Combe Inc. v. Dr. Aug. Wolff GMBH & Co.

Full title: COMBE INC., Plaintiff, v. DR. AUGUST WOLFF GMBH & CO. KG ARZNEIMITTEL…

Court: United States District Court, E.D. Virginia, Alexandria Division.

Date published: May 23, 2019

382 F. Supp. 3d 429 (E.D. Va. 2019)

Opinion

Civil Action No. 1:17-cv-00935

2019-05-23

COMBE INC., Plaintiff, v. DR. AUGUST WOLFF GMBH & CO. KG ARZNEIMITTEL, Defendant.

Anna Balishina Naydonov, Elizabeth Diane Ferrill, Finnegan Henderson Farabow Garrett & Dunner LLP, Washington, DC, for Plaintiff. Michael Abbott Grow, Arent Fox PLLC, Washington, DC, for Defendant.

T.S. Ellis, III, United States District Judge

Anna Balishina Naydonov, Elizabeth Diane Ferrill, Finnegan Henderson Farabow Garrett & Dunner LLP, Washington, DC, for Plaintiff.

Michael Abbott Grow, Arent Fox PLLC, Washington, DC, for Defendant.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

T.S. Ellis, III, United States District Judge

Plaintiff, a manufacturer and seller of personal care products for women and men and the owner of the registered trademark VAGISIL, brings this action pursuant to 15 U.S.C. § 1071(b) seeking reversal of the Trademark Trial and Appeal Board decision dismissing plaintiff's opposition to defendant's application to register the VAGISAN mark.

After completion of discovery, pretrial proceedings, and full briefing, the matter came before the Court on December 4, 2018 for a bench trial. There, the parties stipulated various undisputed facts and presented substantial testimonial and documentary evidence. Each party presented two witnesses. Testifying for plaintiff were a marketing and consumer survey expert, Mr. Hal Poret, and plaintiff's senior vice president and general manager of over-the-counter brands for North America, Ms. Stacey Feldman. In response, defendant presented its own marketing and consumer survey expert, Dr. Itamar Simonson, and defendant's head of international key account management, Ms. Angela May Thevessen. After the close of evidence and presentation of counsel's oral arguments on December 7, 2018, the matter was taken under advisement. Set forth here, pursuant to Rule 52, Fed. R. Civ. P., are the Court's findings of fact and conclusions of law.

I. FINDINGS OF FACT

1. Plaintiff, Combe Inc., is a Delaware corporation that manufactures and sells personal care products for women and men.

2. Plaintiff has sold a variety of feminine personal care products for the vaginal area under the VAGISIL mark in the United States since 1973. Products sold by plaintiff under the VAGISIL mark have included medicated anti-itch creams and wipes, intimate washes, emergency contraceptives, moisturizers, deodorant powders, and screening kits.

3. On October 17, 1978, plaintiff obtained its first trademark registration for the VAGISIL mark. That registration bears United States Registration No. 1104172 and covers "pharmaceutical preparations, namely, medicated creams" in Class 5. Plaintiff also subsequently obtained registrations for the VAGISIL mark that cover powders, wipes, washes, moisturizers, and screening kits. 4. The United States Patent and Trademark Office ("PTO") did not require evidence of acquired distinctiveness or "secondary meaning" for any of plaintiff's trademark applications for the VAGISIL mark.

5. Since 1991, products sold using the VAGISIL mark in the United States have totaled over one billion dollars in sales revenue. In addition, since 1991 the VAGISIL brand has had between 55% and 60% market share in the feminine anti-itch cream segment and between 50% and 69% market share in the feminine powder segment. Since 2010, the VAGISIL brand has had at least 94% market share in the medicated wipes segment.

6. Marketing data from retailers that have sold VAGISIL-marked products during the past year reflect that in the broad intimate-health product category, VAGISIL-marked anti-itch cream ranks fifth in sales revenue and VAGISIL-marked wash ranks sixth. In the anti-itch cream segment of the intimate-health category, VAGISIL-marked anti-itch cream is ranked first in both units sold and sales revenue. In the anti-itch wipe segment, VAGISIL-marked wipes occupy the top two ranks in both units sold and sales revenue. In the intimate wash segment, VAGISIL-marked washes are ranked first and second in sales revenue and first and third in units sold. In the intimate powder segment, VAGISIL-marked powder ranks first in both units sold and sales revenue. In the moisturizing lubricant segment, VAGISIL-marked moisturizing lubricant ranks third in both units sold and sales revenue.

7. VAGISIL-marked products are sold nationwide in a wide assortment of dollar stores, drug stores, grocery stores, mass merchandising stores, pharmacies, and e-tailers.

8. On average, the prices of VAGISIL-marked products range between $ 3.00 and $ 7.00.

9. Consumers of VAGISIL-marked products are women who fall into a wide variety of age groups and socioeconomic backgrounds.

10. Consumers exercise different levels of care in selecting and purchasing feminine personal care products for the vaginal area. Some consumers conduct research before buying such products, and certain companies, including plaintiff, use doctors to educate consumers about different vaginal conditions and products that treat those conditions. Other consumers purchase these products hastily because consumers are embarrassed by the products and the highly personal conditions they treat, and thus the consumers do not want to spend much time at the shelf.

11. Since 1993, plaintiff has spent over $ 350 million advertising and promoting VAGISIL-marked products to the general consuming public through numerous forms of media, including print, television, radio, and the Internet. This level of spending is similar to that of other major sellers of intimate-health products, including KY and Plan B, and exceeds the advertising spending by some of VAGISIL's key competitors, such as SUMMER'S EVE and REPLENS.

12. In addition, since 2008, plaintiff has spent over $ 75 million on other forms of marketing to promote VAGISIL-marked products, including events, public relations campaigns, coupons, samples, and displays in stores.

13. Since the 1980's, plaintiff has advertised and promoted VAGISIL-marked products with over ninety different commercials that have aired on prominent television stations and cable networks. 14. Since the 1970's, plaintiff has advertised and promoted VAGISIL-marked products on over one hundred radio stations, including many prominent networks and satellite radio stations.

15. Since the late 1970's, plaintiff has advertised and promoted VAGISIL-marked products in a wide variety of printed publications, including consumer print books, women's health and beauty books, trade publications, and prominent, nationally circulated magazines and newspapers.

16. Plaintiff also digitally advertises and promotes VAGISIL-marked products on health and beauty websites, including Wed MD and Vogue.com, and social media websites, including Facebook, Twitter, and Instagram. Between June 2016 and May 2018, plaintiff's advertisements for VAGISIL-marked products on Facebook generated approximately 124 million impressions, i.e. the advertisements appeared on Facebook users' computer screens approximately 124 million different times. Since April 2015, plaintiff's website Vagisil.com, which advertises and promotes VAGISIL-marked products, has been viewed over 7.2 million times.

17. The VAGISIL brand has been mentioned in several popular, nationally televised television programs, including South Park, Saturday Night Live , and The Big Bang Theory. The VAGISIL brand was the subject of an entire skit on Saturday Night Live.

18. The VAGISIL brand has also been the subject of unsolicited media attention in articles in leading national publications, including The New York Times , The Wall Street Journal , and The Washington Post.

19. Defendant, Dr. August Wolff GmbH & Co. KG Arzneimittel, is a German limited liability partnership that manufactures and sells medicinal, cosmetic, and dermatological products.

20. In 1998, defendant began to sell in Germany a variety of feminine personal care products for the vaginal area under the VAGISAN brand. Defendant later expanded its sale of VAGISAN-branded products into other countries in Europe, the Middle East, and Asia. Defendant owns trademark registrations for the VAGISAN mark in Germany and more than a dozen other countries. Products sold by defendant under the VAGISAN brand include, among others, suppositories to restore vaginal pH and to regenerate vaginal flora, moisturizing cream, intimate wash lotion, protective ointment, and nutritional supplements. In Europe, parts of the Middle East, and Asia, defendant sells VAGISAN-branded products in pharmacies, health and beauty stores, drugstores, and online shops.

21. Defendant is not currently selling, and has never in the past sold, VAGISAN-branded products in the United States.

22. On January 24, 2012, defendant filed Application Serial No. 79111922 with the PTO to register the mark VAGISAN for "pharmaceutical preparations, namely, vaginal moisturizers, vaginal anti-fungal preparations, vaginal washes; sanitary preparations for medical use; diet pills, diet capsules, diet liquid medications" in Class 5 and "soaps, perfumery, essential oils, cosmetics, hair lotions" in Class 3.

23. Defendant's registration application displays the VAGISAN mark in standard characters and does not include any stylization, colors, company/house name, packaging, or any other elements.

24. On March 7, 2013, plaintiff filed Opposition No. 91209708 before the Trademark Trial and Appeal Board ("the TTAB") against defendant's application for the VAGISAN mark.

25. On June 19, 2017, the TTAB issued a final decision dismissing plaintiff's opposition against defendant's application for the VAGISAN mark on the ground that plaintiff had failed to prove (i) that defendant's use of the VAGISAN mark in United States commerce would create a likelihood of confusion with plaintiff's VAGISIL mark or (ii) that defendant lacked a bona fide intent to use the VAGISAN mark in United States commerce.

26. On August 21, 2017, plaintiff filed this action appealing the TTAB's decision to this Court pursuant to 15 U.S.C. § 1071(b)(1) and asserting additional counts of trademark infringement, unfair competition, and dilution under the Lanham Act; trademark infringement under the Virginia Trademark and Service Mark Act § 59.1-92.12, et seq. ; and trademark infringement and unfair competition under Virginia common law.

27. On January 2, 2018, defendant moved to dismiss all counts of plaintiff's Complaint, except for the count appealing the TTAB's decision, on the ground that defendant currently does not use the VAGISAN mark in United States commerce and has no imminent plans to do so. By Memorandum Opinion and Order issued March 28, 2018, defendant's motion to dismiss was granted, and all counts of the Complaint were dismissed except for plaintiff's appeal of the TTAB decision. See Combe Inc. v. Dr. August Wolff GmbH & Co. KG Arzneimittel , 309 F. Supp. 3d 414 (E.D. Va. 2018).

28. After all Counts of plaintiff's Complaint other than the TTAB appeal had been dismissed, plaintiff retained Mr. Hal Poret to conduct two consumer surveys to test: (i) whether the VAGISIL mark is famous ("the Fame Survey") and (ii) whether there is a likelihood of confusion between the VAGISIL and VAGISAN marks ("the Confusion Survey").

29. Mr. Poret is a professional consumer survey researcher and consultant who has designed, conducted, and analyzed over one thousand consumer surveys, hundreds of which involved trademarks. Mr. Poret's surveys and expert opinions have been admitted as evidence by numerous federal courts, including the Eastern District of Virginia.¹

1 See, e.g. , Valador, Inc. v. HTC Corp. , 241 F. Supp. 3d 650, 669 n.22 (E.D. Va.), aff'd , 707 F. App'x 138 (4th Cir. 2017) (noting that "defendants' expert, Hal Poret, conducted a survey in accordance with well-settled and accepted methods in trademark infringement cases, finding that there was 0% confusion among potential consumers").

30. The Fame Survey questioned three hundred United States consumers, both women and men, age eighteen and older, and tested both aided and unaided awareness of the VAGISIL mark in connection with vaginal care products. Unaided awareness is a measure of the number of consumers who express knowledge of a brand name without being shown the brand name. Aided awareness is a measure of the number of consumers who express recognition of a brand name after being shown the brand name.

31. In the unaided portion of the survey, respondents were asked to list all brands of vaginal care products that they had ever seen or heard of. In total, 38.7% of all respondents named VAGISIL in this unaided portion of the survey. VAGISIL was named more frequently than any other brand, with SUMMER'S EVE being named the second greatest number of times by 26% of all respondents.² 32. Next, in the aided portion, respondents were shown a total of eight brand names, one at a time and in random order, and were asked whether or not they had ever seen or heard of the name in connection with any vaginal care product. The brand names shown to respondents in this portion included VAGISIL; other well-known brands MONISTAT and SUMMER'S EVE; lesser-known brands REPLENS, LUVENA, SWEET SPOT, and VAGI-GARD; and a fictitious brand VAGIZOX, which served as the control. In total, VAGISIL was recognized by more respondents than any other brand name shown in the survey, with 90% of all respondents recognizing VAGISIL. Only 5.3% of respondents stated they had seen or heard of the control, VAGIZOX, in connection with any vaginal care product. Accordingly, the Fame Survey reported a net awareness rate of 85% for VAGISIL; that is, it reflected that 85% of respondents recognized VAGISIL specifically because of the mark's fame and not because of other factors.

2 It is also worth noting that in the unaided portion of the Fame Survey VAGISIL was named by 52.3% of women and 24.1% of men who were surveyed. Remarkably, the 24.1% male unaided awareness level of VAGISIL is comparable to the 26% total unaided awareness level of the second most named brand, SUMMER'S EVE.

33. The Confusion Survey questioned four hundred female consumers of vaginal moisturizers, vaginal washes, or vaginal anti-fungal products.

34. The Confusion Survey utilized the well-established Eveready ³ format for likelihood of confusion surveys. Specifically, the survey first presented the VAGISAN mark to the two hundred respondents who were in the test group and then asked those respondents a series of standard Eveready questions to measure various forms of confusion between VAGISAN and other brand names.⁴ The Confusion Survey never mentioned or showed the VAGISIL mark to respondents.

3 See Union Carbide Corp. v. Ever-Ready, Inc. , 531 F.2d 366 (7th Cir. 1976), cert. denied , 429 U.S. 830, 97 S.Ct. 91, 50 L.Ed.2d 94 (1976) (approving what is now known as the " Eveready " test).

4 For example, respondents were asked (i) what company or brand they think puts out the products they were just shown, (ii) whether they think the company that makes the products they were just shown makes any other products that they know of (and what the names of those other products are), and (iii) whether they think the products they were just shown are affiliated with, sponsored by, or approved by any other company or brand they know of (and what the names of those other companies or brands are).

35. The Confusion Survey showed the test group respondents the VAGISAN mark as the mark appears on defendant's application to register the mark: in a plain, block letter format, without any trade dress, stylization, special font, colors, company name, packaging, or any other distinctive elements and with the description "vaginal moisturizers, vaginal washes and vaginal anti-fungal products."

36. The Confusion Survey was also administered to two hundred respondents in a control group, who were shown the fictitious name VAGIPUR instead of VAGISAN. VAGIPUR was shown to the control group respondents in a plain, block letter format, without any trade dress, stylization, special font, colors, company name, packaging, or any other distinctive elements and with the description "vaginal moisturizers, vaginal washes and vaginal anti-fungal products."

37. When presented with the VAGISAN mark, 37% of the respondents in the test group named VAGISIL in response to the confusion questions. In the control group, 18% of the respondents named VAGISIL after they were shown the VAGIPUR mark and were asked the confusion questions. Accordingly, the Confusion Survey reported a net confusion rate of 19% between VAGISAN and VAGISIL; that is, it reflected that 19% of respondents confused VAGISAN with VAGISIL specifically because of the marks' similarity and not because of other factors.

38. Defendant did not present any survey evidence to rebut plaintiff's survey measurements of the VAGISIL mark's fame or the rate of confusion between the VAGISAN and VAGISIL marks. Instead, defendant introduced Dr. Itamar Simonson's criticisms of the Fame and Confusion Surveys.

39. Dr. Simonson is a professor of marketing and consumer decision-making and has an expertise in the field of researching consumer behavior, including consumer survey methodology. Dr. Simonson has also conducted many consumer surveys.

40. Dr. Simonson submitted an expert report and gave testimony at trial in which Dr. Simonson opined that the two surveys submitted by plaintiff were unreliable because the surveys were flawed in certain key respects. In particular, Dr. Simonson opined that the Fame Survey used an improper control and that the Confusion Survey failed to replicate market conditions and used an improper control.

41. Defendant submitted evidence reflecting that certain marks that are similar to VAGISIL are used by third parties in connection with goods sold in United States commerce.⁵ The following table includes the names of these third-party marks, as well as the evidence of sales revenues and advertising expenditures that correspond to each of the marks.

5 It appears that that defendant discovered many of these third-party "vagi-" brands by scouring the Internet and physical retail stores during this litigation. Defendant's counsel's paralegal admitted that he "possibly" spent more than ten hours over a period of two months searching on the Internet and "six to ten hours" searching in brick-and-mortar stores in the Washington, DC area for third-party products that include "vagi-" in the prefix of their names.

Mark Sales Revenues Advertising Nature of Products Expenditures 1. VAGITONE $3,278 in 2014; None. Anti-itch and antiodor $3,237 in 2015; formula. $1,300 in 2016; $943 in 2017; and $346 in 2018. 2. VAGIFRESH $690 in 2015; None. Moisturizer, wash. $958 in 2016; and $500 in 2018. 3. VAGI-CURE $1,440 in 2013; None. Anti-itch cream. $15,629 in 2014; $14,760 in 2015; $0 in 2016; and $3,788 in 2017. Product has been discontinued. 4. VAGI-CLEAR $3,517 in 2015; $988 in 2015; Anti-itch formula. $3,796 in 2016; $850 in 2016; $3,666 in 2017; and $869 in 2017; $1,392 in 2018. and $342 in 2018. 5. VAGIKOOL $24.95 in 2015; $1,800 in 2016; Reusable cold packs. $29,006 in 2016; $10,467 in 2017; $54,475 in 2017; and $12,194 in 2018. $39,724 in 2018. 6. VAGIFIRM None beyond one None. Tightening product purchased by supplements. defendant's counsel. 7. VAGI-SOOTHE $2,109 in 2015; $988 in 2015; Anti-itch formula. $3,505 in 2016; $850 in 2016; $5,309 in 2017; and $869 in 2017; and $1,358 in 2018. $342 in 2018. 8. VAGI-PAL None beyond one None. Anti-irritation deep product purchased by treatment bath. defendant's counsel. 9. VAGICAL $271 in 2016; None. Anti-irritation insert. $367 in 2017; and $175 in 2018. 10. VAGICARE None beyond one None. Probiotic pill to product purchased by prevent irritation. defendant's counsel. 11. VAGI-SITZ None beyond one None. Bath kits for draining product purchased by cysts. defendant's counsel.

There is no evidence that products sold under the brands VAGI-CLEAR, VAGISOOTHE, VAGIKOOL, VAGITONE, VAGI-FIRM, VAGICAL, or VAGIPAL are sold in any brick and mortar stores. Further, defendant's counsel purchased the VAGICARE product from a Canadian website using Canadian dollars and could not identify any United States websites or stores that carry the product.

42. In addition, during the underlying proceeding before the TTAB, defendant's counsel purchased one product sold under each of the following marks: VAGICREAM, VAGI-CARE, and VAGI-SILK. There is no evidence that these products continue to be sold in the United States, that these products have ever been sold to anyone in the United States other than defendant's counsel, or that these products have been advertised to consumers in the United States.

43. Plaintiff purchased the marks VAGISTAT, VAGISTAT-1, VAGISTAT-3 in May 2017 and has begun to sell VAGISTAT products in packaging labeled "VAGISTAT by VAGISIL Brand." Those marks were previously used by a different company in connection with the sale of a product to treat yeast infections.

44. Several prominent mass merchandising and drug stores, with nationwide locations, sell a generic vaginal anti-itch cream that is labeled "Vagicaine" and is also marked by the store's private-label brand (e.g. , EQUATE, UP & UP, RITE AID, and CVS). These stores have sold millions of dollars in "Vagicaine" products since 2009. Specifically, Rite-Aid has sold $ 2,545,090 of RITE-AID-branded "Vagicaine" products since 2009; CVS has sold $ 1,380,979 of CVS-branded "Vagicaine" products since 2016; Target has sold [sales figure submitted under seal] of UP & UP-branded "Vagicaine" products since 2013; and Walmart has sold $ 8,622,304 of EQUATE-branded products since 2014. There is no evidence that any of these stores advertise their private-label "Vagicaine" products to consumers.

45. The following table includes additional marks that third parties have registered and applied to register with the PTO. There is no evidence that any products have ever been sold to anyone in the United States in connection with these marks or that these marks have been advertised to consumers in the United States.

Mark Registration/Application Nature of Goods or Services 1. VAGILIFT Live registration. Performing cosmetic surgical procedures for vaginal rejuvenation. 2. VAGI-WAVE Live registration. Tampons, feminine hygiene pads, vaginal exercisers and dilators, and physical therapy advice. 3. VAGI-HEX Live registration. Pharmaceutical preparations in tablet form for the treatment of vaginal conditions. 4. VAGI-GARD Live registration. Topical treatment for yeast infections, medicated douches, medicated powder, medicated creams, lubricants, and suppositories. 5. VAGIFEM Live registration. Pharmaceutical, gynecological preparations and applicators for the treatment of menopausal disorders. 6. VAGITOCIN Live registration. Pharmaceutical preparations in the form of gel for local application in the vagina to treat fragile, bleeding, or atrophic mucous membranes. 7. VAGITAMIN Application pending. Vitamin supplements, vitamin tablets, vitamins, gummy vitamins, and multi-vitamin preparations. 8. VAGISERT Application pending. Vaginal antifungal preparations. 9. VAGICLICK Live registration. Medical apparatus for application of flowable pharmaceutical preparations for the skin. 10. VAGICIAN Live registration. Body waxing services for the human body. 11. VAGICONS Live registration. Belts, bottoms, hats, hoods, jackets, ties, and tops.

46. Defendant also submitted PTO records reflecting that the following applications and registrations have been cancelled, are expired, or are otherwise inactive or "dead": VAGICAP, VAGICLEANZ, VAGISOOTHIES, VAGI-NEX, VAGINEX, VAGIPILLS, VAGIFORM, VAGICULT, VAGIDINE, VAGICORT, VAGINOLA, VAGILIA, VAGITENE, VAGINOSTIC, VAGITROL, VAGISPROIN, VAGIMINE, VAGISEC, VAGISEC PLUS, VAGITAB, VAGISOL, VAGILAC, VAGINOME, VAGISTEAM, VAGITRACE, VAGIPEARL, VAGISSAGE, VAGINUE, VAGIWEAR, VAGISOL, VAGINORM, VAGIANT, VAGINOL, VAGITARIAN, VAGI-C, VAGIGRAPH, VAGIMETER, VAGICON, and VAGICANE.

II. CONCLUSIONS OF LAW

Pursuant to 15 U.S.C. § 1071(b)(1), a party may initiate a civil action in the district court to contest the TTAB's decision that a mark is or is not registrable. In a § 1071(b) action, the district court acts as the finder of fact and has authority independent of the PTO to grant or cancel registrations. Swatch AG v. Beehive Wholesale, LLC , 739 F.3d 150, 155 (4th Cir. 2014) (citing 15 U.S.C. § 1071(b)(1) ). The parties to the action have the right to admit the PTO record as well as any new evidence not presented to the PTO that is admissible under the Federal Rules of Evidence and Civil Procedure. Id. (citing 15 U.S.C. § 1071(b)(3) ). Importantly, when new evidence is submitted in addition to the PTO record, the district court must exercise de novo review of the entire factual record. Id. at 156 (citing Kappos v. Hyatt , 566 U.S. 431, 444–45, 132 S.Ct. 1690, 182 L.Ed.2d 704 (2012) ). Precisely this occurred here; both parties submitted substantial new evidence in addition to the PTO record, and thus de novo review of the entire factual record is required in this case.

In the instant § 1071(b) action, plaintiff contends that the TTAB erred in determining that defendant's VAGISAN mark is not likely to cause confusion with plaintiff's VAGISIL mark and that the VAGISAN mark is therefore registrable. The Lanham Act provides that denial of an application to register a mark is warranted when the mark "so resembles a mark registered in the Patent and Trademark Office ... as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive." 15 U.S.C. § 1052(d). The Fourth Circuit in Swatch , 739 F.3d at 158, explained that the following nine factors guide the determination whether an applied-for mark is likely to cause confusion with a registered mark:

(1) the strength or distinctiveness of the senior mark as actually used in the marketplace;

(2) the similarity of the two marks to consumers;

(3) the similarity of the goods or services that the marks identify;

(4) the similarity of the facilities used by the markholders;

(5) the similarity of advertising used by the markholders;

(6) the defendant's intent;

(7) actual confusion;

(8) the quality of the defendant's product; and

(9) the sophistication of the consuming public.

It is well-settled that these nine factors are not all equally important and that certain factors may not be relevant in a given case. George & Co. LLC v. Imagination Entertainment Ltd. , 575 F.3d 383, 393 (4th Cir. 2009) (citing Anheuser-Busch, Inc. v. L & L Wings, Inc. , 962 F.2d 316, 320 (4th Cir. 1992) ).

Before proceeding to apply these factors, it is important to resolve the parties' threshold dispute whether the likelihood of confusion analysis in this § 1071(b) action should be based on the VAGISAN mark as depicted and described in the application or as used in the marketplace. Plaintiff's argument is correct; because this case involves only the issue whether defendant's mark, VAGISAN, is registrable, the likelihood of confusion factors must take into account only the depiction and description of the VAGISAN mark contained in defendant's application to register the mark, not defendant's actual use of VAGISAN in the marketplace.

It is clear that in a registration opposition proceeding before the TTAB, the likelihood of confusion analysis considers only "the mark as shown in the application and as used on the goods described in the application ... not the mark as actually used." B & B Hardware, Inc. v. Hargis Indus., Inc. , ––– U.S. ––––, 135 S.Ct. 1293, 1307, 191 L.Ed.2d 222 (2015).⁶ And as another district court in this district has sensibly concluded, this principle does not change when the TTAB's registrability determination is appealed to a district court pursuant to § 1071(b). Seacret Spa Int'l v. Lee , No. 1:15CV405(JCC/IDD), 2016 WL 880367, at *3 (E.D. Va. Mar. 8, 2016) (citing B & B Hardware , 135 S. Ct. at 1300 ). Specifically, the Seacret Spa court held that when a district court in a § 1071(b) action determines the registrability of a mark, the district court, "like the TTAB, must look to the full scope of usages disclosed within the four corners of the application." Id. Put simply, "the court, like the T.T.A.B. itself, is looking only to the mark as shown in the application and as used on those goods or services listed in the application." 3 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 21:20 (5th ed. 2019). Consequently, for purposes of determining the registrability of an applied-for mark in a § 1071(b) action, "[a]ny limitations on usage must appear in the application," and "evidence of the mark's actual, narrower use in the marketplace is irrelevant." Seacret Spa , 2016 WL 880367, at *3.

6 See also Octocom Sys., Inc. v. Houston Computer Servs., Inc. , 918 F.2d 937, 942 (Fed. Cir. 1990) (collecting cases); In re Elbaum , 211 U.S.P.Q. 639, 640 (T.T.A.B. 1981).

The Seacret Spa court's reasoning is persuasive, and thus guides the analysis here, because if the district court determines that the applied-for mark is registrable, the markholder receives rights and protections in the mark that cover all possible presentations and usages of the mark, restricted only by the limitations in the application itself.⁷ For example, "[i]f the registrant ... obtains a standard character mark without claim to any particular font, style, size or color, the registrant is entitled to depictions of the standard character mark regardless of font style, size, or color." Citigroup Inc. v. Capital City Bank Group Inc. , 637 F.3d 1344, 1353 (Fed. Cir. 2011). In other words, if the registration does not limit the appearance of the mark to certain trade dress, stylization, trade channels of distribution, the markholder can change such uses at any time.⁸ Accordingly, a district court in a § 1071(b) action should not determine that an applied-for mark is not likely to cause confusion based on how the mark is currently used in the marketplace unless the registration application limits the markholder to such usage. Instead, the district court, like the TTAB, must base its likelihood of confusion analysis on the description of the applied-for mark and any limitations on usage that are included in the application.

7 See 15 U.S.C. § 1057(b) (providing that registration on the Principal Register "shall be prima facie evidence ... of the owner's exclusive right to use the registered mark in commerce on or in connection with the goods or services specified in the certificate , subject to any conditions or limitations stated in the certificate ") (emphasis added).

8 See Kimberly-Clark Corp. v. H. Douglas Enterprises, Ltd. , 774 F.2d 1144, 1147 (Fed. Cir. 1985) ("It is settled ... that a distinction in trade dress cannot weigh against likelihood of confusion with respect to the registration of a simple word mark ... [because] such dress might well be changed at any time; only the word mark itself is to be registered."); Seacret Spa , 2016 WL 880367, at *4 ("Because the ‘SECRET’ mark is registered as a standard character mark, its registration extends to a rendition of the word ‘secret’ in the same font as the ‘SEACRET’ mark, complete with a wave design element in the middle of the ‘E’.").
Moreover, when the only issue before the district court in a § 1071(b) action is whether an applied-for mark is registrable, it will often be true, as is the case here, that the applicant's mark has not yet been used in United States commerce. Otherwise, one would expect an infringement claim to accompany the registrability challenge. In those circumstances, there is no reliable evidence of actual use for the district court to consider. To illustrate this point, in the instant case, defendant proposes that the likelihood of confusion analysis should take into account the packaging for VAGISAN products sold by defendant in foreign commerce and the packaging for "gray market" VAGISAN products that are made available for sale on Amazon.com by third parties without authorization by defendant. But this packaging is not reliable evidence of actual marketplace use in the United States because one of defendant's executives, Ms. Thevessen, testified that defendant has not yet determined what stylized format, wording, design elements, colors, or other distinctive trade dress defendant will use to display the VAGISAN mark on packaging in United States commerce.

Seeking to avoid this conclusion, defendant argues that the Fourth Circuit in Swatch , 739 F.3d at 160, rejected this approach by holding that a district court in a § 1071(b) action should consider actual marketplace use. This argument fails because Swatch involved not only an appeal of the TTAB's registrability determination pursuant to § 1071(b), but also a trademark infringement claim, which renders Swatch distinguishable. See Swatch , 739 F.3d at 159-61. Although "the same likelihood-of-confusion standard applies to both registration and infringement," see B & B Hardware , 135 S. Ct. at 1306, the standard is applied differently to an infringement claim. In a trademark infringement claim, the focus of the likelihood of confusion inquiry is whether "the defendant's actual practice is likely to produce confusion in the minds of consumers about the origin of the goods or services in question." CareFirst of Maryland, Inc. v. First Care, P.C. , 434 F.3d 263, 267 (4th Cir. 2006) (emphasis added). In sharp contrast, the issue at the heart of a registrability determination is whether a mark, as delimited in the application, is registrable. See Octocom Sys., Inc. v. Houston Computer Servs., Inc. , 918 F.2d 937, 942 (Fed. Cir. 1990) ("The issue in [a registration] opposition is the right of an applicant to register the mark depicted in the application for the goods identified therein."). In short, when a case involves only an appeal of the TTAB's registrability determination pursuant to § 1071(b) and not also an infringement claim, it makes little sense to consider actual use of the mark, for the reasons already stated. The Fourth Circuit's consideration of such actual use evidence in a § 1071(b) action in Swatch is therefore distinguishable and is not controlling here.

To recapitulate, analysis whether VAGISAN is likely to cause confusion with VAGISIL, and is thus ineligible for registration, proceeds by applying each of the nine likelihood of confusion factors to the parties' marks. And pursuant to persuasive precedent from the Eastern District of Virginia, this analysis considers only the descriptions of VAGISAN and the limitations on usage of the mark that are included in defendant's application, without taking into account any narrower uses of the VAGISAN mark in the marketplace.

1. The Strength or Distinctiveness of the Senior Mark

The first likelihood of confusion factor focuses on the strength of the senior mark, VAGISIL, and is " ‘paramount’ in determining the likelihood of confusion." Grayson O Co. v. Agadir Int'l LLC , 856 F.3d 307, 314–15 (4th Cir. 2017) (quoting Pizzeria Uno Corp. v. Temple , 747 F.2d 1522, 1527 (4th Cir. 1984) ). As the Fourth Circuit has explained, in general, "the stronger the mark, the greater the likelihood that consumers will be confused by competing uses of the mark." George & Co. , 575 F.3d at 393. In addition, a strong, famous mark is entitled to a broad scope of protection against confusingly similar marks. Kenner Parker Toys Inc. v. Rose Art Indus., Inc. , 963 F.2d 350, 354 (Fed. Cir. 1992) ; Mobil Oil Corp. v. Pegasus Petroleum Corp. , 818 F.2d 254, 258 (2d Cir. 1987) ; 2 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 11:73 (5th ed. 2019).

A mark's "strength" is measured by "the degree to which a consumer in the relevant population, upon encountering the mark, would associate the mark with a unique source." CareFirst , 434 F.3d at 269. In this respect, the Fourth Circuit has explained that a mark's strength is evaluated in light of the mark's "conceptual strength" and "commercial strength," id. , each of which is separately considered below.

A.

A mark's "conceptual or inherent strength" is determined by "focus[ing] on the linguistic or graphical ‘peculiarity’ of the mark [ ] considered in relation to the product, service, or collective organization to which the mark attaches." Id. (internal citation omitted). This inquiry considers both (i) the degree to which the mark describes the product branded by the mark and (ii) the similarity of other marks in the same field of merchandise. See Grayson O , 856 F.3d at 315 (citing George & Co. , 575 F.3d at 393–94 ; CareFirst , 434 F.3d at 270 ).

With respect to the first factor, a mark's conceptual strength "is determined in part by its placement into one of four categories of distinctiveness: (1) generic; (2) descriptive; (3) suggestive; or (4) arbitrary or fanciful." George & Co. , 575 F.3d at 393–94. In the instant case, the parties dispute whether plaintiff's VAGISIL mark is suggestive or descriptive. Suggestive marks "connote, without describing, some quality, ingredient, or characteristic of the product." Sara Lee Corp. v. Kayser-Roth Corp. , 81 F.3d 455, 464 (4th Cir. 1996). In contrast, descriptive marks "merely describe a function, use, characteristic, size, or intended purpose of the product." Id. Put another way, "if the mark imparts information directly, it is descriptive," but "[i]f it stands for an idea which requires some operation of the imagination to connect it with the goods, it is suggestive." Pizzeria Uno , 747 F.2d at 1528. The significance of a mark's classification into one of these two categories is that suggestive marks are deemed to be inherently distinctive and strong, whereas descriptive marks are not inherently distinctive and are thus comparatively weak. Sara Lee , 81 F.3d at 464 ; Pizzeria Uno , 747 F.2d at 1527.

As an initial matter, it is important to note that the PTO has issued plaintiff numerous registrations for the VAGISIL mark without requiring proof of secondary meaning. In so doing, the PTO concluded the mark was suggestive, not descriptive, and the PTO's determination creates a rebuttable presumption that VAGISIL is in fact suggestive. See Synergistic Int'l, LLC v. Korman , 470 F.3d 162, 172 (4th Cir. 2006) ; Pizzeria Uno , 747 F.2d at 1529, 1533.

Evaluation of the VAGISIL mark confirms that is a suggestive mark, as the PTO concluded. As a whole, "Vagisil" is a made-up word that does not directly "describe a function, use, characteristic, size, or intended purpose" of the feminine care products sold under the mark. Sara Lee , 81 F.3d at 464. Instead, the word "Vagisil" merely "connotes" the vaginal area of a woman's body; a consumer must exercise some degree of "imagination to connect it with the goods," namely feminine care and hygiene products. Id. ; Pizzeria Uno , 747 F.2d at 1528. And importantly, the mark's suffix "-sil" appears to be meaningless and does not describe or even connote a feminine care product. This feature of the mark further weighs in favor of finding that VAGISIL is suggestive and not descriptive. See Syntex Labs., Inc. v. Norwich Pharmacal Co. , 315 F. Supp. 45, 51 (S.D.N.Y. 1970), aff'd , 437 F.2d 566 (2d Cir. 1971) (holding that the strong suggestiveness of the prefix "vag-" is attenuated when joined with the suffix "-trol" to form the word "Vagitrol"). Accordingly, the PTO's conclusion that VAGISIL is a suggestive mark is amply supported. But concluding that VAGISIL is a suggestive mark does not end the analysis of the mark's conceptual strength. The Fourth Circuit has made clear that even a suggestive mark's conceptual strength is diminished if many third parties in the same field as the mark holder have used a portion of the text of the senior mark in their own marks. CareFirst , 434 F.3d at 269-70 (concluding that the conceptual strength of the suggestive mark "CareFirst" was weakened by "substantial third-party use of the words ‘Care,’ ‘CareFirst,’ ‘First,’ and ‘First Care’ in the health care industry"); 2 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 11:90 (5th ed. 2019) ("Third party registrations are relevant to prove that some segment of the composite marks which both contesting parties use has a normally understood and well-recognized descriptive or suggestive meaning, leading to the conclusion that that segment is relatively weak.").

In the instant case, defendant has submitted evidence of approximately sixty-six third-party marks containing "vagi-" as a prefix, including sixteen marks used in commerce, eleven live registrations and applications, and thirty-nine applications and registrations that are cancelled, expired, or otherwise "dead." See supra , Findings of Fact ¶¶ 41–16. This evidence reflects that a significant number of third parties have at one time or another used a mark that is similar to VAGISIL, many of which were used in connection with vaginal care products. Indeed, the number of similar third-party marks introduced by defendant in this case is comparable to the amount that other courts have deemed sufficient to weaken the distinctiveness of a mark. See Petro Stopping Centers, L.P. v. James River Petroleum , Inc. , 130 F.3d 88, 93 (4th Cir. 1997) (citing cases holding that approximately seventy third-party registrations of similar marks weakened the distinctiveness of the plaintiff's mark). Thus, this evidence of similar third-party marks serves to weaken the conceptual strength of VAGISIL because if the "mark were truly a distinctive term, it is unlikely that so many other businesses in the ... industry would independently think of using the same mark or similar variants of it." CareFirst , 434 F.3d at 270.

In response, plaintiff advances two arguments aimed at showing that this evidence of third-party marks is entitled to little weight. First, plaintiff argues that these third-party marks are not sufficiently similar to VAGISIL to weaken VAGISIL's conceptual strength because none of them contain a "vagi-" prefix, then an "s", then a vowel, and then a consonant. This argument fails because plaintiff has pointed to no authority for the proposition that third-party marks must share such a close degree of similarity to be relevant. To the contrary, courts have concluded that extensive third-party use of a component of the plaintiff's mark, such as the prefix, undermines the conceptual strength of the plaintiff's mark as a whole.⁹

9 See, e.g. , Water Pik, Inc. v. Med-Sys., Inc. , 726 F.3d 1136, 1152 (10th Cir. 2013) (holding that the widespread use of "Sinu-" or "Sinus-" prefixes in third-party marks weakened the distinctiveness of the plaintiff's SINUCLEANSE mark); CareFirst , 434 F.3d at 270 (holding that substantial use of the words "Care" and "First" in third-party marks, and not only use of "CareFirst" and "FirstCare," weakened the conceptual strength of the plaintiff's "CAREFIRST" mark); The Land-O-Nod Co. v. Paulison , 220 U.S.P.Q. 61, 66–67 (T.T.A.B. 1983) (holding that the common use of "Chiro-" in third-party marks weakened the distinctiveness of the applicant's CHIRO-MATIC mark).

Second, plaintiff argues that the evidence of similar third-party marks submitted by defendant is insufficient to weaken the strength of VAGISIL because most of these third-party marks are commercially insignificant. This argument fails in the conceptual strength context¹⁰ because it is well-established that, even without evidence of commercial impact, the widespread use of similar marks by third parties is "powerful on its face" to weaken the inherent distinctiveness of the plaintiff's mark. Juice Generation, Inc. v. GS Enterprises LLC , 794 F.3d 1334, 1339 (Fed. Cir. 2015).¹¹ This is so because similar third-party marks "are useful to demonstrate the sense in which a term is used in ordinary parlance and they can show that a particular term has been adopted by those engaged in a certain field or industry and that said term has less than arbitrary significance with respect to certain goods or services." Spraying Sys. Co. v. Delavan, Inc. , 762 F. Supp. 772, 778 (N.D. Ill. 1991), aff'd , 975 F.2d 387 (7th Cir. 1992). In short, the evidence of third-party marks in this case is significant, despite the commercial insignificance of many of the marks, because it diminishes the "linguistic or graphical ‘peculiarity’ " of plaintiff's VAGISIL mark, which it the focus of the conceptual strength inquiry.

10 But as explained below, infra part 1.B, the commercial insignificance of the third-party marks in this case does render the third-party marks immaterial with respect to measuring VAGISIL's commercial strength.

11 See also Petro Stopping Ctrs. , 130 F.3d at 93 ("The company maintains that only proof that third parties actually use the term PETRO would be relevant. We disagree. The frequency with which a term is used in other trademark registrations is indeed relevant to the distinctiveness inquiry under the first likelihood of confusion factor.").
Indeed, courts have even recognized that expired and cancelled registrations of third-party marks are relevant to show the inherent indistinctiveness of the senior mark. See, e.g. , Anthony's Pizza & Pasta Int'l, Inc. v. Anthony's Pizza Holding Co. , 95 U.S.P.Q.2d 1271, 1277 (T.T.A.B. 2009) (term "ANTHONY'S" weak and indistinctive in light of third-party registrations, including "8 third-party cancelled/expired registrations"), aff'd , 415 F. App'x 222 (Fed. Cir. 2010).

In sum, although VAGISIL is a suggestive mark, the prevalent use of similar "vagi-" prefix marks by third parties weakens the conceptual strength of VAGISIL. Analysis thus proceeds to consider the second prong of the inquiry into VAGISIL's overall strength, namely the VAGISIL mark's commercial strength.

B.

The commercial strength of a mark is measured by determining whether, in the marketplace, "a substantial number of present or prospective customers understand the designation when used in connection with a business to refer to a particular person or business enterprise." CareFirst , 434 F.3d at 269. The Fourth Circuit has explained that evaluation of the commercial strength of a mark should take into account the same factors used in the conceptually similar "secondary meaning" test, namely "(1) advertising expenditures; (2) consumer studies linking the mark to a source; (3) sales success; (4) unsolicited media coverage of the product; (5) attempts to plagiarize the mark; and (6) the length and exclusivity of the mark's use." Grayson O , 856 F.3d at 316 (quoting Perini Corp. v. Perini Const., Inc. , 915 F.2d 121, 125 (4th Cir. 1990) ).

Here, the evidence submitted by plaintiff demonstrates that the VAGISIL brand has enjoyed considerable commercial success and that a substantial portion of the consuming public understands the VAGISIL mark as signifying a particular source for vaginal care products. VAGISIL-branded products are sold nationwide in a wide assortment of dollar stores, drug stores, grocery stores, mass merchandising stores, pharmacies, and e-tailers. See supra , Findings of Fact ¶ 7. Since 1991, VAGISIL-branded products have generated over one billion dollars in sales revenue and have commanded between 55% and 60% market share in the feminine anti-itch cream segment and between 50% and 69% market share in the feminine powder segment. And since 2010, VAGISIL has had at least 94% market share in the medicated wipes segment. See id. ¶ 5.

Perhaps even more telling of VAGISIL's commercial success is the persuasive marketing data from the past year reflecting that in the broad intimate-health product category, which includes segments in which VAGISIL-branded products do not compete, VAGISIL-branded anti-itch cream ranks fifth in sales revenue and VAGISIL-branded wash ranks sixth. In the segments where VAGISIL-branded products do compete, the data are even more impressive. In the anti-itch cream segment, VAGISIL-branded anti-itch cream is ranked first in both units sold and sales revenue. In the anti-itch wipe segment, VAGISIL-branded wipes occupy the top two ranks in both units sold and sales revenue. In the intimate wash segment, VAGISIL-branded washes are ranked first and second in sales revenue and first and third in units sold. In the intimate powder segment, VAGISIL-branded powder ranks first in both units sold and sales revenue. In the moisturizing lubricant segment, VAGISIL-branded moisturizing lubricant ranks third in both units sold and sales revenue. See id. ¶ 6.

In addition to sales, the VAGISIL brand has attained substantial public recognition through widespread advertising and unsolicited media attention. Plaintiff has marketed the VAGISIL brand for over forty years through numerous forms of media, including print, television, radio, the Internet, events, public relations campaigns, coupons, samples, and displays in stores. See id. ¶¶ 11–16. Since 1993, plaintiff has spent over $ 425 million advertising and promoting VAGISIL-branded products to the general consuming public, which is comparable to the amount spent on advertising by other major intimate-health brands, including KY and Plan B, and exceeds spending by some of VAGISIL's key competitors, such as SUMMER'S EVE and REPLENS. See id. ¶¶ 11–12. Prominent television stations have aired over ninety VAGISIL commercials, over one-hundred radio stations have featured VAGISIL promotions, and online advertisements for VAGISIL generate millions of impressions by Internet users each year. See id. ¶¶ 13, 14, 16. In addition, the VAGISIL brand has been mentioned in several popular, nationally televised television programs, including episodes of South Park and The Big Bang Theory , and the VAGISIL brand was the subject of an entire skit on Saturday Night Live . See id. ¶ 17. VAGISIL products have also been the subjects of articles in leading national publications, including The New York Times, The Wall Street Journal , and The Washington Post . See id. ¶ 18.

This evidence convincingly demonstrates that the VAGISIL brand is one of the most commercially successful brands for intimate-health products on the market and has enjoyed substantial, nationwide publicity. The sales figures alone exceed those that courts have found to be indicative of fame. See, e.g. , Nina Ricci, S.A.R.L. v. E.T.F. Enterprises, Inc. , 889 F.2d 1070, 1073 (Fed. Cir. 1989) (holding that NINA RICCI was a strong mark for perfume, clothing, and accessories based on $ 200 million in sales over 5 years of use); Specialty Brands, Inc. v. Coffee Bean Distributors, Inc. , 748 F.2d 669, 674 (Fed. Cir. 1984) (SPICE ISLANDS famous mark based on forty years of use, annual sales of $ 25,000,000 for seasonings and $ 12,000,000 for teas). Further evidence of VAGISIL's commercial strength is the evidence plaintiff introduced of the survey conducted by Mr. Hal Poret designed to test whether the VAGISIL mark is famous. See supra , Findings of Fact ¶¶ 30–32. The results of the Fame Survey provide substantial empirical data to reinforce the conclusion that the VAGISIL brand is well-recognized by consumers. In the unaided portion of the Fame Survey, three hundred United States consumers, both men and women, were asked to list all brands of vaginal care products that they had ever seen or heard of. VAGISIL was named by a total of 38.7% of the survey respondents, which was a far greater level of unaided awareness than any other brand that was mentioned.¹² In the aided portion of the Fame Survey, the survey respondents were shown eight brand names, one at a time and in random order, and were asked whether they had ever seen or heard of the name in connection with any vaginal care product. The eight names included VAGISIL, along with an assortment of well-known vaginal care product names, lesser known but commonly available product names, relatively unknown product names, and a fictional product name, VAGIZOX, which served as the control. VAGISIL was recognized by 90% of all respondents, which was a far higher rate of recognition than any of the other seven brand names received. Only 5.3% of respondents stated they had seen or heard of the control, VAGIZOX, in connection with any vaginal care product. Accordingly, the Fame Survey reported a net awareness rate of 85% for VAGISIL; i.e. it reflected that 85% of respondents recognized VAGISIL specifically because of the mark's fame and not because of other factors.

12 For reference, the second most frequently named brand in the unaided portion of the survey was SUMMER'S EVE, which was listed by only 26% of all respondents. Notably, the unaided awareness level by all respondents with respect to SUMMER'S EVE, the second most frequently named brand, is similar to the 24.1% unaided awareness level by male respondents with respect to VAGISIL.

The results of the Fame Survey are substantial and certainly support a finding that the VAGISIL mark is commercially famous. Indeed, the 38.7% level of unaided awareness and the 85% rate of net aided recognition that VAGISIL received in the survey far surpass the degree of recognition from similar surveys that courts have found sufficient to demonstrate the commercial strength of a mark. See, e.g. , Ringling Bros.-Barnum & Bailey Combined Shows, Inc. v. Utah Div. of Travel Dev. , 955 F. Supp. 605, 613 (E.D. Va. 1997), affd , 170 F.3d 449 (4th Cir. 1999) (holding that a 41% rate of aided recognition was sufficient to show that the plaintiff's mark was famous); Schieffelin & Co. v. Jack Co. of Boca , 850 F. Supp. 232, 242–43 (S.D.N.Y. 1994) (holding that a 1% unaided recognition rate and a 72% aided recognition rate were sufficient to support a finding that the plaintiff's mark was commercially famous). Therefore, the results of the Fame Survey are persuasive, empirical evidence of the VAGISIL mark's commercial strength and, along with the powerful sales and marketing data examined above, clearly demonstrate that VAGISIL is a famous mark that enjoys substantial commercial strength.

Seeking to diminish the force of this evidence, defendant argues (i) that the existence of similarly named third-party marks weaken the commercial strength of VAGISL and (ii) that plaintiff's Fame Survey suffers from certain defects that call into question the level of consumer awareness of VAGISIL reported by the survey. These arguments are dispensed with in turn.

First, defendant argues that the use of marks containing a "vagi-" prefix by third parties demonstrate that plaintiff's VAGISIL mark is commercially weak. This argument fails because it is well-established that the existence of similar third-party marks alone is insufficient to demonstrate that a mark is commercially weak; the impact of third-party marks on the commercial strength of the senior mark depends wholly on the extent to which they are used and promoted by third-parties and are recognized by consumers.¹³ The reason for this principle is simple. Evidence that other sellers in the same industry as the plaintiff extensively use marks that are similar to the plaintiff's mark shows that "consumers may not ‘associate [the plaintiff's] mark with a unique source,’ " and that plaintiff's mark is thus commercially weak. Variety Stores , 888 F.3d at 664 (quoting Grayson O , 856 F.3d at 315 ). But if third-party marks are not in use in the market or have meager sales and advertising figures, it is unlikely that consumers are familiar with the marks, which renders them irrelevant to the question whether consumers associate the plaintiff's mark with a unique source.¹⁴ CAE, Inc. v. Clean Air Eng'g, Inc. , 267 F.3d 660, 685 (7th Cir. 2001) ; JIPC Mgmt., Inc. v. Incredible Pizza Co. , No. CV0804310MMMPLAX, 2008 WL 11336396, at *15 (C.D. Cal. Aug. 26, 2008) ; Electropix v. Liberty Livewire Corp. , 178 F. Supp. 2d 1125, 1130 (C.D. Cal. 2001) ; Johnson & Johnson , 2009 WL 4075364, at *13 (T.T.A.B. Nov. 2, 2009).

13 See McGraw-Edison Co. v. Walt Disney Prods. , 787 F.2d 1163, 1171 (7th Cir. 1986) ; Scarves by Vera, Inc. v. Todo Imports Ltd. , 544 F.2d 1167, 1173 (2d Cir. 1976) ; Select Auto Imports Inc. v. Yates Select Auto Sales, LLC , 195 F. Supp. 3d 818, 833 (E.D. Va. 2016) ; Lexington Mgmt. Corp. v. Lexington Capital Partners , 10 F. Supp. 2d 271, 281 (S.D.N.Y. 1998) ; Miss Universe, Inc. v. Little Miss U.S.A., Inc. , No. C80-1364A, 1981 WL 40568, at *3 (N.D. Ga. June 19, 1981).

14 It is worth observing that the relevance of third-party marks differs under the conceptual strength and commercial strength inquiries. As explained above, supra part l.A, evidence of similar third-party marks is relevant to weaken the conceptual strength of the plaintiff's mark even without evidence showing that the third-party marks are commercially significant. In contrast, the commercial strength of the plaintiff's mark is not diminished by third-party marks that are not extensively used, sold, and advertised in the marketplace. This apparent incongruity is entirely sensible given the entirely distinct issues that are at the heart of the conceptual strength and commercial strength inquiries. Conceptual strength focuses on the linguistic uniqueness of the plaintiff's mark, whereas commercial strength hinges on whether consumers in fact associate the plaintiff's mark with a unique source. CareFirst , 434 F.3d at 269. Clearly, the mere existence of similar marks is sufficient to show that the plaintiff's mark is not unique but sheds little light, without more, on whether consumers' perceptions have been impacted. Thus, although the third-party marks submitted by defendant may weaken the conceptual strength of VAGISIL even without evidence of extensive use, sales, advertising, or recognition by consumers of the third-party marks, that evidence is insufficient to weaken the strong commercial strength of VAGISIL.

Here, the evidence demonstrates that the third-party marks relied upon by defendant are commercially insignificant and thus do not materially weaken the commercial strength of plaintiff's VAGISIL mark. Defendant presented no evidence to establish that the majority of the sixty-six third-party "vagi-" prefix marks submitted by defendant have ever been sold or promoted in the United States. Indeed, thirty-nine of these marks appear to correspond to registrations and applications that are cancelled, are expired, or are otherwise "dead," while eleven others appear to be live registrations or applications for which there is no evidence that the marks have ever been used in commerce. See supra , Findings of Fact ¶¶ 45–46.

Furthermore, fourteen of the fifteen marks that defendant established are used by third parties in commerce¹⁵ appear to have achieved extremely limited sales, promotion, and recognition by consumers. With respect to seven of these marks, the only sale defendant could identify was a single purchase made by defendant's defense team during the course of litigating this case. Id. ¶¶ 40–41. Five other marks appear to have generated between $ 0 and $ 5,309 in annual sales and between $ 0 and $ 988 in annual advertising spending during the last five years. Id. ¶ 40. Another brand, VAGI-CURE, garnered between $ 0 and $ 15,629 in sales from 2013 to 2017, with no evidence of advertising expenditures, but the product sold in connection with that brand has since been discontinued. Id. The VAGIKOOL brand has generated slightly higher sales and advertising numbers—between $ 25 and $ 54,475 in annual sales and between $ 0 and $ 12,194 in annual advertising since 2015. Id. But the evidence also reflects that VAGIKOOL is used in connection with the sale and promotion of reusable cold packs, which are not closely competitive with the types of personal care products that are sold under the VAGISIL brand. In sum, the evidence reflects that none of these third-party marks are used in commerce in connection with products that are extensively sold, advertised, or otherwise recognized by consumers and that some of these marks were used in connection with products that are not sufficiently similar to VAGISIL-branded products.¹⁶ Thus, the evidence of third-party use submitted by defendant is entitled to minimal weight and certainly does not diminish the strong evidence of VAGISIL's commercial strength. See CAE , 267 F.3d at 685 (7th Cir. 2001) ; Scarves by Vera , 544 F.2d at 1173–74.

15 Aside from the fifteen third-party marks discussed above, defendant also presented evidence with respect to the use and promotion of the VAGISTAT, VAGISTAT-1, and VAGISTAT-3 brands. But the evidence also reflects that in May 2017 plaintiff purchased these VAGISTAT marks and has begun to sell VAGISTAT products in packaging labeled "VAGISTAT by VAGISIL Brand." As a consequence, the VAGISTAT marks are no longer third-party marks, and the packaging in which VAGISTAT products are now sold actually reinforce the strength of the VAGISIL brand. Moreover, even before plaintiff bought them, the VAGISTAT marks were used in connection with yeast-infection products, with which plaintiff's products did not compete. Thus, evidence concerning the use of the VAGISTAT brands is irrelevant to the analysis of VAGISIL's commercial strength.

16 The conclusion that the third-party marks relied upon by defendant are commercially insignificant finds further support in the Fame Survey conducted by Mr. Poret. In the unaided portion of the survey, less than 1% of respondents named any third-party "vagi-" prefix mark as a brand that they had seen or heard of. One respondent out of three hundred, or 0.3% of all respondents, named Vagicare, and one respondent out of three hundred, or 0.3% of all respondents, named Vaginex. This empirical evidence reinforces the conclusion that the third-party marks submitted by defendant are commercially insignificant.

In addition to these commercially insignificant marks, defendant also introduced evidence that several prominent mass merchandising and drug stores have generated substantial sales in connection with a generic vaginal anti-itch cream that is sold in packaging labeled "Vagicaine." Notwithstanding the significant sales of these "Vagicaine" products, this evidence is entitled to little weight because the packaging of each of these generic "Vagicaine" products prominently features the store's private-label brand (i.e. , EQUATE, UP & UP, RITE AID, and CVS). Courts have sensibly concluded in trade dress infringement cases that consumers focus on these well-known, store-specific signatures to identify the store as the source of the generic, private-label products and to distinguish these generic products from the name brand products they are designed to imitate. See McNeil Nutritionals, LLC v. Heartland Sweeteners, LLC , 511 F.3d 350, 353–54, 361, 368 (3d Cir. 2007) ; Conopco, Inc. v. May Dep't Stores Co. , 46 F.3d 1556, 1568 (Fed. Cir. 1994) ; Warner Lambert Co. v. McCrory's Corp. , 718 F. Supp. 389, 398 (D.N.J. 1989). Applied here, this principle points persuasively to the conclusion that consumers do not think of "Vagicaine" as a source-identifying brand; rather, consumers recognize the "Vagicaine" label only as a signal that the generic product seeks to imitate VAGISIL's anti-itch cream.¹⁷ Thus, the presence of the "Vagicaine" label on the packaging of generic, private-label vaginal anti-itch cream products does not materially diminish VAGISIL's strength as a source-identifier because consumers recognize the store-specific signatures, and not the "Vagicaine" label, as the source-identifying element of the "Vagicaine" products' trade dress.

17 To illustrate this point, in the unaided portion of the Fame Survey, 1.7% of the respondents named EQUATE, Walmart's private-label store brand for generic products, as a brand they had seen or heard of in connection with the sale of vaginal care products. Not a single respondent named "Vagicaine" as a brand name that they knew.

Defendant's second set of arguments aimed at attacking the evidence of VAGISIL's commercial strength include two criticisms of the methodology utilized by Mr. Poret's Fame Survey. First, defendant argues that the aided portion of the Fame Survey used a poor control, which caused the survey to report an inflated net rate of consumer recognition of VAGISIL. In particular, according to defendant's expert, Dr. Simonson, the "VAGIZOX" control does not sound like a plausible name for a vaginal care product because "-zox" is an uncommon suffix for health, beauty, or feminine care products. Thus, Dr. Simonson explained, the VAGIZOX control would not adequately measure or capture spurious awareness—i.e. the tendency of respondents to answer that they are aware of a brand name even if they have not have heard of it—because respondents would be able to tell VAGIZOX is a fake brand name.

This criticism is unpersuasive. The VAGIZOX control included a "vagi-" prefix, like other marks registered in connection with vaginal care products. And plaintiff adduced evidence at trial that other marks including "-zox" and "-ox" have been registered and used in connection with pharmaceutical products.¹⁸ In the end, defendant presented no empirical evidence to support its argument that consumers find the "-zox" suffix strange or to demonstrate that a different control would have detected a higher level of spurious awareness.¹⁹ Indeed, Mr. Poret testified that the 5.3% awareness rate recorded in connection with the VAGIZOX control is consistent with his experience of measuring spurious awareness in brand recognition surveys. Thus, it appears that there is no basis to conclude that VAGIZOX was an implausible brand name or that it was an ineffective control.

18 Although none of these other "-zox" and "-ox" marks are used in connection with vaginal care products, Mr. Poret convincingly testified that a proper control cannot be too similar to an existing vaginal care product because respondents might claim to have heard of the control term because they have confused it with an existing brand name that the respondents have actually heard of. Under those circumstances, the control would not capture only answers reflecting spurious awareness, and the net awareness rate of the subject mark would thus be improperly underreported by the survey.

19 Notably, a review of the third-party registrations that defendant introduced in this case reveal that much stranger brand names have been registered in connection with products that treat vaginal conditions, such as VAGI-HEX, VAGICIAN, VAGITARIAN, and VAGICULT.

Second, defendant argues that the aided portion of the Fame Survey should have used more than one control to detect and measure spurious, mistaken awareness. According to Dr. Simonson, every potential brand name has certain idiosyncrasies that prevent it from serving as a proper control. Thus, it would have been preferable for the Fame Survey to have included additional fictitious control names so that the recognition rates of each of these controls could have been averaged together to produce a more reliable estimate of the spurious, mistaken awareness that occurred in the survey.

Although Dr. Simonson's recommendation appears to be conceptually sensible, it does little to undermine the reliability of the Fame Survey. Indeed, defendant presented no empirical evidence, such as a survey implementing Dr. Simonson's views, to demonstrate that using multiple controls would have measured a materially higher rate of spurious awareness. Furthermore, as Mr. Poret testified at trial, the inclusion of existing but relatively unknown brand names in the aided portion of the Fame Survey served a control-like function and showed that the VAGIZOX control was not materially underestimating the rate of spurious awareness present in the survey. In particular, the fact that the survey reported meager recognition rates of between 5% and 18% for SWEETSPOT, LUVENA, and VAGI-GARD demonstrated that the survey was not reporting inflated, artificially high recognition rates for unknown marks. Thus, the 85% net recognition rate of VAGISIL cannot be explained by spurious awareness, and there is no ground on which to conclude that the inclusion of additional control terms in the Fame Survey would have materially altered this result.

In any event, the force of defendant's two criticisms of the aided portion of the Fame Survey is significantly blunted by the convincing results of the survey's unaided portion, which defendant has not attempted to attack. Clearly, the fact that 38.7% of survey respondents named VAGISIL as a brand name they knew without any type of prompting goes a long way to dispel the concern that the aided portion's report of an 85% net awareness rate of VAGISIL is unreliable.

In summary, the above evidence of VAGISIL's sales, marketing, unsolicited media attention, and brand recognition by consumers persuasively demonstrates that VAGISIL is a famous mark that has attained substantial commercial strength in the marketplace.

C.

Importantly, the parties dispute whether conceptual strength or commercial strength carries more weight in the overall determination of a senior mark's strength. On this issue, the Fourth Circuit has held that "in certain circumstances, the commercial strength of a mark can be more important than the conceptual strength." Variety Stores, Inc. v. Wal-Mart Stores, Inc. , 888 F.3d 651, 663 (4th Cir. 2018). Specifically, "if a mark has sufficient commercial strength such that consumers would associate the mark with a unique source, it may be considered strong despite its conceptual weakness." Id. (internal citation and quotation marks omitted).²⁰ This is so because "the ultimate inquiry under [the strength of the senior mark] factor [is] ‘the degree to which the designation is associated by prospective purchasers with a particular source.’ " Renaissance Greeting Cards, Inc. v. Dollar Tree Stores, Inc. , 405 F. Supp. 2d 680, 690–91 (E.D. Va. 2005), aff'd , 227 F. App'x 239 (4th Cir. 2007) (quoting Petro Stopping Ctrs. , 130 F.3d at 93 ).

20 For example, "marks such as AMERICAN airlines, PAYLESS shoe stores, FORD autos and KENTUCKY FRIED CHICKEN fast-food outlets would, at birth, have been characterized as inherently ‘weak’ terms. But they have all become famous and well recognized by the consuming public." 2 McCarthy on Trademarks and Unfair Competition § 11:83 (5th ed. 2019).

As already discussed, plaintiff's evidence convincingly proves that VAGISIL has attained considerable commercial strength and that consumers associate the VAGISIL brand with plaintiff's products. Thus, assuming arguendo that the conceptual strength of the VAGISIL mark is weakened by the existence of similar third-party marks, see supra part l.A, the powerful evidence of VAGISIL's commercial strength is clearly sufficient here to demonstrate that VAGISIL is a strong, distinctive mark overall. Accordingly, the first likelihood of confusion factor, namely the strength of the senior mark, weighs heavily in favor of plaintiff.

2. The Similarity of the Two Marks

The second factor focuses on "whether there exists a similarity in sight, sound, and meaning which would result in confusion" between the two marks at issue. George & Co. , 575 F.3d at 396.²¹ In this respect, the comparison of the two marks requires consideration of the marks as whole words instead of focusing on the similarities or dissimilarities of component parts of the marks. Swatch , 739 F.3d at 159. And importantly, for the reasons already explained, when registrability is the only issue in a case, the actual use or appearance of the marks in the marketplace is irrelevant, and it is the marks as shown in the registration applications which must be considered. See Seacret Spa , 2016 WL 880367, at *3–4. Thus, because both marks are registered as standard character marks, which do not limit the marks to any particular appearance or stylization in commerce, the manner in which the VAGISIL and VAGISAN marks are currently displayed to consumers in the marketplace does not influence the similarity factor in this registrability determination. Id. at *4.

21 Plaintiff argues that less similarity between marks is required to satisfy the second factor in cases where the parties' products are similar. The Fourth Circuit has rejected this argument on the ground that "the similarity of the parties' products is already accounted for under the third factor of the analysis." Swatch , 739 F.3d at 160.

Compared as a whole, the marks VAGISIL and VAGISAN are closely similar in sight and sound. Both marks consist of "Vagis", then a vowel, and then a consonant. Indeed, only two of the seven letters differ between the marks. When read or heard, the words could easily be confused with one another, especially because the difference in spelling and pronunciation does not come until the very end of the words. Thus, although the marks are not identical, because VAGISIL and VAGISAN "are perceived similarly by the eye and ear," the marks are sufficiently similar in sight and sound to confuse consumers.²² Sara Lee , 81 F.3d at 465 ; see also Washington Speakers Bureau, Inc. v. Leading Authorities, Inc. , 33 F. Supp. 2d 488, 497 (E.D. Va. 1999), aff'd , 217 F.3d 843 (4th Cir. 2000) ("[A]bsolute identity is not necessary for infringement; all that is necessary is enough similarity between the marks to confuse consumers.").

22 Indeed, this conclusion is further supported by the empirical evidence submitted by plaintiff in the form of the Confusion Survey, see infra part 7. In particular, the Confusion Survey shows that a net rate of 19% of the respondents who took the survey confused VAGISAN with VAGISIL specifically because of the similarity of the marks.

The meaning evoked by the two marks is also similar. Both VAGISIL and VAGISAN connote the vaginal area. See supra part 1.A. One of defendant's executives, Ms. Thevessen, testified at trial that the "-san" suffix of VAGISAN also refers to the Latin term for "health." But she admitted that she did not know whether consumers would also perceive VAGISAN to connote "health" because defendant has not conducted any consumer studies on the issue. In any event, it is doubtful whether the additional intended connotation of VAGISAN, i.e. health, would be recognized by consumers given the average consumer's lack of Latin fluency and in light of what is clearly the dominant connotation of both marks: the vaginal area of a woman's body. Thus, the marks share a highly similar meaning.

Accordingly, the close similarities in sight, sound, and meaning between the marks point persuasively to the conclusion that VAGISAN is sufficiently similar to VAGISIL to confuse consumers.

Defendant attempts to counter this conclusion by arguing that the "vagi-" prefix that is common to both VAGISIL and VAGISAN is weak, and thus consumers will distinguish the terms based on minor differences in the non-common components of the marks. Even assuming arguendo that the "vagi-" prefix is descriptive and thus conceptually weak, this argument fails for several reasons.

First, this argument conflicts with the "anti-dissection" rule that the Fourth Circuit has adopted with respect to analyzing the similarity of marks. As noted above, the "anti-dissection" rule "requires consideration of the marks as a whole, rather than the component parts of the marks." Sweetwater Brewing Co., LLC v. Great Am. Restaurants, Inc. , 266 F. Supp. 2d 457, 462 (E.D. Va. 2003) ; see Swatch , 739 F.3d at 159. Indeed, even in cases that involve so-called "descriptive prefix" marks, courts have confirmed that it is improper to focus on the marks' non-common components when evaluating the marks' similarity. See e.g. , Water Pik, Inc. v. Med-Sys., Inc. , 726 F.3d 1136, 1155–56 (10th Cir. 2013) ("We therefore compare the full words ‘SinuCleanse ’ and ‘SinuSense,’ not just the components ‘Cleanse ’ and ‘Sense.’ ").

Second, courts that have held that consumers were likely to differentiate between marks based on minor differences reached that result based on the existence in the marketplace of multiple marks including the same descriptive component for similar goods. See, e.g. , Jack Wolfskin Ausrustung Fur Draussen GmbH & Co. KGAA v. New Millennium Sports, S.L. U. , 797 F.3d 1363, 1374 (Fed. Cir. 2015) (holding that evidence of third-party use of similar marks on similar goods "can show that customers have been educated to distinguish between different marks on the basis of minute distinctions."); In re Nat'l Data Corp. , 753 F.2d 1056, 1060 (Fed. Cir. 1985) ("Where consumers are faced with various usages of descriptive words, our experience tells us that we and other consumers distinguish between these usages."). These cases are distinguishable because, as explained supra part 1.B, defendant has presented no evidence that any other mark including a "vagi-" prefix is extensively used in commerce in connection with vaginal care products. To the contrary, the evidence makes clear that VAGISIL is the only "vagi-" prefix mark that is currently used in United States commerce in connection with the sale of vaginal care products of which most consumers are aware. Thus, because consumers are not faced with multiple "vagi-" prefix marks in the marketplace, there is no basis here on which to conclude that consumers are conditioned to distinguish between VAGISIL and VAGISAN based on minor differences in the non-common portions of the marks.²³

23 Again, this conclusion is reinforced by the empirical evidence supplied by the Confusion Survey, see infra part 7. Specifically, the Confusion Survey shows that third-party marks have not conditioned consumers to distinguish between "vagi-" prefix marks that are as similar as VAGISIL and VAGISAN because a net rate of 19% of the respondents who took the survey confused VAGISAN with VAGISIL, notwithstanding the respondents' real-world exposure to other products sold in the marketplace.

Third, the other cases cited by defendant that concluded there was no likelihood of confusion between marks containing the same so-called "descriptive prefix" are inapposite here because the marks at issue in those cases contained non-common components that were sufficient to make the marks as a whole dissimilar.²⁴ Here, for the reasons already stated, when compared as a whole, VAGISAN is confusingly similar to VAGISIL. Thus, unlike in the cases cited by defendant, the slight differences between the parties' marks here—the last two letters and a dubious reference to the Latin word sanitas —are insufficient to make the marks dissimilar overall.

24 See, e.g. , Water Pik , 726 F.3d at 1155–56 (holding that "SinuSense" and "SinuCleanse" were not similar when compared as a whole); Land-O-Nod , 220 U.S.P.Q. at 66–67 (holding that "applicant's addition of the term ‘MATIC’ to the term ‘CHIRO’ is sufficient to render the resulting mark ‘CHIRO-MATIC,’ considered as a whole , distinguishable from and registrable over opposer's marks ‘-CHIRO-’ and ‘CHIROPRACTIC’ ") (emphasis added); Alltel Corp. v. Actel Integrated Commc'ns, Inc. , 42 F. Supp. 2d 1265, 1271 (S.D. Ala. 1999) (holding that ACTEL and ALLTEL were dissimilar as a whole based on the clearly different meanings invoked by the terms).

In sum, when compared as a whole, VAGISAN is confusingly similar to VAGISIL, and defendant has not shown that consumers have been educated to distinguish between the marks based on minor differences. Accordingly, the second factor in the likelihood of confusion analysis weighs in favor of plaintiff.

3. The Similarity of the Goods or Services that the Marks Identify

The third factor focuses on the similarity of the goods sold in connection with the parties' marks. With respect to this factor, the Fourth Circuit has explained that "the goods in question need not be identical or in direct competition with each other" to be deemed similar. George & Co. , 575 F.3d at 397. Instead, because "confusion may arise even where products are merely ‘related,’ the court is to consider ‘whether the public is likely to attribute the products and services to a single source.’ " Renaissance Greeting Cards, Inc. v. Dollar Tree Stores, Inc. , 227 F. App'x 239, 244 (4th Cir. 2007) (quoting CAE , 267 F.3d at 679 ). Put another way, the parties' goods are similar if they "are often made by the same manufacturer." Seacret Spa , 2016 WL 880367, at *5. And, once again, because this case involves only the issue of registrability, "it is the mark ... as used on the goods described in the application which must be considered, not the mark as actually used." Id. at *3 (comparing the goods listed in the defendant's application with the goods covered under the plaintiff's mark).

Here, the registrations for VAGISIL and registration application for VAGISAN reflect that certain goods covered by the parties' marks are identical, such as vaginal moisturizers and vaginal washes. In addition, the VAGISAN application lists "cosmetics," which is broad enough to encompass VAGISIL's registration for "cosmetics – namely, powders for feminine use." See In re Hughes Furniture Indus., Inc. , 114 U.S.P.Q.2d 1134, 1137 (TTAB 2015). Accordingly, because several of the goods covered by the parties' marks are identical, the third likelihood of confusion factor, similarity of the goods identified by the marks, weighs in favor of plaintiff.²⁵

25 It is important to note that plaintiff is not required to prove similarity as to every good listed in VAGISAN's registration application and VAGISIL's registrations. Instead, the similarity of goods factor is satisfied if plaintiff shows relatedness between any item encompassed by both marks. See Gen. Mills, Inc. & Gen. Mills Ip Holdings II, LLC v. Fage Dairy Processing Indus. S.A. , 100 U.S.P.Q.2d 1584 (T.T.A.B. 2011) ; Apple Computer v. Tvnet.net, Inc. , 90 U.S.P.Q.2d 1393 (T.T.A.B. 2007) (citing Tuxedo Monopoly Inc. v. General Mills Fun Group , 648 F.2d 1335, 209 U.S.P.Q. 986, 988 (CCPA 1981) ).

4. The Similarity of the Facilities Used by the Markholders

The fourth factor, similarity of the parties' facilities, examines "how and to whom the respective goods of the parties are sold," that is, "whether ‘both products [are] sold in the same ‘channels of trade.’ ’ " Rosetta Stone Ltd. v. Google, Inc. , 676 F.3d 144, 155 (4th Cir. 2012) (quoting 4 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 24:51 ). In the context of a registration appeal, "the similarities in channels of trade must be analyzed based on the channels of trade contemplated by the application." Seacret Spa , 2016 WL 880367, at *5. If a registration application "does not delimit any specific trade channels of distribution, no limitation will be applied." Id. Here, neither plaintiffs VAGISIL registrations nor defendant's VAGISAN application define or restrict the channels of trade in which the marks will be used in connection with the distribution of goods. Accordingly, it must be presumed that the parties' marks cover all channels of trade and that the parties' facilities are thus identical. Id. at *6. The fourth factor therefore weighs in favor of plaintiff.

5. The Similarity of the Advertising Used by the Markholders

On a similar note, the fifth factor, similarity of the parties' advertising, focuses on whether the parties "both employ similar media and target the same consumers" in connection with promoting their marks and marketing their products. Sara Lee , 81 F.3d at 466. Again, because "a registration proceeding deals with the mark as described and limited in the application," analysis of this factor is based on the usages described within VAGISIL's registrations and VAGISAN's registration application. Seacret Spa , 2016 WL 880367, at *6. Accordingly, because the parties' marks are not delimited to any particular trade channels or advertising media in their registrations, no limitations will be implied. Id. at *6. The parties' advertising models are thus presumed to employ similar media and target the same consumers, which causes the fifth factor to weigh in favor of plaintiff.

6. Defendant's Intent

The sixth likelihood of confusion factor focuses on whether "the junior user intended to capitalize on the good will associated with the senior user's mark." CareFirst , 434 F.3d at 273. The defendant's intent is sometimes "a major factor" in infringement cases because a defendant who "inten[ds] to confuse the buying public ... generally attempts to make his signs, advertisements, etc., to resemble the other's so as deliberately to induce confusion." Pizzeria Uno , 747 F.2d at 1535. Yet, importantly, even in the trademark infringement context, the converse is not true, as the Fourth Circuit has held that a "good faith belief that a subsequently-adopted mark will not lead to confusion ... is no defense if a court finds actual or likelihood of confusion." Id. Indeed, another district court in this district has observed that "while this factor can be extremely probative in infringement cases, it is rarely relevant to the likelihood of confusion in registration proceedings." Seacret Spa , 2016 WL 880367, at *6.

Here, plaintiff does not contend that defendant applied to register the VAGISAN mark with the intent to capitalize on the goodwill associated with plaintiff's VAGISIL mark. To the contrary, the evidence reflects that defendant has registered and used the VAGISAN mark since 1998 in other countries, including Germany, the United Kingdom, the Netherlands, and Poland. In those countries, defendant has displayed the VAGISAN mark on packaging that does not appear to be designed to resemble the trade dress used in connection with VAGISIL products.²⁶ In any event, although this evidence supports a finding that defendant did not apply to register the VAGISAN mark with the intent to confuse consumers, the factor is accorded only minimal weight. See Pizzeria Uno , 747 F.2d at 1535 (holding that a "good faith belief that a subsequently-adopted mark will not lead to confusion ... is no defense if a court finds actual or likelihood of confusion"); Seacret Spa , 2016 WL 880367, at *6 (observing that the intent factor "is rarely relevant to the likelihood of confusion in registration proceedings").

26 Plaintiff correctly argues that evidence concerning foreign trademark uses and registrations are irrelevant and inadmissible for purposes of determining whether a likelihood of confusion exists in the United States. See Fuji Photo Film Co. v. Shinohara Shoji Kabushiki Kaisha , 754 F.2d 591, 599 (5th Cir. 1985) (collecting cases) (holding that the district court erred in admitting "evidence of the parties' foreign trademark usage and occurrences" because "[t]he concept of territoriality is basic to trademark law; trademark rights exist in each country solely according to that country's statutory scheme"). Yet, in the related context of the Anti–Cybersquatting Consumer Protection Act, 15 U.S.C. § 1125(d), courts have concluded that evidence that the defendant owns trademark rights for the mark at issue in several countries can support a finding that the defendant did not register a domain name that is similar to the plaintiff's mark with a bad faith intent to profit from the plaintiffs protected mark. See, e.g. , Valador, Inc. v. HTC Corp. , 241 F. Supp. 3d 650, 673 (E.D. Va. 2017), aff'd , 707 F. App'x 138 (4th Cir. 2017). Thus, evidence concerning defendant's registration and use of the VAGISAN mark outside of the United States is admitted in this case solely for the purpose of demonstrating defendant's good faith intent in applying to register the VAGISAN mark in the United States and is excluded in all other respects.

7. Actual Confusion

The seventh factor in the likelihood of confusion analysis is evidence of actual confusion between the senior and junior marks. Like the first factor, the strength of the senior mark, actual confusion is an important factor that weighs heavily in the overall likelihood of confusion analysis. See Sara Lee , 81 F.3d at 467 ("If the strength of the senior mark is the alpha of infringement analysis, then evidence of actual confusion is surely the omega; where the defendant in an infringement case has elected to use a mark similar to that of a competitor's distinctive mark, and, as a result, has actually confused the public, our inquiry ends almost as soon as it begins."). The Fourth Circuit has explained that "[a]ctual confusion can be demonstrated by both anecdotal and survey evidence." George & Co. , 575 F.3d at 398.

Here, plaintiff did not present any anecdotal evidence, such as reports by United States consumers expressing confusion between the VAGISIL and VAGISAN marks. It is true that "the absence of any evidence of actual confusion over a substantial period of time ... creates a strong inference that there is no likelihood of confusion." CareFirst , 434 F.3d at 269. But importantly, "[e]vidence of the number of instances of actual confusion must be placed against the background of the number of opportunities for confusion before one can make an informed decision as to the weight to be given the evidence." George & Co. , 575 F.3d at 398 (quoting 4 MCCARTHY ON TRADEMARKS AND UNFAIR COMPETITION § 23:14 ). Accordingly, because defendant has not begun to use its VAGISAN mark in United States commerce, a lack of anecdotal evidence of confusion is unsurprising and is thus not meaningful in this case.²⁷ See Hasbro, Inc. v. Mattel, Inc. , 2002 WL 834489, at *6 (T.T.A.B. Apr. 30, 2002) (non-precedential) ("[T]he lack of [actual confusion] evidence is not meaningful in this case because applicant has not yet begun to use its mark, and therefore we would not expect to find evidence of actual confusion.").

27 Seeking to avoid this conclusion, defendant argues that the absence of anecdotal evidence is meaningful based on evidence that a "gray market" version of defendant's VAGISAN products has been available for purchase on Amazon.com without defendant's authorization since October 2016. This argument fails because defendant introduced no evidence that any consumer has ever seen or purchased this "gray market" VAGISAN product, other than one purchase made by defendant's counsel.

In the absence of anecdotal evidence, plaintiff submitted a survey conducted by plaintiff's expert, Mr. Hal Poret, designed to test whether there is a likelihood of confusion between the VAGISIL and VAGISAN marks. See supra , Findings of Fact ¶¶ 33–37. Mr. Poret administered the Confusion Survey to four hundred female consumers of vaginal moisturizers, vaginal washes, or vaginal anti-fungal products. The Confusion Survey utilized the well-established Eveready ²⁸ methodology²⁹ and first displayed the VAGISAN mark to the respondents who were in the test group and then asked those respondents a series of standard questions to measure various forms of confusion between VAGISAN and other brand names.³⁰ After each of these questions, the survey gave the respondents the opportunity to explain why they selected their answer. The Confusion Survey never mentioned or showed the VAGISIL mark to the survey respondents. Importantly, the test group respondents were shown the VAGISAN mark as the mark appears on defendant's application to register the mark, i.e. in a plain, block letter format, without any trade dress, stylization, special font, colors, company name, packaging, or any other distinctive elements. After being presented with the VAGISAN mark, 37% of the respondents in the test group named VAGISIL in response to the confusion questions.

28 See Union Carbide Corp. v. Ever–Ready, Inc. , 531 F.2d 366 (7th Cir. 1976), cert. denied , 429 U.S. 830, 97 S.Ct. 91, 50 L.Ed.2d 94 (1976) (approving what is now known as the "Eveready " test).

29 The only exception to this statement is that the survey respondents in the test group of the Confusion Survey were not shown the VAGISAN mark as it appears in the marketplace. See Valador, Inc. v. HTC Corp. , 242 F. Supp. 3d 448, 464 (E.D. Va. 2017) (citing 6 McCarthy on Trademarks § 32:173 (4th ed. 2014) ), aff'd , 707 F. App'x 138 (4th Cir. 2017) (noting that under the Eveready methodology, "the respondent is shown the entire mark as used in the marketplace"). But as explained below, it was appropriate for Mr. Poret to diverge from the typical Eveready format in this respect to obtain an accurate measurement of whether the VAGISAN mark, as delimited in the application, is likely to cause confusion, which is the key issue in this registrability action.

30 For example, respondents were asked (i) what company or brand they think puts out the products they were just shown, (ii) whether they think the company that makes the products they were just shown makes any other products that they know of (and what the names of those other products are), and (iii) whether they think the products they were just shown are affiliated with, sponsored by, or approved by any other company or brand they know of (and what the names of those other companies or brands are).

The Confusion Survey was also administered to respondents in a control group, who were shown the fictitious name VAGIPUR instead of VAGISAN. VAGIPUR was shown to the control group respondents in a plain, block letter format, without any trade dress, stylization, special font, colors, company name, packaging, or any other distinctive elements. In the control group, 18% of the respondents named VAGISIL after they were shown the VAGIPUR mark and were asked the confusion questions. Accordingly, the Confusion Survey reported a net confusion rate of 19% between VAGISAN and VAGISIL; i.e. , it reflected that 19% of respondents experienced confusion that was caused specifically by the marks' similarity and not by other factors.

The Confusion Survey provides persuasive empirical evidence of actual confusion between the parties' marks. First, the Eveready survey format used by the Confusion Survey has been widely accepted by courts and commentators as a reliable approach for surveying consumer confusion.³¹ The reliability of the Eveready format used in the Confusion Survey stems, in part, from the fact that the survey displays only the allegedly infringing mark and at no point shows or mentions the senior mark.³² This feature is particularly important where, as here, the plaintiff's senior mark is strong and widely-recognized. See Valador , 242 F. Supp. 3d at 464.

31 See, e.g. , Valador, Inc. v. HTC Corp. , 242 F. Supp. 3d 448, 456 n.8 (E.D. Va. 2017), aff'd , 707 F. App'x 138 (4th Cir. 2017) ("noting that "the Eveready method is one of the standard, accepted approaches for surveying likelihood of confusion"); E & J Gallo Winery v. Proximo Spirits, Inc. , No. 1:10-CV-00411 LJO, 2011 WL 5922090, at *3 (E.D. Cal. Nov. 28, 2011) ("[T]he parties agree that the Eveready method or the ‘standard survey format’ is the ‘gold standard’ for determining whether there is a likelihood of confusion."); U.S. Polo Ass'n, Inc. v. PRL USA Holdings, Inc. , 800 F. Supp. 2d 515, 535 (S.D.N.Y. 2011), aff'd , 511 F. App'x 81 (2d Cir. 2013).

32 See Kargo Glob., Inc. v. Advance Magazine Publishers, Inc. , No. 06 CIV 550 JFK, 2007 WL 2258688, at *8 (S.D.N.Y. Aug. 6, 2007) (noting that the Eveready format's display of a single mark "is much more reliable" because it "ensur[es] that respondents are not made artificially aware of the other party's trademark") (internal quotations marks omitted).

Second, the 19% net confusion rate reported by the Confusion Survey exceeds survey results that courts have found to be strong evidence of actual confusion. See Mutual of Omaha Ins. Co. v. Novak , 836 F.2d 397, 400 (8th Cir. 1987), cert. denied , 488 U.S. 933, 109 S.Ct. 326, 102 L.Ed.2d 344 (1988) (giving significant weight to survey evidence showing confusion level of between ten and eleven percent); Henri's Food Products Co., Inc. v. Kraft, Inc. , 717 F.2d 352, 358 (7th Cir. 1983) (collecting cases holding that survey evidence indicating ten to twelve percent confusion was sufficient to demonstrate actual confusion); Quality Inns Int'l, Inc. v. McDonald's Corp. , 695 F. Supp. 198, 219 (D. Md. 1988) (holding that a 16.3% confusion rate demonstrated that "an appreciable number of consumers are likely to be confused"); Miles Laboratories Inc. v. Naturally Vitamin Supplements Inc. , 1986 WL 83319, at *12 (T.T.A.B. March 31, 2011) ("[S]urveys disclosing likelihood of confusion ranging from 11% to 25% have been found significant."). Accordingly, in light of the reliable methodology utilized by the survey and the substantial level of confusion between VAGISIL and VAGISAN reported by the survey, the Confusion Survey is persuasive and reliable evidence of actual confusion between the parties' marks.

Seeking to avoid this conclusion, defendant advances three criticisms aimed at discrediting the reliability of the Confusion Survey. First, defendant argues that the Confusion Survey is entitled to little weight because the survey did not replicate the marketplace conditions under which consumers are most likely to encounter the VAGISAN mark. In particular, defendant's expert, Dr. Simonson, testified that in order to obtain an accurate measurement of consumer confusion, the survey needed to display the VAGISAN mark (i) with the packaging used by defendant in other countries and (ii) side-by-side with the VAGISIL mark, as the products might appear if they were sold on the same shelf of a store.

This criticism is unpersuasive because it ignores the principle, already discussed extensively, that "the relevant inquiry in a registration proceeding involves the mark and usage described in the application rather than as they appear in the marketplace." Seacret Spa , 2016 WL 880367, at *2 (holding that the probative value of an expert's testimony was "severely limited" because the testimony was based on a study of the use of the "SEACRET" mark in the marketplace rather than the uses described in the application). To be sure, it is true that in the context of a trademark infringement claim, a likelihood of confusion survey must replicate market conditions. See, e.g. , Water Pik , 726 F.3d at 1146–47 ; Valador , 242 F. Supp. 3d at 462. But this is so "for the simple reason that trademark infringement occurs if ‘the defendant's actual practice is likely to produce confusion in the minds of consumers about the origin of the goods or services in question." Valador , 242 F. Supp. 3d at 462 (emphasis in original) (quoting CareFirst , 434 F.3d at 267 ). In contrast, in the instant § 1071(b) action, the only issue is whether the mark for which defendant applied is registrable, not whether any particular use of defendant's mark constitutes infringement.³³ Accordingly, as the TTAB has repeatedly and correctly concluded, the only scientifically sound format for a survey designed to test the likelihood confusion in a registrability proceeding is to display the applicant's mark as it appears in the registration application. See OMS Invs., Inc. v. Cent. Garden & Pet Co. , 2006 WL 2066583, at *14 (T.T.A.B. July 10, 2006) ; Miles Labs , 1986 WL 83319, at *15 (TTAB 1986) ; McDonough Power Equip., Inc. v. Weed Eater, Inc. , 208 U.S.P.Q. 676, 685 (T.T.A.B. 1981).

33 Furthermore, it would be arbitrary to select as a stimulus the packaging used by defendant for VAGISAN products sold in Europe, as proposed by Dr. Simonson. One of defendant's executives, Ms. Thevessen, testified at trial that defendant has not made a final determination of how the VAGISAN mark will be used or how the VAGISAN mark will appear on packaging on products should they be permitted to be sold in the United States. More importantly, if defendant's application is granted, defendant will be free to use the mark and display it on packaging differently from how defendant does so elsewhere because defendant has applied to register the VAGISAN mark as a standard character mark, without any restriction on stylization or usage. See Citigroup , 637 F.3d at 1353 ; Kimberly-Clark , 774 F.2d at 1147.

The second criticism argued by defendant is that the Confusion Survey used an improper control. The parties agree that an effective control serves the essential function of estimating the rate of survey responses appearing to reflect confusion that are in fact explained by survey "noise," i. e. some factor other than genuine confusion. Of particular importance in this case, which involved the well-known mark VAGISIL, the control needed to measure the survey's "demand effect," which relates to the number of survey respondents who would guess VAGISIL in response to the survey stimulus because they believe it is the expected answer. The parties also agree that to be effective, the selected control needed to be as similar to the test stimulus, VAGISAN, as possible but could not share the characteristic whose influence is being tested. Dr. Simonson, defendant's expert, opined at trial that the control used by Mr. Poret, VAGIPUR, violated these principles and was thus an improper control because it was not sufficiently similar to VAGISAN.

This criticism is entitled to little, if any, weight. Although it is conceivable that a different, more similar control could have been selected, the control used by Mr. Poret was adequate to measure survey "noise." In this case, plaintiff's hypothesis was that VAGISAN is confusingly similar to VAGISIL because both marks (i) begin with "vagis-", (ii) followed by a vowel (iii) and then a consonant. Thus, an effective control needed to be similar to VAGISAN without meeting those three criteria. VAGIPUR satisfied this requirement. Like VAGISAN, it began with "vagi" and thus also connoted the vaginal area of a woman's body. And as Mr. Poret noted, using a control beginning with "vagi" would capture the tendency of VAGISAN's "vagi-" prefix or the topic of vaginal care products to cause people to name VAGISIL. In addition, VAGIPUR held constant the length of VAGISAN and also included a consonant, then a vowel, and then a consonant after "vagi." Thus, the VAGIPUR control maintained a proper degree of similarity to the test stimulus without overlapping with the three criteria constituting the hypothesis tested by the survey.³⁴

34 Mr. Poret testified that he also intended the "-pur" suffix of the VAGIPUR control to invoke the concept of "purity," which would be similar to the reference to "health" that defendant argued is invoked by the "-san" suffix of VAGISAN. In response, Dr. Simonson opined that the "-pur" suffix sounded foreign and appeared to be a reference to cities in other countries that end with "-pur." Neither suggestion is persuasive because neither party presented evidence to show what meaning, if any, consumers would attribute to the "-pur" suffix, and the connotations proposed by Dr. Simonson and Mr. Poret are far from self-evident. In any event, it was not important for the control to include this additional similarity to VAGISAN because, as previously noted supra part 2, it is doubtful that consumers detect VAGISAN's alleged reference to the Latin word for health.

The results of the Confusion Survey further support the conclusion that the survey used an effective control. Mr. Poret testified that based on his substantial experience conducting Eveready studies, the 18% "noise" level captured by VAGIPUR was certainly robust. Furthermore, because defendant did not conduct its own survey using a different control, there is no empirical basis on which to conclude whether or to what degree a different control would have captured a materially higher level of survey noise. Accordingly, the evidence at trial convincingly establishes that the Confusion Survey used an adequate, even if imperfect, control and that the 19% net rate of confusion reliably reported the confusion caused by the similarity of VAGISAN to VAGISIL and not by other factors.

Defendant's third criticism is that plaintiff's expert, Mr. Poret, improperly tabulated certain survey respondents as confused when calculating the gross and net rates of confusion reported by the Confusion Survey. In this respect, defendant argues that a survey respondent in the test group who answered VAGISIL in response to the three Eveready questions designed to test confusion should not have been counted as confused if the respondent's answer to the follow-up "why?" question reflects that the respondent answered VAGISIL for reasons other than the similarity of VAGISAN. This argument fails because, as both parties' experts (Mr. Poret and Dr. Simonson) testified, an effective control already serves the function of estimating the number of respondents that will name VAGISIL for reasons other than the similarity between VAGISAN and VAGISIL. Thus, sifting through the respondent's open-ended responses, in addition to using a control, would improperly double-count survey "noise."

Moreover, as Mr. Poret persuasively explained at trial, the methodology advocated by defendant is scientifically inferior to using an effective control. Survey respondents are lay people who do not always know the actual reason that VAGISIL came to their mind when they saw the survey stimulus, and they often give explanations that are difficult to interpret objectively. Measuring the rate of confusion among survey respondents in the manner proposed by defendant is thus unreliable and calls for subjective judgment by the survey-administrator as to whether the respondents were truly confused.³⁵

35 At trial, Mr. Poret recalculated the test group's net confusion rate by examining the respondents' explanations for answering VAGISIL instead of subtracting the control group's result. Tellingly, this methodology yielded an even higher net rate of confusion of 21.5%. This exercise persuasively reinforces the notion that calculating survey "noise" based on the survey respondents' explanations is inferior to using an effective control.

As already discussed, the Confusion Survey used an effective control term to account for survey "noise." Accordingly, because all four hundred respondents were asked identical questions after the test group was shown VAGISAN and the control group was shown VAGIPUR, the difference between the two groups' rate of naming VAGISIL must be caused by the similarity of VAGISAN to VAGISIL and not by other reasons. Put differently, by using an effective control, the Confusion Survey scientifically and objectively estimated the number of test group responses that were given for reasons other than genuine confusion. It is therefore unnecessary and unreliable to attempt to measure survey "noise" by evaluating the respondents' own suggestions of why they answered as they did.

In sum, the Confusion Survey submitted by plaintiff is reliable and is powerful evidence of actual confusion. Accordingly, the seventh factor weighs heavily in plaintiff's favor.

8. The Quality of Defendant's Product

The Fourth Circuit has explained that the eighth factor, the quality of defendant's products, is rarely applicable. Sara Lee , 81 F.3d at 467. Specifically, "[c]onsideration of the quality of the defendant's product is most appropriate in situations involving the production of cheap copies or knockoffs of a competitor's trademark-protected goods." Id. This is so because "[i]f a defendant markets a product under a mark similar to that affixed by a competitor to a commodity of like nature but superior manufacture, that the defendant's product is markedly inferior is likely to be highly probative of its reliance on the similarity of the two marks to generate undeserved sales." Id. Here, neither party presented evidence or argument on this factor, and there is no basis on which to conclude that defendant's products are "markedly inferior" and rely heavily on similarity to plaintiff's VAGISIL products "to generate undeserved sales." Thus, this factor is inapplicable in the instant case.

9. The Sophistication of the Consuming Public

Like factor eight, the Fourth Circuit has also held that the ninth factor, the sophistication of the consuming public, is often inapplicable. Id. As the Fourth Circuit explained in Sara Lee , "[b]arring an unusual case, buyer sophistication will only be a key factor when the relevant market is not the public at-large." Id. In particular, this factor is only relevant to the likelihood of confusion analysis if the typical consumer in the relevant market "possesses an expertise" regarding the product at issue or is "more sophisticated about that product than those who comprise the market for other ordinary retail goods." Id. (citing Perini , 915 F.2d at 127–28 ). But if a mark "targets both highly knowledgeable and average consumers, ‘the issue of likelihood of confusion will usually revolve around the less knowledgeable consumer.’ " Seacret Spa , 2016 WL 880367, at *6 (quoting 4 MCCARTHY, TRADEMARKS AND UNFAIR COMPETITION § 23:100 (4th ed. 2014) ).

Here, the evidence does not show persuasively that the typical consumer of vaginal care products sold in connection with the parties' marks is any more or less sophisticated than the typical consumer of ordinary retail goods. On the one hand, the evidence reflects that some consumers are embarrassed about vaginal dryness and other conditions treated by the parties' products. These consumers do not want to talk about the conditions treated by the parties' products, not even with physicians, and do not want to be seen buying the products that treat those conditions. As a result, such consumers often purchase the parties' products—which are sold over the counter for as little as $ 3.00—hastily without spending much time at the shelf to examine the products carefully. On the other hand, the evidence also shows that some consumers spend time prior to purchase carefully researching the parties' products and the conditions that those products are designed to treat. For example, plaintiff and other companies provide resources that educate women in a confidential setting about vaginal dryness and how the parties' products help to treat this condition.

Viewed in its entirety, this evidence reflects that consumers of the parties' products vary widely in their sophistication about the parties' products and the degree of care with which consumers purchase such products, as is characteristic of the market for most ordinary retail goods. Accordingly, because this factor does not clearly favor either party, and because it appears that the parties' products target the public at-large, the ninth likelihood of confusion factor is not entitled to significant weight in this case.

10. Conclusion

For the reasons stated above, application of the likelihood of confusion factors to the evidence submitted at trial convincingly supports a finding that the VAGISAN mark that defendant applied to register "so resembles a mark registered in the Patent and Trademark Office," to wit: plaintiff's VAGISIL mark, "as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive." See 15 U.S.C. § 1052(d). In particular, the considerable overall strength of the VAGISIL mark, the similarity between VAGISIL and VAGISAN, the similarity of the goods identified by the marks, the similarity of the parties' facilities, the similarity of the parties' advertising, and evidence of actual confusion between VAGISIL and VAGISAN weigh heavily in favor of a determination that confusion would likely result if VAGISAN were registered. The remaining factors—defendant's good faith intent, the quality of defendant's products, and consumer sophistication—are not entitled to significant weight in this case.

Therefore, pursuant to 15 U.S.C. § 1071(b)(1), the evidence compels the conclusion that defendant's VAGISAN mark is not registrable and that the ruling of the TTAB dismissing plaintiff's opposition to defendant's application to register the VAGISAN mark must be reversed.

Accordingly, judgment for plaintiff in this matter is granted.

An appropriate judgment will enter.

The Clerk is directed to send a copy of these Findings of Fact and Conclusions of Law to all counsel of record.