased against Henri’s and in favor of Kraft on the question of likelihood of confusion. On the other hand, we do not think the district court erred in giving some weight to the Hardin survey, as corroborating the finding of no confusion. Kraft points to several flaws in the survey which caution against giving it too much weight; we cannot conclude, however, that the survey should be ignored.

Kraft’s expert criticized the survey because the sites selected for interviews contained supermarkets in which Henri’s products were visible to the shopper before the interview. He also criticized as unrealistic the fact that respondents were shown the Henri’s Yogowhip jar for five seconds, testifying that the average person would take less time to purchase a product off the supermarket shelf (Tr. 1261-1263). Kraft argues that the survey should have been conducted using a ,jar bearing 'the word Yogowhip alone, rather than an actual jar reading “Henri’s Yogowhip.” Finally, Kraft claims that the question: “Who puts out this product?” is misleading because respondents who answered “Henri’s” could have been referring to Henri’s as a licensee of Kraft.

These criticisms are certainly not without merit. However, Kraft had the opportunity to conduct and introduce its own survey to remedy the timing and site selection inadequacies, but chose not to do so. With respect to the survey question, this Court in Union Carbide, supra, 531 F.2d at 385-386, found the trial court clearly erred by not crediting a survey posing similar questions: “Who do you think puts out the lamp shown here?” and “Please name any other products put out by the same concern which puts out the lamp shown here.” Although

*358this phraseology may not be the best way to ask the question, it does not mandate rejection of the survey. Neither does the fact that respondents were shown a jar marked “Henri’s Yogowhip.” In McDonough Power Equip., Inc. v. Weed Eater, Inc., 208 U.S. P.Q. 676, 684-685 (T.T.A.B.1981), relied on by Kraft, the introduction of features other than the marks was just one of a long line of factors the Board noted when it concluded that the survey was of no use.

In interpreting the survey, the district court correctly discounted the “don’t knows” and found that only 7.6% of those interviewed confused the source of Yogowhip with the source of Miraole Whip. In Union Carbide, supra, 531 F.2d at 387-388, this Court counted those who did not know who put out the product shown, but who named other products put out by the same concern in response to a later question. The Court held that it was not necessary to name “Union Carbide” to indicate likelihood of confusion; associating the products with a single though anonymous source was sufficient. The survey in this case allowed not only the “don’t knows” but all respondents who did not answer Kraft or Miracle Whip to the first question an opportunity to indicate association with a single anonymous source by asking them to name other products put out by the same company that puts out Yogowhip. In fact, an additional .6% named Miracle Whip.2

The remaining question is whether the district court was correct in holding that a 7.6% confusion finding weighs against infringement. We hold that it was. Kraft has pointed to no case in which a 7.6% figure constituted likelihood of confusion. See James Burrough, supra, 540 F.2d at 279 n. 23 (referring to a prior case showing 11%); Union Carbide, supra, 531 F.2d at 386 (referring to a prior case showing 11.4%); McDonough, supra, 208 U.S.P.Q. at 685 (11%); Jockey Internat’l, Inc. v. Burkard, 185 U.S.P.Q. 201, 205 (S.D.Cal.1975) (11.4%). In Grotrian, Helfferich, Schulz, Th. Steinweg Nachf v. Steinway & Sons, 365 F.Supp. 707, 716 (S.D.N.Y.1973), affirmed, 523 F.2d 1331 (2d Cir.1975), the court held as strong evidence of likelihood of confusion a survey showing that 8.5% of the people interviewed confused the names Steinway and Grotrian-Steinweg, and 7.7% perceived a business connection between the two companies; it is not clear whether or not the figures were combined. In James Burrough Ltd. v. Sign of Beefeater, Inc. (Beefeater II), 572 F.2d 574, 578 (7th Cir.1978), this Court credited a survey showing that 10% of the respondents confronted with a Sign of the Beefeater restaurant sign mentioned that liquor came to mind; only 6% mentioned Beefeater specifically as a liquor that came to mind (Kraft Reply Br. B-14). Again, it is not clear which figures the Court deemed sufficient to evidence likelihood of confusion. In contrast, Henri’s points to Wuv’s Internat’l, Inc. v. Love’s Enterprises, Inc., 208 U.S.P.Q. 736, 756 (D.Colo.1980), where the court found 9% to be a “questionable amount of confusion” and in combination with other factors found no infringement; and S.S. Kresge Co. v. United Factory Outlet, Inc., 598 F.2d 694, 697 (1st Cir.1979), where the court found it unconvincing that 7.2% of the respondents believed that “The Mart” and “K-Mart” were owned by the same people, though the 7.2% figure was discredited because 5.7% of the respondents *359reached the same conclusion with respect to “The Mart” and “Kings Department Store.”

Despite the survey’s flaws, we conclude that the district court did not err in considering it, and the court correctly found that the 7.6% finding is a factor weighing against infringement.

C. Intent

A finding of fraudulent intent or bad faith is not essential to prove infringement where likelihood of confusion already exists. Tisch Hotels, Inc. v. Americana Inn, Inc., 350 F.2d 609, 613 (7th Cir.1965). One can readily infer likelihood of confusion, however, when there is proof of intentional copying because then “the adoption itself indicates that defendants expected that likelihood to their profit.” Id. See also Processed Plastic, supra, 675 F.2d at 857; American Chicle Co. v. Topps Chewing Gum, Inc., 208 F.2d 560, 563 (2d Cir.1953). Kraft argues that the evidence presented at trial supports a finding of intentional copying.

The district court found that while there was “an intent to compete” on the part of Henri’s, there was no intentional and deliberate copying of the Miracle Whip mark, no intent “to compete by misleading the consuming public” (App. 16). For various characterizations of the intentional copying standard, see McCarthy, Trademarks and Unfair Competition § 23:33 (1973) (“intent to get a free ride”); Tisch Hotels, supra, 350 F.2d at 613 (adoption of the senior user’s name “deliberately with a view to obtaining some advantage from [the senior user’s] investments in promotion and advertising”); Fleischmann Distilling Corp. v. Maier Brewing Co., 314 F.2d 149, 157 (9th Cir.1963) (adoption with the purpose of “capitalizing] upon the popularity of the name chosen”), certiorari denied, 374 U.S. 830, 83 S.Ct. 1870, 10 L.Ed.2d 1053. The evidence indicates that Les O’Rourke, Executive Vice President of Henri’s, selected the names Yogonaise and Yogowhip. At the time he chose the name Yogowhip, he was aware of the leading position of Miracle Whip on the market; he also chose the name long before production began with no specific idea how the product would be made. O’Rourke testified, however, that he chose the name because he thought it would describe the product: Yogo for yogurt, and “salad dressing is a whipped dressing” (Tr. 855). He further testified that he had always heard spoonable salad dressings referred to as whipped products, and that was why he chose the word “whip.” Despite Kraft’s characterization of this rationale as implausible (Br. 42), the district court evidently credited it. Kraft also points to Henri’s desire to put its Yogowhip product next to Kraft’s Miracle Whip product on the supermarket shelf as evidence of an intent to get a “free ride.” There was testimony, however, indicating that it is common practice in the food industry for a company to desire placement near the leading brand (Tr. 971). Finally, Henri’s distinguished Yogowhip by advertising it as an alternative to the high fat products in the marketplace: it contains yogurt, is low in fat, and tastes good (Tr. 857, 913-914). Henri’s also took special care to design a label that did not resemble Kraft’s (Tr. 856-857; 886-888).

On this record we cannot hold the district court clearly erroneous in its finding against Kraft on the issue of intentional copying. Though, as noted above, a finding of wrongful intent is not necessary to an ultimate finding of infringement, our evaluation of the record does mean that wrongful intent should not be counted as a factor in favor of infringement in this case.

Finally, Kraft asks this Court to consider that it has been the exclusive user of the term Whip in the midwest for almost fifty years. In its rejection of secondary meaning, the district court explained the fact of exclusivity by inferring that most users “got scared” when Kraft made moves to stop their use of Whip (App. 12). This record indicates that, for whatever reasons, the word Whip has not been in use, though derivatives of the word do appear. Even taking exclusivity of use into account, however, we think the scales dip against infringement. The marks are not confusingly *360similar; the survey evidence supports a finding of no likelihood of confusion and there is insufficient evidence to support a finding of intentional copying. The district court was therefore not clearly erroneous in finding no likelihood of confusion.

This holding does not necessarily conflict with the holdings of other courts that have considered similar challenges. In Kraft-Phenix Cheese Corp. v. Goldfarb, 7 F.Supp. 199 (S.D.Cal.1934), the court held Wonder Whip infringed Miracle Whip, and in Kraft-Phenix Cheese Corp. v. R.E. Robertson, Inc., 9 F.Supp. 125 (E.D.Mich.1934), the court held Wonder Mix infringed Miracle Whip; both marks bore greater similarity to Miracle Whip than the mark in this case because of the similarity in meaning of the words Wonder and Miracle. In Kraft Cheese Co. v. Leston Co., 43 F.Supp. 782, 783 (E.D.Mo.1941), the court held that Kraft was entitled to the exclusive use of both Miracle Whip and Whip, and that Leston’s Salad Whip infringed Kraft’s rights. The court in that case was either presented with more convincing evidence than the district court in this case; or what is more likely, based on the transcript of that court’s remarks at the close of the case (included in the record by Kraft), is that after finding that Whip had acquired secondary meaning, the court found likelihood of confusion as a matter of law, without requiring evidence to prove that the use of the word Whip in Leston’s Salad Whip would generate source confusion with Miracle Whip. In this case it is plain that Yogowhip does not generate source confusion with Miraole Whip regardless of whether Whip has secondary meaning. See Union Carbide, supra, 531 F.2d at 381; National Football League v. Wichita Falls Sportswear, 532 F.Supp. 651, 659 (W.D. Wash.1982). The Leston case is not the final word on the subject; in National Dairy Prods. Corp. v. Parman-Kendall Corp., 122 U.S.P.Q. 332 (T.T.A.B.1959), a trademark was successfully obtained for Avo-Whip salad dressing over Kraft’s opposition. According to an official of Parman-Kendall, the company “got scared” and signed a consent decree before the case was litigated (App. 12). On the facts of this case, and the law as applied in this Circuit, we affirm the district court’s finding of no infringement.

Mislabeling and Fraud

The district court refused to enjoin Henri’s from making certain representations on the labels of its yogurt products. Among other things, Kraft objects to Henri’s representation that Yogowhip is a salad dressing despite the fact that, according to Kraft, the Federal Standard of Identity for salad dressing does not permit yogurt as an ingredient, 21 C.F.R. §§ 130.8(a), 169.150. Because “the alleged improprieties have stopped,” the district court saw no need to either “evaluate the practices,” or “in the exercise of discretion,” enter an injunction (App. 23). Kraft has not convinced us that this refusal was an abuse of discretion.

In addition, Kraft argues that the district court abused its discretion in refusing to declare void and order cancellation of Henri’s July 7, 1976 trademark application for federal registration of Yogowhip as salad dressing. That application has been stayed pending resolution of this case. In August of 1981 Henri’s filed another application to register Yogowhip as a reduced calorie dressing, for which there is no Standard of Identity, and in which yogurt is therefore a permitted ingredient. Kraft has shown no irreparable injury from Henri’s July 7 filing of the trademark application that would necessitate an injunction; if the application is for some reason granted despite the presence of yogurt in the product, Kraft can object to it at that time.

For the foregoing reasons, the judgment of the district court is affirmed.

Notes

. It is also not necessary to review the district court’s finding that Whip is a descriptive word. Regardless of whether it is generic (as argued by Henri’s), descriptive (as found by the district court), or suggestive (as argued by Kraft), there is insufficient evidence of likelihood of confusion between Yogowhip and Miracle Whip or Whip to constitute infringement, see infra.

. In Union Carbide, supra, 531 F.2d at 387, the Court counted those respondents who first said that they did not know who put out the product shown, which was an Ever-Ready lamp or mini-bulb, but who named “batteries or flashlights” when asked to name other products put out by the same concern that put out the lamp or mini-bulb. Because both Union Carbide and the alleged infringer Ever-Ready Inc. put out flashlights, those respondents who answered “flashlights” alone, rather than “batteries and/or flashlights” should not have been counted as confused; such a response does not indicate that the interviewee has associated the producer of the lamp or mini-bulb with a “single anonymous source” such as Union Carbide, because both Union Carbide and Ever-Ready make flashlights. It is not clear from the Court’s opinion whether or to what extent the Court relied on the answer “flashlights” alone. Because both Henri’s and Kraft make salad dressing, those who named salad dressing as another product put out by the company that puts out Yogowhip cannot be counted as associating Yogowhip with a single, anonymous source, i.e., Kraft.



Dissenting Opinion

COFFEY, Circuit Judge,

dissenting in part.

I respectfully dissent from the majority’s opinion because I find it ironic that a company owning a strong trademark after spending millions of dollars promoting the same and having successfully defended it in the courts can now be deprived of this *361valuable asset by a competitor who has provided the court with a wholly inadequate basis to justify such a result. I believe it is likely that the consuming public will be confused into believing that Henri’s “Yogowhip” dressing “comes from the same source” (is made by the same company) as “Miracle Whip” salad dressing, an impermissible situation since the underlying purpose of trademark law is to prevent confusion. I reach this conclusion since I am firmly convinced that it is clear that the word “whip” has developed a secondary meaning and a weighing of the factors necessary in a decision whether or not one trademark infringes upon another trademark clearly indicates that “Yogowhip” infringes upon both “whip” and “Miracle Whip.”

I. Secondary Meaning

The district court found, without stating adequate legal reasoning, that the word “whip” was “merely descriptive.” “A merely descriptive term specifically describes a characteristic or ingredient of an article. It can, by acquiring a secondary meaning, i.e., becoming distinctive of the applicant’s goods (15 U.S.C. § 1052(f)), becomes a valid trademark.” Miller Brewing Co. v. G. Heileman Brewing Co., 561 F.2d 75, 79 (7th Cir.1977). See also Walt-West Enterprises, Inc. v. Gannett Co., Inc., 695 F.2d 1050, 1056 (7th Cir.1982). “Once a term ceases to be descriptive, for example, and commences to indicate the origin of a product, [as I am convinced it has in this fact situation], it has acquired a secondary meaning. Secondary meaning confers title and results in the attachment of the common law right to prevent infringing imitation of the mark in its trademark meaning.” E. Kintner and J. Lahr, An Intellectual Property Law Primer, at 257 (1982).

Several factors are relevant in the determination of whether or not a term has attained a secondary meaning: “The amount and manner of advertising, volume of sales, the length and manner of use, direct consumer testimony and consumer surveys.” Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d 366, 380 (7th Cir.1976). A brief history of “Miracle Whip” is required before engaging in the secondary meaning analysis. “Miracle Whip” salad dressing was originally introduced in 1933 and it became the market leader shortly thereafter and is still the best-selling salad dressing in the nation. The district court found that “Kraft has sold more than six billion units of ‘Miracle Whip’ Salad Dressing and has spent in excess of 70 million dollars in the promotion of the product.” Since the introduction of “Miracle Whip,” Kraft has vigorously protected its “Miracle Whip” trademark and the record is replete with examples of Kraft’s successful protective efforts including litigation when deemed necessary. See Kraft-Phenix Cheese Corp. v. Goldfarb, 7 F.Supp. 199 (S.D.Cal.1934) (“Wonder Whip” found to infringe “Miracle Whip”); Kraft Cheese Co. v. Leston Co., Inc., 43 F.Supp. 782 (E.D.Mo.1941) (“Leston’s Salad Whip” found to infringe “Miracle Whip”). But see National Dairy Products Corp. v. Parman-Kendall Corp., 122 U.S.P.Q. 332 (T.T.A.B.1959) (opposition to registration of “Avo-Whip” dismissed).1

Obviously the amount of advertising expenditures is not probative of secondary meaning unless those advertising efforts are successful in entrenching the word sought to be protected in the minds of the consuming public. When the consuming public equates the word “whip” with “Miracle Whip” salad dressing as opposed to being descriptive of the product, “whip” has acquired a secondary meaning. Therefore, consumers’ attitudes toward a trademark are the chief inquiry in a secondary meaning case of this type. Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 380. See also Carter-Wallace, Inc. v. Proctor & Gamble Co., 434 F.2d 794, 802 (9th Cir.1970). “[W]hen a company causes the public to *362associate a certain word with that company’s business, that word has a secondary meaning and receives the full protection of the law of trademark and unfair competition.” American Footwear Corporation v. General Footwear Company Limited, 609 F.2d 655, 663 (2d Cir.1979), cert. denied, 445 U.S. 951, 100 S.Ct. 1601, 63 L.Ed.2d 787 (1980). Our court has held that:

“A party attempting to appropriate the exclusive use of a descriptive term must persuade the court that the consuming public treats the term as a declaration that the item or service bearing the term originates at a single source, and thereby rebut the presumption that the public regards the term as merely describing an attribute of the item or service.”

Walt-West Enterprises, Inc. v. Gannett Company, Inc., 695 F.2d at 1058. For secondary meaning to attach to a certain word, the consuming public must give primary significance to that meaning which associates the word with a particular company’s business. See Kellogg Co. v. National Biscuit Co., 305 U.S. 111, 118, 59 S.Ct. 109, 113, 83 L.Ed. 73 (1938); American Footwear Corporation v. General Footwear Company Limited, 609 F.2d at 671.

The record discloses that the two surveys conducted by Henri’s indicate that the consuming public gives primary significance to that meaning of “whip” which, associates “whip” with Kraft’s spoonable salad dressing business — “Miracle Whip.” It is highly significant that in response to questions designed to ascertain the meaning of the word “whip,” consumers in both surveys responded “Miracle Whip” more often than any other response. In the first survey, the Bendikas Survey, 38% of the consumers surveyed responded “Miracle Whip” when asked the following question: “When thinking about mayonnaise-type spoonable salad dressings, what does the word ‘whip’ mean to you?” 36% of the consumers surveyed provided the next most frequent response, answering with various terms (lighter, creamy, smooth, fluffy, etc.) descriptive of the physical properties of the product.2 In the second survey conducted, the Hardin Survey, 30.5% of the consumers surveyed responded “Miracle Whip” when asked the following question: “When looking at this product (unidentified ‘Miracle Whip’ jar) what meaning does the term ‘whip’ convey to you?” 30.3% of the consumers surveyed provided the next most prevalent response, “mayonnaise.” 13.7% of the consumers surveyed responded “salad dressing.” Henri’s contends that the “mayonnaise” responses and the “salad dressing” responses should be combined to obtain a total for “spoona-ble salad dressing products” which would then exceed the “Miracle Whip” responses. This is a weak attempt to reduce the significance of the “Miracle Whip” responses— that the predominant meaning of “whip” to consumers is “Miracle Whip Salad Dressing.” There is no basis for combining these two categories because none of the survey questions were meant to elicit a response from consumers in the context of “spoona-ble salad dressing” and it is common knowledge that not all salad dressing is spoonable and not all mayonnaise is used as spoonable salad dressing.

The record establishes that the surveys introduced into evidence convincingly proved that the primary meaning consumers have given the word “whip” is Kraft “Miracle Whip.” This is not surprising considering the vast sums of money Kraft has spent promoting “Miracle Whip,” the fact Kraft has sold over six billion jars of “Miracle Whip” and that Kraft has virtually exclusively used the word “whip” in the context of the spoonable salad dressing market for approximately fifty years. Considering this proof it is more than evident that the district court’s finding that “whip” has no *363secondary meaning is clearly erroneous. It is obvious the district court failed to give proper consideration to this issue. This is not surprising considering the district court began its analysis of the issue by stating it was “convinced” the issue was a “red herring.” Nor did the court recognize that “[t]he crucial question in a case involving ‘secondary meaning’ always [has been] whether the public is moved in any degree to buy an article because of its source.” Walt-West Enterprises v. Gannett Co., Inc., 695 F.2d at 1062 n. 14 (quoting American Footwear Corporation v. General Footwear Company Limited, 609 F.2d at 663). Consequently, the district court ignored the foregoing test and failed to apply it when analyzing whether or not “whip” was infringed by Henri’s use of “whip” by naming its product “Yogowhip.”

II. Infringement

The district court correctly ruled that the test for trademark infringement is the likelihood of confusion among consumers and that their interest is the paramount consideration. James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d 266, 274 (7th Cir.1976). However, the district court has failed to protect the interest of the consuming public by failing to focus on the right of consumers not to be confused and, for some unknown reasons without foundation in the record, giving undue weight to Henri’s putative interests in the use of the word “whip.” The court also failed to require that Henri’s comply with its duty as a latecomer to avoid creating confusion in the market. Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 382.

The questions relevant to the infringement issue are: (1) whether or not “Yogow-hip” infringes “Miracle Whip,” and (2) (because I believe that the word “whip” has attained a secondary meaning) whether or not “Yogowhip” infringes Kraft’s use of the word “whip.” The following analysis is germane to both questions.

As the majority points out, several factors must be considered in the determination of whether there is a likelihood of consumer confusion: the type of trademark in issue, the similarity of design, the similarity of products, identity of retail outlets and purchasers, identity of advertising media utilized, defendant’s intent, and actual confusion. Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 381-82. The strength of the allegedly infringed mark is also relevant to this inquiry. Tisch Hotels, Inc. v. Americana Inn, Inc., 350 F.2d 609 (7th Cir.1965).

While the district court found that “Miracle Whip” was an “extremely strong trademark” it failed to give this factor the required weight mandated by caselaw and refused, without logical reasoning, to recognize that

“[a] mark that is strong because of its fame or its uniqueness, is more likely to be remembered and more likely to be associated in the public mind with a greater breadth of products or services, than is a mark that is weak . .. . ”

James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d at 276. See also James Burrough Ltd. v. Sign of the Beefeater, Inc., 572 F.2d 574, 577 (7th Cir.1978); AMF, Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir.1979) (strong trademark inherently distinctive and will be given widest ambit of protection). “The strength or distinctiveness of a mark determines both the ease with which it may be established as a valid trademark and the degree of protection it will be accorded.” Beer Nuts, Inc. v. Clover Club Foods Co., 711 F.2d 934 at 939 (10th Cir.1983) (quoting McGregor-Doniger, Inc. v. Drizzle, Inc., 599 F.2d 1126, 1131 (2d Cir.1979). The fame of “Miracle Whip” and “whip” creates the opportunity for others to ride on and benefit from Kraft’s goodwill “in the effort to attract consumers to goods and services which may be thought by consumers to be in some way associated with [Kraft]. That fame creates the ‘commercial magnetism’ referred to by the Supreme Court in Mishewaka Mfg. Co. v. Kresge Co., 316 U.S. 203 [62 S.Ct. 1022, 86 L.Ed. 1381] (1942).” James Burrough Ltd. v. Sign of *364the Beefeater, Inc., 540 F.2d at 277.3 The majority, while recognizing the strength of “Miracle Whip” as a trademark fails to consider this factor in its analysis. While conceding that three of the factors to be considered favor Kraft, namely, the similarity of products, the identity of retail outlets and purchasers, and the identity of advertising media utilized, the majority depreciates their significance in failing to discuss them. These factors are significant as they have a direct bearing on the ultimate question: whether or not there is a likelihood of consumer confusion. They failed to consider that “[converging marketing channels increase the likelihood of confusion.” Beer Nuts, Inc. v. Clover Club Foods Co., 711 F.2d 934 at 941 (quoting AMF, Inc. v. Sleekcraft Boats, 599 F.2d at 353). Also, that “[c]onfusing similarity is most likely when the products themselves are very similar.” Id. Such is the case here since we have two similar products sold side-by-side in grocery stores. Furthermore, without citation, the majority considers three other factors: similarity of the trademarks, evidence of confusion and the intent of the alleged infringer to be the “more important” factors. I disagree not only with the majority’s deference to merely three out of the eight relevant factors to be considered in an infringement case, but also with their conclusion that these factors do not weigh in favor of infringement, a conclusion they somehow reach without setting forth a legal, logical basis.

In their analysis of the similarity factor, despite recognizing that a supermarket shopper makes a selection quickly and without careful examination, the majority proceeds to perform its own “examination” of the product labels giving substantial weight to factors such as the number of syllables in “Miracle Whip” and “Yogowhip” and the color and styling differences in the products’ labels. However, caselaw recites that “[a] side-by-side comparison of the marks is not the proper test. The test is the consumer’s state-of-mind when faced with the marks individually.” Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 382 (emphasis added). “Though the marks must be compared, they must be compared in light of what occurs in the marketplace, not in the courtroom.” James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d at 275. The majority’s “examination” is not the kind that would be undertaken by the typical consumer in the marketplace as “[t]he consuming public is unlikely ever to be presented with the opportunity for such comparison.” Id. “Buyers typically exercise little care in the selection of inexpensive items that may be purchased on impulse. Despite a lower degree of similarity, these items are more likely to be confused than expensive items which are chosen carefully.” Beer Nuts, Inc. v. Clover Club Foods Co., at 941 (citing Sun-Fun Products, Inc. v. Suntan Research Development, Inc., 656 F.2d 186, 191 (5th Cir.1981)). Spoonable salad dressing is an inexpensive item which will be purchased on “impulse.” The item also has a high rate of purchase frequency and consequently a scan of the available products and quick selection will be made rather than the lengthy purchase decision associated with expensive items. The fact Henri’s merchandising experts intentionally positioned “Yogowhip” next to “Miracle Whip” on retail grocery shelves further induced impulse purchase decisions to purchase “Yogowhip” instead of “Miracle Whip.” Also, the majority’s “examination” is clearly contradicted by the obvious confusion shown by a significant percentage of consumers in the survey admitted into evidence.

While no evidence of actual consumer “confusion” was presented to the court, “confusion” evidence in the form of a consumer survey was admitted. Such evidence is probative of the likelihood of confusion among consumers. The evidence of confu*365sion contained in the Hardin Survey showed that 7.6% of the consumers surveyed believed that “Yogowhip” was produced by Kraft or whatever company produced “Miracle Whip” salad dressing. Despite this, the district court erroneously ruled that such a level of confusion is not sufficient to find infringement.

Our court in James Burrough Ltd. v. Sign of the Beefeater, Inc. recognized that “the percentage of likely confusion required may vary from case-to-case . . .. ” 540 F.2d at 279. There the court found 15% confusion not to be de minimis and set no specific required level of confusion. Id. In that case, the products of the respective parties were not related (gin and roast beef dinners) and indeed direct competition between products is not necessary to a finding of infringement. However, “the more closely products are related, the more likely sources may be confused.” Union Carbide Corp. v. Ever-Ready, Inc., 531 F.2d at 382. Such is the case here where both products are spoonable salad dressings and it is entirely conceivable consumers would think “Yo-gowhip” was a reduced calorie salad dressing or a yogurt based salad dressing being introduced by the producers of “Miracle Whip.” In a case such as we are presented with, where the evidence shows that only roughly half as many consumers are confused than in James Burrough, Ltd., this lower confusion level is nonetheless extremely significant and should be accorded more weight because the products are in competition with each other and it is much more likely that this lower level of confusion will have a greater adverse impact upon the party seeking to protect its trademark (Kraft). While in James Burrough Ltd., supra, the only likely harm which would occur would be for some imbibers of “Beefeater” gin to switch to another brand if served an unsatisfactory meal at the “Sign of the Beefeater” restaurant, here Kraft stands to lose a significant portion of its “spoonable salad dressing” market share to Henri’s if some of the 7.6% “confused” consumers purchase “Yogowhip” because of a perceived connection with Kraft products. Without doubt, brand loyalty is an extremely important factor in consumer purchases. If it were not, Kraft would not have invested over 70 million dollars in advertising.

The majority adopts and gives great weight to Henri’s argument that Kraft has not presented eases to the court in which a 7.6% level of confusion constituted a likelihood of confusion. However, each case must be decided on its own facts and “the percentage of likely confusion required may vary from case to case, ...” James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d at 279. “As we said in G. LeBlanc Corporation v. H & A Selmer, Inc., [310 F.2d 449 (7th Cir.1962) ] we are not aided by citation of other registrations, the evidence frequently being so varied on the question of likelihood of confusion.” James Burrough Ltd. v. Sign of the Beefeater, Inc., 540 F.2d at 279 n. 24. Our court has recognized “the right of the public to be free of confusion,” Id. at 274 (emphasis added), and that “[a] trademark owner has a property right only insofar as is necessary to prevent consumer confusion as to who produced the goods and to facilitate differentiation of the trademark owner’s goods.” Walt-West Enterprises, Inc. v. Gannett Company, Inc., 695 F.2d at 1057 (emphasis added) (quoting International Order of Job’s Daughters v. Lindeburg & Co., 633 F.2d 912, 919 (9th Cir.1980), cert. denied, 452 U.S. 941, 101 S.Ct. 3086, 69 L.Ed.2d 956 (1981)). It is clear that these cases are concerned with the existence of some “confusion” rather than a specific amount of confusion. Indeed the underlying purpose of trademark law is to prevent confusion. The survey evidence unquestionably proves that some “confusion” exists in the minds of consumers. This “confusion” is sufficient to establish the likelihood the public will be moved in some degree to purchase “Yogowhip” due to the misperception it is manufactured by Kraft. See Walt-West Enterprises v. Gannett Co., Inc., 695 F.2d at 1062 n. 14.

“A newcomer to a product field has an infinity of nonconflicting trademarks from which to choose, and he has a legal duty to select a mark which is totally dissimilar to trademarks already being used in the field.” *366Jockey International, Inc. v. Burkard, 185 U.S.P.Q. 201, 208 (S.D.Cal.1975). Accord, Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 392. “One who adopts a mark similar to another already established in the marketplace does so at his peril, . .. All doubts must be resolved against him.” Beer Nuts, Inc. v. Clover Club Foods Co., at 941 (citations omitted). Henri’s could have selected a dissimilar trademark with a minimum of effort. I would hold that there is clearly a sufficient amount of evidence to find that the amount of consumer confusion as to the origin of “Yogowhip” should cause this factor to weigh heavily against Henri’s on the issue of infringement.

Finally, the district court somehow, without reason and legal basis recited in the record, found persuasive Henri’s position that it did not intend to infringe upon Kraft’s trademark. The court instead found that Henri’s intended to compete with Kraft but did not intend to compete by misleading the consuming public to believe that “Yogowhip” was distributed by Kraft. The court reached this result because it was convinced Henri’s was “equally concerned with maintaining the distinctiveness of its products.” However, the evidence introduced at trial does not support this conclusion and clearly demonstrates that Henri’s paid no heed whatsoever to its legal duty as a latecomer in the spoonable salad dressing market to avoid all consumer confusion. See Union Carbide Corporation v. Ever-Ready, Inc., 531 F.2d at 382.

At trial Leslie O’Rourke, Henri’s executive vice president, testified that when he selected the name “Yogowhip” he was not aware of the “whipping” process used in making spoonable salad dressings. This factor not only negates the district court’s finding that the word “whip” was used by Henri’s in order to describe the process of making “Yogowhip,” but further indicates Henri’s intent to infringe on the “Miracle Whip” trademark as O’Rourke was aware of the leading position “Miracle Whip” held in the spoonable salad dressing market. Furthermore, O’Rourke had previously been informed of the “Miracle Whip” trademark and had been warned that the use of the term “Yogowhip” might in fact infringe upon that mark. O’Rourke also testified that he directed his brokers and salesmen to have “Yogowhip” placed on supermarket shelves immediately next to “Miracle Whip” products for the purposes of inducing “impulse” purchases of “Yogowhip” by individuals who might otherwise purchase “Miracle Whip.” This is a further attempt by Henri’s to seek the benefit of “Miracle Whip’s” reputation for quality. One of Henri’s foodbrokers also testified that it was his opinion that the “whip” in “Yogow-hip” was “undoubtedly conceived to give the impression that it is a similar product to the leader in the market, Miracle Whip.” I also believe that it is more than a mere coincidence that the most distinctive part of the “Yogowhip” label, the word “Yogow-hip,” appears in a lettering style virtually identical to that used by Kraft in the most distinctive portion of the “Miracle Whip” label, the words “Miracle Whip.”

While it is true that a party need not show intent to establish a “likelihood of confusion,” wrongful intent is viewed as presumptive evidence of the likelihood of confusion. See Processed Plastic Co. v. Warner Communications, Inc., 675 F.2d 852, 857 (7th Cir.1982); Beer Nuts, Inc. v. Clover Club Foods Co., at 941. “Important in determining whether the second comer’s entrance into the market creates possible confusion, is any evidence of conscious imitation of the first comer’s product.” Harold F. Ritchie, Inc. v. Chesebrough-Pond’s, Inc., 281 F.2d 755, 758 (2d Cir.1960) (emphasis added). Since the evidence clearly establishes that a key member of Henri’s management team knew of the possibility of an infringement problem and yet encouraged the placement of the product in such a way as to maximize the potential “confusion” in the mind of the consuming public, the district court’s decision that there was no intent to deceive is clearly erroneous.

I would reverse the district court and find that “whip” has developed a secondary meaning when used in the context of a spoonable salad dressing, and that Henri’s use of the word “Yogowhip” infringed upon *367Kraft’s use of the words “Miracle Whip” and “whip.” I believe that the evidence clearly shows that the relevant factors, when balanced, easily weigh in Kraft’s favor and show that it is likely that a number of consumers will be confused as to the source of “Yogowhip.” As a result Henri’s will increase its profits at Kraft’s expense and some consumers whose expectations as to “Yogowhip” are not fulfilled will blame Kraft (by not purchasing Kraft products) and not Henri’s for his or her dissatisfaction. Since the interests of the consuming public are paramount, and on the state of this record I am convinced there is a significant amount of consumer confusion, I would reverse since we have a duty to protect the public.

. This case is distinguishable because there was nothing in the record suggesting that “the public uses ‘Whip’ or would recognize the word as meaning ‘Miracle Whip.’ ” 122 U.S.P.Q. at 332. The survey evidence discussed infra shows that a significant percentage of the public recognizes “whip” as meaning “Miracle Whip.”

. The district court found this survey biased and did not accept it into evidence for the purpose of determining the likelihood of confusion on the issue of infringement. The preceding question asked of the consumers and two other questions purportedly biased the survey in favor of Kraft by suggesting certain answers. However, there is no logical or legally sound reasoning set forth in the court’s ruling why the results of the survey should not be probative as to the secondary meaning issue. In any event the second survey conducted provides a sufficient basis for a finding of secondary meaning.

. Our court’s opinion in James Burrough Ltd. v. Sign of the Beefeater, Inc., should be given substantial deference since we were privileged to have Chief Judge Markey of the U.S. Court of Customs and Patent Appeals (now the U.S. Court of Appeals for the Federal Circuit) sit on the panel and author the opinion, giving our court the considerable benefit of his expertise in trademark law.