OPINION AND ORDER

DENISE COTE, District Judge.

In this lawsuit, a book publisher alleges that the prominent technology company Apple Inc. ("Apple") has infringed its trademark and created a likelihood of reverse confusion, in that consumers will likely believe that its books are in fact published by or affiliated with Apple. The plaintiffs J.T. Colby & Company, Inc. d/b/a Brick Tower Press, J. Boylston & Company, Publishers LLC and iPicturebooks LLC, (collectively "J.T. Colby & Company") employ the unregistered trademark ibooks and bring this trademark infringement action against Apple in connection with Apple's use of the mark iBooks to designate Apple's e-reader software. Apple moves for summary judgment on all of the plaintiffs' claims. The defendant's motion is granted. The plaintiffs have failed to present sufficient evidence that their ibooks mark is entitled to trademark protection or that their mark is likely to suffer from reverse confusion with Apple's iBooks mark.

BACKGROUND

The following facts are either undisputed or taken in the light most favorable to the plaintiffs, unless otherwise indicated.

I. 1999: Launch of ibooks Imprint

Two publishing companies created by Byron Preiss ("Preiss") — ibooks, inc. and Byron Preiss Visual Publications ("BPVP") — launched the ibooks imprint in 1999. As explained to the public, the imprint was intended to take advantage of opportunities in publishing provided by the Internet. In May 1999, Publishers Weekly ran an article describing the forthcoming imprint:

In August of 1999, Preiss put out a press release introducing the new imprint. In the press release, Preiss explained that

Preiss filed two applications with the United States Patent and Trademark Office ("PTO") in 1999 (the "1999 Applications"). In one, Preiss sought to register the mark IBOOKS in connection with the sale of "Books" for trademark protection (the "IBOOKS Application"). In the other application, Preiss sought to register the mark "IBOOKSINC.COM" in connection with an "Internet website providing information and text about printed publications and for purchasing printed publications." The PTO no longer has complete records pertaining to the 1999 Applications. Based on the records that do exist, however, it appears that the IBOOKS Application was initially rejected on two grounds: (1) two other trademark registrants owned similar marks,1 and (2) the mark was considered deceptively misdescriptive. In 2002, Preiss submitted a response to the PTO's refusal to register the IBOOKS mark. Preiss explained that "Applicant's mark is for `books, namely, a series of fiction books; non-fiction books in the field of science.'" Preiss also argued that the IBOOKS mark was not "misdescriptive," denying that the mark would suggest a connection to the Internet. Preiss stated that "consumers, when seeing the mark on the books, will not think it is an electronic book found on the Internet."

The PTO also refused to register the IBOOKSINC.COM mark. The application was denied on three grounds: (1) two other trademark registrants owned similar marks; (2) the mark was merely descriptive; and (3) the mark was deceptively misdescriptive.2 In his response to the PTO's Office Action, Preiss explained that "Applicant's mark is for `Computerized on-line ordering service in the field of printed publications' and `Providing a website on global computer networks featuring information in the field of printed publications.'" The response further stated that "Applicant's services are equivalent to an electronic retail store. [T]his is clearly different from books themselves and from creating books, which is, in essence, equivalent to publishing." Preiss also argued that the mark was not "misdescriptive," stating:

The 1999 Applications were abandoned.

From 1999 to 2005, Preiss and his companies used the ibooks imprint in connection with the publication of works in the science-fiction, horror, and fantasy genres as well as graphic novels. According to the plaintiffs, and as illustrated by samples of physical books submitted in connection with this motion practice by the defendant, the ibooks mark appears on the spine of a book, the back cover, and on one or several of the first inside pages of a book. With respect to their ebooks, the owner of the plaintiffs, John Colby ("Colby") has explained that the ibooks mark appears on an inside page of ebooks since ebooks do not have spines and generally do not have a back cover.

Whether it is a physical book or an ebook, in almost all instances, the ibooks mark appears as a component of a larger mark. The word ibooks appears underneath an image of a light bulb with the letter "i" inside it.3 According to Colby, prior to 2011, plaintiffs' ibooks mark was "exclusively" depicted in all lowercase letters. The following image is the most common depiction of plaintiffs' mark:

On some physical books, the light bulb logo appears without the word ibooks.

Ibooks, inc. also owned two domain names that used the word "ibooks" as a component of web addresses: www.ibooksinc.com and www.ibooks.net. These domain names appeared on the copyright pages of at least some of the books published by ibooks, inc. between 1999 and 2006. Precisely how these websites were used or what content they hosted is undisclosed by the record. At some point in time before this action was filed both websites became inactive.

Preiss operated ibooks, inc. from 1999 through 2005. During this time, Colby had only limited contact with Preiss and his companies. In their only business transaction, Preiss purchased the rights to one of the titles published under Colby's Brick Tower imprint. On July 9, 2005, Preiss died in a car accident. His companies fell on hard times and on February 22, 2006, ibooks, inc. and BPVP filed for Chapter 7 bankruptcy liquidation.

II. 2006: Colby's Acquisition of the ibooks Imprint

The plaintiffs purchased the assets of ibooks, inc. and BPVP on December 13, 2006 for $125,000. The assets included all publishing rights, copyrights, trademarks, rights and licenses to software programs, computer hardware, manuscripts, and over 300,000 books. Through this transaction, the plaintiffs acquired the ibooks imprint. As Colby has explained, the $125,000 purchase price reflected his "estimate of the fair market value of the sales potential of the existing inventory in the next two years out."4

Since acquiring Preiss' companies, the plaintiffs have continued to publish books under the ibooks imprint. The plaintiffs sell both physical books and ebooks. A spreadsheet that Colby compiled and produced in discovery indicates that 98.17% of the books sold under the ibooks and ipicturebooks imprints from 1999 through 2012 have been physical books, while 1.83% have been ebooks. In the years Colby has owned the ibooks imprint, the sales of products bearing the ibooks mark have been modest. During this time period, annual net sales were as high as $118,049, and as low as negative $28,876.

III. 2010: Apple adopts the iBooks Mark

Apple is a technology company offering computer hardware products, including the iPhone, iPad, and iPod, and computer software products such as iTunes and iBooks. By the end of 2009, Apple was developing an e-reader software program. By January 2010, Apple was actively considering "iBooks"5 for the name of its new product. Apple employed counsel to conduct a trademark search in connection with Apple's clearance process for finalizing the selection of the iBooks mark. Apple's search, the adequacy of which is disputed by the plaintiffs, did not uncover plaintiffs' existence or any current use of ibooks by the plaintiffs.

Around this time, Apple was aware that the mark IBOOK was being used by a company called Family Systems Ltd. ("Family Systems").6 Family Systems owned a trademark registration for "IBOOK" with a priority date of October 8, 1996, in connection with "computer hardware and software used to support and create interactive, user-modifiable electronic books." On January 13, 2010, Apple contacted Family Systems to negotiate an assignment from Family Systems to Apple of Family System's registration in the IBOOK trademark. On January 27, Apple and Family Systems entered into an Assignment Agreement that provided as follows:

Shortly after executing the assignment, Apple announced that it would be offering the iBooks e-reader software. IV. Colby Contacts Apple

Two days after Apple's announcement, on January 29, 2010, Colby emailed a Public Relations Director at Apple (the "January 29 Email"). The email reads as follows:

(Emphasis supplied.) Around February 1, Colby spoke with Glen Gunderson ("Gunderson") — counsel for Apple — by telephone. Colby recalled that during the conversation he asked Gunderson if Apple would be interested in "the exclusive use of [the plaintiffs'] books on the iPad." According to Colby, Gunderson indicated that he would put Colby in contact with someone at Apple, but Colby never received a follow-up call from anyone at Apple.

Sometime in 2011, the plaintiffs began using a slightly different version of the ibooks imprint on some of the books they publish. In particular, the plaintiffs began capitalizing the letter "B" in their imprint, so that the imprint appears as "iBooks."7

On June 15, 2011, the plaintiffs filed this law suit against Apple. The complaint asserts a claim for false designation of origin in violation of Section 43(a)(i)(A) of the Lanham Act, 15 U.S.C. § 1125(a)(i)(A), as well as state law claims of infringement of common law trademark and unfair competition, wrongful misappropriation, unjust enrichment and conversion. Following the close of discovery,8 the parties served cross-motions for summary judgment on December 21, 2012. The plaintiffs move for partial summary judgment on the ground that the assignment of Family System's IBOOK trademark registration to Apple is an invalid assignment in gross.9 The defendant moves for summary judgment with respect to all of the plaintiffs' claims. The parties' motions were fully submitted on February 5, 2013.

DISCUSSION

Summary judgment may not be granted unless the submissions of the parties taken together "show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c). The moving party bears the burden of demonstrating the absence of a material fact question, and in making this determination the court must view all facts in the light most favorable to the non-moving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247 (1986); Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). When the moving party has asserted facts showing that the non-movant's claims cannot be sustained, the opposing party must "set forth specific facts showing that there is a genuine issue for trial," and cannot rest on mere "allegations or denial" of the movant's pleadings. Fed.R.Civ.P. 56(e); Hicks v. Baines, 593 F.3d 159, 166 (2d Cir. 2010). Nor can a non-movant "rely on mere speculation or conjecture as to the true nature of the facts to overcome a motion for summary judgment." Id.

I. Trademark Infringement

Plaintiffs assert rights in the unregistered mark ibooks when used as an imprint for books that they publish.10 They claim that Apple infringed plaintiffs' mark when it adopted the iBooks mark in connection with e-reader software.

Pursuant to Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a),

Section 43(a) protects both registered and unregistered trademarks from infringement. "[T]he general principles qualifying a mark for registration under §2 of the Lanham Act are for the most part applicable in determining whether an unregistered mark is entitled to protection under §43(a)." Two Pesps, Inc v. Taco Cabana, Inc., 505 U.S. 763, 768 (1992). In order to prevail on a claim of trademark infringement, the plaintiff most prove ownership of a protectable trademark and likelihood that an appreciable number of ordinary prudent purchasers will be confused by the similarity of the plaintiff and defendant's marks. See Savin Corp. v. Savin Group, 391 F.3d 439, 456 (2d Cir. 2004).

A. Classification of the Mark

In order to be protectable, a mark must be "`distinctive' and not `generic'" Christian Louboutin S.A. v. Yves Saint Laurent Am. Holdings, Inc., 696 F.3d 206, 216 (2d Cir. 2012), such that the mark is "capable of distinguishing the products it marks from those of others." Lane Capital Mgm't, Inc. v. Lane Capital Mgm't, Inc., 192 F.3d 337, 344 (2d Cir. 1999). Along the spectrum of distinctiveness, marks fall into one of roughly four categories. The categories of trademarks, listed in order of the unprotectable to most protectable, are (1) a generic mark; (2) a descriptive mark; (3) a suggestive mark; and (4) an arbitrary or fanciful mark. See Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976).

When a mark is unregistered, the party claiming trademark infringement has the burden of proving that the mark is protectable. See Courtenay Commc'ns Corp. v. Hall, 334 F.3d 210, 217 (2d Cir. 2003). The classification of a mark is a factual question that focuses on "how the purchasing public views the mark." Lane Capital Mgm't, 192 F.3d at 344; see also Star Indus., Inc. v. Bacardi & Co. Ltd., 412 F.3d 373, 382 (2d Cir. 2005); Courtney Commc'ns Corp., 334 F.3d at 215 (noting that for classification of mark "critical fact" is "how the purchasing public views" the mark). It is important to recognize that

Lane Capital Mgm't, 192 F.3d at 344.

In some cases the distinctiveness or non-distinctiveness of a mark may be self-evident from the examination of the mark and the product on which it appears. Id. at 348. In other cases, a party may seek to demonstrate the public's understanding of a term through other evidence, including "purchaser testimony, consumer surveys, dictionary definitions, trade journals, newspapers, and other publications." In re Reed Elsevier Prop. Inc., 482 F.3d 1376, 1378 (Fed. Cir. 2007); see also Borinquen Biscuit Corp. v. M.V. Trading Corp., 443 F.3d 112, 118 & n.4 (1st Cir. 2006). When the only evidence of the mark's appropriate classification is the mark and the product itself, the party bearing the burden will only be able to prevail "if the purchasing public could have only one view, or if one competing view is dominant enough to permit a reasonable conclusion by the preponderance standard." Lane Capital Mgm't, 192 F.3d at 348. In other words, "[t]he party bearing the burden of proof . . . has no right to take the case to jury if a favorable verdict could only be the product of surmise, speculation, and conjecture." Id. at 346 (citation omitted).

Classifying a mark is a delicate task, and one that carries great significance in a suit for trademark infringement. The importance of the determination stems from the varying degrees of protection and of required proof that accompany each category. A generic mark, for instance, is entitled to no trademark protection at all. A mark is generic if it "refers, or has come to be understood as referring, to the genus of which the particular product is a species." Abercrombie & Fitch Co., 537 F.2d at 9.

The next two categories — descriptive and suggestive marks — represent the "broad middle ground where most of the trademark battles are fought." West & Co., Inc v. Arica Inst., Inc., 557 F.2d 338, 342 (2d Cir. 1977).

Id. (quoting Stix Products v. United Merchants & Mfrs., Inc., 295 F.Supp. 479, 488 (S.D.N.Y. 1968)). A descriptive mark is entitled to trademark protection only if "the descriptive meaning of a word becomes subordinate and the term instead becomes primarily a symbol of identification." PaperCutter, Inc. v. Fay's Drug Co., Inc., 900 F.2d 558, 562 (2d Cir. 1990); Thompson Med. Co., Inc. v. Pfizer Inc., 753 F.2d 208, 216 (2d Cir. 1985). In other words, a descriptive mark is entitled to protection only to the extent that it acquires and maintains secondary meaning among consumers. See Thompson Med. Co., Inc., 753 F.2d at 216; Abercrombie & Fitch Co., 537 F.2d at 10 (indicating that trademark protectability can be lost). A suggestive mark is automatically entitled to trademark protection.

The final category of marks includes marks that are either fanciful — "words invented solely for their use as trademarks," — or arbitrary — common words "applied in an unfamiliar way." Genesee Brewing Co., Inc v. Stroh Brewing Co., 124 F.3d 137, 143 (2d Cir. 1997). These marks are inherently strong and do not require proof of secondary meaning.

To determine whether secondary meaning exists, a court considers whether the primary significance of the mark to the consuming public is to "identify the source of the product rather than the product itself." Mana Products, Inc. v. Columbia Cosmetics Mfg., Inc., 65 F.3d 1063, 1070 (2d Cir. 1995) (citation omitted). The party claiming trademark infringement bears the burden of proving secondary meaning, a task that "entails vigorous evidentiary requirements." 20th Century Wear, Inc. v. Sanmark-Stardust, Inc., 747 F.2d 81, 90 (2d Cir. 1984) (citation omitted). Factors that are relevant to a secondary meaning determination include

Genessee Brewing Co., 124 F.3d at 143 n.4 (citation omitted).

The plaintiffs claim the word "ibooks" is a protectable trademark because it "could" be suggestive of books with ideas, and because it is a suggestive mark, they do not have to demonstrate the existence of secondary meaning. Before addressing this assertion, it is important to note that the plaintiffs are not asserting that Apple has infringed the entirety of their mark as it appears on their books. As Colby admits, when the term ibooks is shown on a book it is almost always displayed as part of a larger mark whose principal element is a light bulb. Arguably, in light of this use, plaintiffs' mark should be viewed as a composite mark consisting of the word ibooks and the light bulb image ("ibooks Logo"). In fact, a composite mark must ordinarily be viewed as a whole to determine the mark's distinctiveness. See Courtney Commc'ns Corp., 334 F.3d at 215. As a composite mark, the ibooks Logo might easily be classified as an inherently distinctive mark. Even though, as discussed below, the word ibooks is descriptive, the light bulb image as a component of a book's imprint is not. The image does not immediately convey information about the ingredients, qualities, or characteristics of the plaintiffs' books, although it may suggest certain attributes. A mark that combines descriptive words with distinctive design elements, when viewed as a whole, can be inherently distinctive. See, e.g., Courtney Commc'ns Corp., 334 F.3d at 216.

In the present case, however, the plaintiffs do not claim protection for their ibooks Logo, but for the word "ibooks." In other words, the plaintiffs seek to "control" the defendant's use of the word that comrpises the plaintiff's mark. Id. at 215 n.3. To the extent the plaintiffs' claim that their use of the word "ibooks" is "separately protectable in its own right," they must demonstrate that the ibooks mark "creates a separate and distinct impression" from the ibooks Logo. Star Indus., 412 F.3d at 382. This is a task they have not even begun to undertake.

Because the plaintiffs' mark is unregistered, it is also their burden to offer evidence from which a reasonable jury could conclude that plaintiff's mark is a protectable mark. This they have failed to do as well. The only admissible evidence presented suggests that the "i" in ibooks was intended by the mark's creator to refer to the Internet and promoted as such. As the 1999 press release explained, the ibooks imprint was established "to take full advantage of the promotional and distribution potential of the [I]nternet through downloadable free chapters, virtual reading groups and message boards between authors and readers." (Emphasis supplied.) The letter "i" in reference to the Internet describes a feature of products published under the ibooks imprint by informing the consumer that the book is available for purchase on the Internet.

The use of the prefix "i" or "I," as a descriptive identifier that refers to the Internet is well recognized. For instance, the trademark office provides the following guidance in the Trademark Manual of Examining Procedure:

T.M.E.P. § 1209.03(d) (emphasis supplied); see also Lahoti v. VeriCheck, Inc., 586 F.3d 1190, 1199 (9th Cir. 2009) (noting that, even if mark is unregistered, PTO's treatment of similar marks may offer evidence of appropriate classification of mark). Indeed, as described below, the PTO concluded that Apple's iBooks mark is merely descriptive of an Internet service and unprotectable without evidence of secondary meaning.

The plaintiffs have submitted no evidence of how consumers perceive the ibooks mark and plaintiffs' Rule 30(b)(6) deponent testified that he was aware of no documents that would show that the letter "i" in ibooks refers to something other than the Internet. Nonetheless, the plaintiffs argue that their ibooks mark is suggestive rather than descriptive.

In support of their argument that the ibooks mark is suggestive, the plaintiffs point to the fact that the mark is almost always accompanied by a light bulb image with a lowercase letter "i" appearing inside the light bulb. The presence of the light bulb, the plaintiffs contend, might suggest to a consumer the concept of ideas. But, as explained above, the plaintiffs are not seeking protection for their ibooks Logo. Accordingly, the suggestive nature of the light bulb image fails to demonstrate that the word ibooks by itself "creates a separate and distinct impression" on consumers. Star Indus., 412 F.3d at 382. Where, as here, the plaintiffs' only evidence of the proper classification of the mark is the mark itself and the product on which it appears, the plaintiffs can only prevail at trial "if the purchasing public could have only one view, or if [the plaintiffs'] view is dominant enough to permit a reasonable conclusion by the preponderance standard." Lane Capital Mgm't, 192 F.3d at 348. Even if the ibooks mark could plausibly suggest the concept of ideas it does not compel that interpretation. The plaintiffs are not entitled to ask the jury to speculate as to how the consuming public perceives their mark.

The plaintiffs next point out that when Preiss attempted to register the IBOOKS mark with the PTO, the application was rejected not on the grounds of descriptiveness, but rather on the grounds of deceptive misdescriptiveness. See 15 U.S.C. § 1052(e). This point is hardly helpful to the plaintiffs. Under the Lanham Act, both descriptive and deceptively misdescriptive marks require proof of secondary meaning to be registered. See 15 U.S.C. §§ 1052(e),(f); see also McCarthy on Trademarks and Unfair Competition, § 11:55 Deceptive and deceptively misdescriptive marks (4th ed. 2013). A mark is classified as deceptively misdescriptive by the PTO when the term (1) misdescribes the goods or services; and (2) consumers are likely to believe the misrepresentation. In Re Phillips-Van Heusen Corp., 63 U.S.P.Q.2d 1047, *1 (T.T.A.B. 2002). "[I]n order for a term to misdescribe goods or services, the term must be merely descriptive, rather than suggestive, of a significant aspect of the goods or services which the goods or services plausibly possess but in fact do not." In Re Phillips-Van Heusen Corp., 63 U.S.P.Q.2d at *4 (emphasis supplied). A determination that a mark is misdescriptive does not constitute proof that the mark is suggestive. Moreover, when Preiss disputed the PTO's finding of misdescriptiveness, he did not claim that the mark suggests "books with ideas" to consumers or that the mark had achieved secondary meaning.

Finally, the plaintiffs contend that the classification of the ibooks mark is a disputed factual question that cannot be resolved on summary judgment. This argument ignores the fact that, as the parties bearing the burden at trial, the plaintiffs must be able to present admissible evidence that can create a genuine dispute of material fact. In this case, the plaintiffs have presented no evidence that the ibooks mark conveys anything to consumers other than "books available for sale on the Internet."11 In other words, the plaintiffs have not presented evidence to support a finding that the mark ibooks is anything other than a descriptive mark.

B. Secondary Meaning

Because the plaintiffs' ibooks mark is at best descriptive, they must present evidence demonstrating that, as of the time when Apple began using the iBooks mark, "the primary significance of the [plaintiffs' mark] in the minds of the consuming public [was] not the product but the producer." 20th Century Wear, 747 F.2d at 90 (citation omitted). In determining whether a mark has acquired secondary meaning, the focus must be on the relevant group of consumers, which are those who would ordinarily consider purchasing the plaintiff's product. Centaur Commc'ns, Ltd. v. A/S/M Commc'ns Inc., 830 F.2d 1217, 1221 (2d Cir. 1987); Lane Capital Mgm't, 192 F.3d at 345 ("[T]he relevant purchasing public is not the population at large, but prospective purchasers of the product."). In the present case, the plaintiffs have failed to raise a question of fact regarding the "rigorous evidentiary requirements" associated with establishing secondary meaning for their mark. 20th Century Wear, 815 F.2d at 10.

1. Advertising Expenditures

The plaintiffs claim that over $616,127 was spent in connection with advertising and marketing activities between the years 1999 through 2006, and approximately $43,000 was spent between 2007 and 2011. These figures come from spreadsheets created by Colby. For several reasons, these figures say little about the existence of secondary meaning for the mark ibooks.

First, both of these figures represent the amount spent on promoting two of the plaintiffs' imprints — ibooks and ipicturebooks. The plaintiffs are not pursuing a claim of trademark infringement with respect to their ipicturebooks mark.12 Accordingly, the only relevant advertising expenditures are those made in connection with products bearing the ibooks imprint. Without a breakdown of advertising expenditures by imprint, the significance of the $616,127 or $43,000 figures is speculative.

Second, the plaintiffs have offered little evidence of how any of this money was spent.13 The relevance of advertising expenditures to the secondary meaning analysis is premised on the idea that advertising may make it more likely that a consumer will identify a given mark with a single source. Cf. Centaur Commc'ns, 830 F.2d at 1221-23. Without evidence of how advertising funds were actually used, it is difficult to conclude that the money contributed to the creation of secondary meaning of the mark.

It is particularly difficult to evaluate the significance of the $616,127 figure. As described by Colby, this figure came from Simon & Schuster, the former distributor of books published by ibooks, inc., and appeared on a spreadsheet that the plaintiffs obtained from the bankruptcy trustee. It is unclear whether this figure represents money spent through a cooperative advertising program or something else. It is also unclear whether the ibooks mark (as opposed to book titles and authors' names, for instance) appeared in any advertising or promotional activities associated with this figure. And, of course, there is no evidence that any of these funds promoted the ibooks mark rather than the ibooks Logo. Accordingly, the plaintiffs have failed to offer any meaningful evidence of advertising expenditures contributing to secondary meaning.

2. Consumer Surveys

Although a plaintiff can establish secondary meaning through a variety of evidence, it is not uncommon for the proponent of secondary meaning to offer "some form of survey of consumer attitudes under actual market conditions." Mattel, Inc. v. Azrak-Hamway Intern., Inc., 724 F.2d 357, 361 (2d Cir. 1983). In this case, the plaintiffs did not offer evidence of a consumer survey to show that secondary meaning has attached to their ibooks mark.

3. Sales Success

The plaintiffs have offered spreadsheets reflecting net sales of over $33 million of books associated with the ibooks imprint between the years 1999 to 2011. The information for the years before 2007 was gathered by Colby after he acquired the assets of ibooks, inc. in the bankruptcy proceeding. Colby did not obtain any business records from the deceased Preiss or the bankrupt entities, but the bankruptcy trustee provided him with some information from Simon & Schuster and Colby wrote to about 15 other distributors to piece together this sales information.

Apple has argued that there are evidentiary problems that prevent these spreadsheets from being received as admissible evidence at trial.14 Assuming, without deciding, that the plaintiffs could overcome those hurdles at trial, the significance of these sales figures diminishes upon closer review.

Preiss died in July 2005, and the sales of products bearing the ibooks mark plummeted. Between 2005 and 2006, net sales of products bearing the ibooks mark fell by over $2,000,000. In the three years preceding Apple's announcement of the iBooks e-reader software, the plaintiffs' net sales were $118,749, negative $28,876, and $48,656. The drop-off in plaintiffs' sales success is significant to the secondary meaning analysis. Secondary meaning refers to an origin-identifying association that consumers have with a mark. Just as secondary meaning can be built-up over time, it can also diminish over time. That which the market has learned can be unlearned. Cf. Landers, Fray & Clark v. Universal Cooler Corp., 85 F.2d 46, 48 (2d Cir. 1936). The undisputed record reflects that, for the four years preceding the defendant's introduction of the iBooks software, the plaintiffs had little success in selling their products to consumers.15

4. Unsolicited Media Coverage

The plaintiffs have submitted no evidence that any newspaper or magazine directed to the public ever provided unsolicited media coverage of the ibooks mark or the businesses associated with the plaintiffs' mark. With a single exception, the only articles mentioning the ibooks imprint ran in the trade publication Publishers Weekly.16 Most of these articles were published between 1999 and 2006, but four of them were published between 2007 and March 10, 2009. In each instance, ibooks was mentioned in a cursory manner in the context of an article focused on other matters.17

5. Intentional Copying

Intentional copying of a plaintiff's mark by others supports a finding of secondary meaning. Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1042 (2d Cir. 1992); Coach Leatherware Co., Inc. v. AnnTaylor, Inc., 933 F.2d 162, 169 (2d Cir. 1991). The plaintiffs have offered no evidence of the intentional copying of the plaintiffs' ibooks mark.

6. Length and Exclusivity of Use

A plaintiff's long and exclusive use of a mark weighs in favor of finding that the mark has acquired secondary meaning. There is no bright line rule with respect to the length of time necessary for a mark to achieve secondary meaning. The significance of the length of use of a mark "is evaluated in light of the product and its consumers." Centaur Commc'ns, 830 F.2d at 1225. Furthermore, the fact that similar marks have been used by third-parties tends to diminish the possibility that the plaintiff's mark has developed secondary meaning. Streetwise Maps, Inc. v. VanDam, Inc., 159 F.3d 739, 744 (2d Cir. 1998).

The ibooks mark has been in use as a publishing imprint over a period of time spanning thirteen years. Its greatest success was in its earliest years, from 1999 to 2005. Beginning in 2006, the plaintiffs' sales declined precipitously, and since that time there have been only modest sales.

In addition to this thin record of use, the plaintiffs' own expert has described the particular challenges that exist in developing brand recognition in the publishing industry. As plaintiffs' publishing industry expert Michael Shatzkin ("Shatzkin") explains, "the role and behavior of `brands' in book publishing is somewhat unlike the way they play out in other consumer goods." Until relatively recently, the target of publishing companies' branding efforts were bookstores, retailer buyers, and book reviewers — those intermediaries that act as gatekeepers to the retail market. Traditionally, "imprints usually denoted a team of editors and marketers within a publishing house and . . . communicat[ed] an editorial philosophy and marketing approach to gatekeepers." Direct marketing to retail consumers by publishers is a relatively recent phenomenon. According to Shatzkin, a publisher's brand is created over time when "publishers deliver a `value' — a kind of book — consistently under an author, imprint, series, or company brand. The audience self-selects around the content, and the value of the brand is created over time by the experiences readers and consumers have with the published books." Thus, while the thirteen years in which the ibooks mark has been in use weighs in favor of secondary meaning, the significance of this length of time is tempered both by the uneven sales record and the particular challenges associated with creating secondary meaning for an imprint. Cf. Centaur Commc'ns, 830 F.2d 1225.

The non-exclusivity of plaintiffs' use also detracts from a finding of secondary meaning. The record demonstrates that plaintiffs' use of the ibooks mark has never been exclusive. The University of Illinois began using "I BOOK" to designate calendar handbooks in August 1988. A Texas-based company used the domain name ibooks.com in 2000. The Texas-based company's ibooks.com website was mentioned in five articles published in Publishers Weekly between March and October of 2000. One of the articles described ibooks.com as "a Web book retailer that sells online access to technical reference books." Also in 2000, Family Systems began using the word IBOOK in commerce to describe "computer hardware and software used to support and create interactive, user-modifiable electronic books." In 2010, the ABDO Publishing Group introduced a "line of interactive picture books, nonfiction, and graphic novels" called "ABDO iBooks." At least four third party entities have used variations of the ibooks mark to denote their own products. Such third party use of the word ibooks tends to diminish the likelihood that the plaintiffs have been able to create secondary meaning in their mark. Streetwise Maps, 159 F.3d at 744; Rockland Exposition, Inc. v. Alliance of Automotive Serv. Providers of New Jersey, 894 F.Supp.2d 288, 323 (S.D.N.Y. 2012) (collecting cases).18

7. Other Evidence of Secondary Meaning

In support of their claim that the ibooks imprint has achieved secondary meaning, the plaintiffs have offered the testimony of two witnesses. Neither witness provides meaningful support for a finding of secondary meaning.

First, the plaintiffs offer the testimony of Shatzkin, whom they describe as an expert on the publishing industry.19 On the basis of a spreadsheet containing subsets of data derived from the same source files that were used to create the spreadsheets of plaintiffs' sales, Shatzkin opined that:

(Emphasis supplied.)

As evidenced by the statement quoted above, Shatzkin's opinion was expressly limited to readers of science-fiction novels. The relevant market for plaintiffs' products, however, is admittedly much broader.20 Furthermore, since the bulk of the ibooks' sales occurred before 2006, Shatzkin's opinion provides little support for a finding of secondary meaning past that date. Finally, the plaintiffs have not shown that Shatzkin is qualified to opine regarding the assumption that he proffers: that any purchaser of a book bearing the ibooks mark "recognized" the imprint. That leap of logic is particularly difficult to make since the dominant feature of the imprint's logo was not the descriptive term ibooks, but the illustration of a light bulb with a stylized "i."21 Accordingly, Shatzkin's opinions — even if admissible — would be of little assistance to the jury in deciding whether the ibooks mark has secondary meaning for a substantial segment of the relevant ordinary consumers.

The plaintiffs have also offered the testimony of Richard Freese, an individual who worked for two distributors of ibooks.22 Freese worked as President of Publishers Group West ("PGW") for some unspecified period of time around 2006. Freese recalled signing ibooks, inc. as a client, and estimated that PGW distributed Preiss' ibooks for some period of time between 2003 and 2006. Freese joined another distributor, National Book Network ("NBN") in 2010. When Freese joined NBN in 2010, NBN was distributing books published under the ibooks imprint. The plaintiffs have not sought to qualify Freese as an expert. At his deposition, Freese testified that "ibooks" had a "strong reputation" at PGW in the category of science fiction and fantasy, and had products "that we knew that our customers were going to want." Freese added that he could infer that consumers recognized the ibooks brand from the fact that book sellers like Barnes & Noble and Walden bought books published by ibooks from PGW.

As his testimony reveals, Freese's opinion assumes that because retailer purchasers were willing to buy books published under the ibooks imprint, individual consumers — by extension — must have recognized the brand. This is pure speculation. As a lay witness, Freese is only capable of testifying to matters of which he has first-hand knowledge. Fed. R. Evid. 701(a) and advisory notes. As a result, Freese's testimony offers scant evidence of secondary meaning.

8. Aggregate Assessment of Secondary Meaning

Weighing each of these factors, the plaintiffs have failed to submit sufficient evidence to raise a genuine issue of material fact with respect to whether a "substantial segment of the relevant group of consumers made the requisite association" between a source and the mark. Centaur Commc'ns, 830 F.2d at 1221-22. The strongest evidence in support of a finding of secondary meaning is the volume of sales of books bearing the ibooks imprint. Those sales were largely made before 2006, however, and the plaintiffs have little evidence of any commercial success since that time. The only other evidence weighing in the plaintiffs' favor is the length of time during which the ibooks mark has been in use. Given the uneven sales record for ibooks products and the particular challenges associated with cultivating brand recognition for a publishing imprint, the weight of this factor is limited. The plaintiffs have not presented evidence from which a jury could find that any serious advertising campaign was undertaken to create recognition of the mark. Similarly, the plaintiffs have offered only minimal evidence of unsolicited media coverage, no consumer surveys, and no evidence of intentional copying. Finally, the plaintiffs' additional sources of evidence contribute little more than speculation regarding how consumers might have once viewed the ibooks mark. Drawing all inferences in the plaintiffs' favor, no reasonable jury could conclude that, as of 2010 when Apple announced its e-reader software, a substantial segment of ordinary consumers in the plaintiffs' market associated the mark "ibooks" with a single source.

This conclusion is entirely consistent with the price paid for the bankrupt entities' assets in 2006. Colby paid $125,000 to acquire all of the intellectual property rights and licenses for software programs, computer hardware, remaining credit balances for advances made to authors and licensors, manuscripts of two publishing companies, and 300,000 copies of physical books. This relatively small sum of money spread over so many assets, and Colby's failure to show that he has achieved any commercial success with the mark since he acquired it, strongly suggest that there was little or no secondary meaning attached to the mark as of 2006 or thereafter. A descriptive mark that has not acquired (or has lost) secondary meaning is not entitled to trademark protection and as a result the plaintiffs' trademark claim must fail.

C. Likelihood of Confusion

The plaintiffs' trademark claim fails for another, entirely independent, reason. The plaintiffs have not presented evidence from which a jury could find a likelihood of confusion.

In addition to proving commercial use of a protectable mark, a plaintiff must demonstrate that, as a result of the defendant's appropriation of a confusingly similar mark, "an appreciable number of ordinary prudent purchasers are likely to be misled, or indeed simply confused" as to the source, affiliation, sponsorship, connection, or identification of the plaintiff or defendant's products. Savin Corp., 391 F.3d at 456 (citation omitted); see also Star Indus., 412 F.3d at 383. The Second Circuit has recognized that various forms of confusion are actionable under the Lanham Act. See Star Indus., 412 F.3d at 383.

Relevant to the case at hand, the Second Circuit has acknowledged that reverse confusion is actionable. See Malletier v. Burlington Coat Factory Warehouse Corp., 426 F.3d 532, 539 n.4 (2d Cir. 2005). "Reverse confusion is the misimpression that the junior user is the source of the senior user's goods." Banff, Ltd. v. Federated Dept. Store, Inc., 841 F.2d 486, 490 (2d Cir. 1988). See also Lang v. Retirement Living Pub. Co., Inc., 949 F.2d 576, 583 (2d Cir. 1991)("Reverse confusion exists when a subsequent user selects a trademark that is likely to cause consumers to believe, erroneously, that the goods marketed by the prior user are produced by the subsequent user."). Reverse confusion is contrasted with the more common form of confusion that occurs when a consumer mistakenly believes that the senior user is the source of the junior user's product.

In order to assess the likelihood that consumers will be confused by the similarity of the opposing party's marks, courts apply the eight factors articulated in Polaroid Corp. v. Polaroid Electronics Corp., 287 F.2d 492 (2d Cir. 1961). In particular, courts consider (1) the strength of the plaintiffs' mark; (2) the similarity of the marks; (3) the proximity of the parties' products; (4) the likelihood that the plaintiff will bridge the gap; (5) actual consumer confusion; (6) bad faith of the defendant in adopting the similar mark; (7) the quality of the defendant's products; and (8) sophistication of the consumer.

In a case where reverse confusion is alleged, additional considerations are relevant. In particular, with respect to the first factor courts should (1) consider the strength of the junior user's mark as well; and (2) recognize that the commercial weakness of the senior user's mark actually makes reverse confusion more likely.23 See A&H Sportswear, Inc. v. Victoria's Secret Stores, Inc., 237 F.3d 198, 230-31 (3d Cir. 2000). With respect to actual consumer confusion, which is commonly demonstrated through the use of consumer surveys, the Second Circuit has held that the senior user's consumers are the appropriate class of consumers to survey in reverse confusion cases. See Sterling Drug, Inc. v. Bayer AG., 14 F.3d 733, 742 (2d Cir. 1994).

1. Strength of the Marks

In assessing the strength of a mark, courts focus "on the distinctiveness of the mark, or more precisely, its tendency to identify the goods sold under the marks as emanating from a particular, although possibly anonymous source." W.W.W. Pharm., 984 F.2d at 572 (citation omitted). This inquiry encompasses two elements: (1) the degree to which the mark is inherently distinctive; and (2) the degree to which it has acquired distinctiveness in the marketplace. Star Indus., 412 F.3d at 384-85. With respect to inherent distinctiveness, courts refer back to the classification of the mark. A descriptive mark is an inherently weak mark. See Virgin Enterps. Ltd v. Nawab, 335 F.3d 141, 148 (2d Cir. 2003). As discussed above, the plaintiffs have offered insufficient evidence to demonstrate that consumers associated the ibooks mark with a single source at the time when Apple began using the iBooks mark. Even if the plaintiffs' mark were classified as suggestive, the lack of evidence of secondary meaning would still indicate that plaintiffs' mark is relatively weak. See Star Indus., 412 F.3d at 385-86.

The plaintiffs make two points regarding the relative strength of the two marks. First, the plaintiffs note that the strength of the junior user's mark is relevant to the reverse confusion analysis. The defendant's iBooks mark has no more conceptual or inherent strength than the plaintiffs' ibooks mark. As the defendant itself explains, the PTO initially rejected Apple's application to register IBOOKS for "expanded goods and services" because the PTO considered the mark to be merely descriptive. On the other hand, the parties do not dispute that the defendant's iBooks mark has achieved secondary meaning. Accordingly, it will be assumed that Apple's iBooks mark has become a strong mark.

Next, the plaintiffs argue that in the two years since Apple announced its iBooks software, the plaintiffs' mark has become commercially weak and, as a result, this factor favors the plaintiffs. The plaintiffs' contention is not supported by their own sales figures. The plaintiffs' sales plummeted in 2006, four years before Apple announced the iBooks mark in 2010. In 2009, the plaintiffs' net sales of products bearing the ibooks imprint were $48,000. Since Apple announced its iBooks software, the plaintiffs' net sales have improved, albeit slightly. In 2010 and 2011, the net sales of books published under the ibooks imprint were $53,667 and $72,300 respectively. As a result, there is no evidence that the commercial weakness of plaintiffs' mark has been influenced by the defendant's use of the mark iBooks.

Moreover, commercial weakness is in some ways a double-edged sword. On the one hand, a commercially weak mark is more vulnerable to reverse confusion. On the other hand, part of what entitles a mark to protection is its ability to serve as an indicator of origin. Accordingly, to the extent a senior user has invested so little in its mark that it has failed to create an association in the minds of consumers between the mark and a source, there is correspondingly less reason to protect the mark. After all, "[t]he chief danger inherent in recognizing reverse confusion claims is that innovative junior users, who have invested heavily in promoting a particular mark, will suddenly find their use of the mark blocked by plaintiffs who have not invested in, or promoted, their own marks." A&H Sportswear, 237 F.3d at 228.

2. Similarity of the Marks

In comparing the two marks, a court should ask "(1) whether the similarity between the two marks is likely to cause confusion and (2) what effect the similarity has upon prospective purchasers." The Sports Auth., Inc. v. Prime Hospitality Corp., 89 F.3d 955, 962 (2d Cir. 1996). Similarity is gauged by looking "at the general impression created by the marks, taking into account all factors that potential purchasers will likely perceive and remember." Lang, 949 F.2d at 581. The fact that the marks use the same word is not dispositive if the differences in the ways the marks are presented in the marketplace make confusion less likely. See Star Indus., 412 F.3d at 386; W.W.W. Pharm., 984 F.2d at 573. In this regard, it is appropriate to consider "the products' sizes, logos, typefaces, and package designs," in addition to any other contextual clues that might serve to distinguish the marks. W.W.W. Pharm., 984 F.2d at 573; see also Malletier, 426 F.3d at 538 ("Lanham Act requires a court to analyze the similarity of the products in light of the way in which the marks are actually displayed in their purchasing context.").

In this case, the parties' marks consist of the same word: IBOOKS. The differences between the marks, however, are significant and derive from both the nature of the marks and the context in which they appear.

Plaintiffs' ibooks mark is almost always accompanied by a lower case "i" enclosed in a light bulb, positioned directly above the mark. From 1999 through 2011, plaintiffs' mark was consistently depicted in all lowercase letters. The lowercase "i," which appears twice — once in the light bulb and once as the first letter in ibooks — is depicted in a font that has a "flag" on the letter "i." Apple's iBooks mark, on the other hand, is written in a different typeface, it is depicted next to an image of an open book against a wood-colored background, and the letter B is capitalized.

The context in which the two marks appear is also quite different. Plaintiffs' mark appears on both their physical books and ebooks. In many, if not all cases, the plaintiffs' mark is in close proximity to contextual information indicating that it is associated with a publisher. This information includes the name of the publishing company, the name of the distributor, the physical and web address of the publisher, the copyright date, and the International Standard Book Number (ISBN), which is a unique nine-digit code used to identify books.24 The Apple iBooks mark, on the other hand, appears in an Apple-branded environment — either preinstalled on Apple devices or on the Apple online App Store.

The plaintiffs emphasize but a single feature of the two marks — the case for the letter B — to argue that the differences are less important than they seem. They argue that the capitalization of the letter B is of little significance in distinguishing the marks. In support of this proposition, the plaintiffs point to two consumer confusion surveys conducted by their expert witness Dr. Susan McDonald ("McDonald"). They point out that McDonald's surveys tested both "ibooks" and "iBooks" and found identical rates of confusion for both versions of the mark. Even if it were appropriate to confine the comparison of the marks to this single letter — and it is not — for reasons discussed at length below, the McDonald surveys are so flawed that they are inadmissible at trial. Fed.R.Evid. 403.

3. Proximity of the Products

The proximity of the products in the marketplace is the next factor to be evaluated. In judging proximity, courts take into account whether the products compete, whether the goods serve the same purpose, and whether they "fall within the same general class, or are used together." Savin Corp., 391 F.3d at 458 (citation omitted). In addition, "[t]he court may consider whether the products differ in content, geographic distribution, market position, and audience appeal." Id. (citation omitted).

Although the products of the parties each relate to books, they are not proximate in the marketplace. Apple is a technology company that offers a computer software called iBooks that enables users to download and read ebooks. The plaintiffs are publishing companies that use the ibooks imprint on their physical books and ebooks. These products do not directly compete. While the ebooks that can be read on Apple's iBooks software may compete directly with the plaintiffs' physical books and ebooks, a consumer could not purchase the defendant's product — software — in place of the plaintiffs' — books.

Second, the parties' products are sold through different channels. The plaintiffs' products are sold in brick-and-mortar stores and on third-party websites like Amazon.com and BarnesandNoble.com. The plaintiffs' products are not available for sale through the defendant's website, iTunes Store or iBookstore. The defendant's e-reader software, on the other hand, is not available in brick-and-mortar stores or through third-party websites. Instead, the iBooks software is only available to consumers as pre-installed software on Apple devices or as a download from Apple's online App store. In the plaintiffs' favor, however, is the fact that the parties' products may appeal to overlapping audiences. For instance, plaintiffs' ebooks and defendant's e-reader software may both appeal to readers of ebooks.

The plaintiffs raise two points. First, they argue that the plaintiffs and defendant offer complementary goods that can, at least as a theoretical matter, be used together, which makes consumer confusion more likely. The plaintiffs have not shown, however, that any consumer is actually able as of today to read the plaintiffs' ebooks on the defendant's e-reader software. The plaintiffs' ebooks are not available for sale on Apple's website, iTunes Store, or Apple's iBookstore. Moreover, over 98% of the plaintiffs' sales over the past thirteen years have been of physical books.

Next, the plaintiffs argue that consumers are likely to mistakenly believe that the defendant's iBooks mark refers to the ebooks sold through Apple's iBookstore rather than the e-reader software. This confusion over the meaning of Apple's iBooks mark will in turn, plaintiffs argue, cause consumer confusion with respect to the plaintiffs' physical and electronic books. Of course, if this is occurring, it should have a virtually identical effect on every publisher of every ebook. The plaintiffs have failed to offer any proof, however, that consumers actually make this mistake. They have offered no evidence that consumers who use Apple's iBooks software to download ebooks have come to believe that Apple has also entered the publishing business and is the publisher of all of the downloaded books, despite the fact that each book bears the imprint of its actual publisher. Indeed, McDonald — plaintiffs' expert — admits that consumers do not regard Apple as a publisher. The plaintiff has offered a total of four examples of instances in which people mistakenly referred to ebooks for sale on the iBookstore as "iBooks."25 This evidence is insufficient to raise a genuine issue of material fact regarding whether Apple's iBooks mark should be interpreted as referring to ebooks rather than Apple's e-reader software for purposes of the proximity analysis.

4. Bridging the Gap

This factor, known as "bridging-the-gap," weighs the likelihood that the plaintiffs will enter the defendant's business or the "average customer's perception of the likelihood that the plaintiff would enter the defendant's market." The Sports Auth., 89 F.3d at 963. There is no evidence that the plaintiffs intend to develop e-reader software or that plaintiffs' customers would expect the plaintiffs to do so.

The plaintiffs nonetheless argue that if Apple had not begun using the iBooks mark the plaintiffs might have been able to "create a powerful niche brand in the digital space." The hypothetical possibility that the plaintiffs might have developed a more robust presence in the "digital space" in the absence of Apple's use of the iBooks mark hardly demonstrates that the plaintiffs were likely to bridge the gap between the publishing and computer software fields. Characterizing the defendant's field as the "digital space" would render this factor meaningless. Many companies operate in the "digital space" without offering remotely similar products.

5. Actual Confusion

Although evidence of actual confusion is especially probative of a likelihood of confusion, a plaintiff does not need to show the existence of actual confusion in order to prevail under the Lanham Act. Savin Corp., 391 F.3d at 459. When parties do offer evidence of actual confusion, or lack thereof, the evidence commonly includes anecdotal evidence of consumer confusion and consumer confusion surveys. See, e.g., Star Indus., 412 F.3d 373 (2d Cir. 2005). In the context of a case where reverse confusion is alleged, the senior user's consumers are the appropriate class of consumers to survey. See Sterling Drug, 14 F.3d at 741; see also Citizens Fin. Group, Inc. v. Citizens Nat. Bank of Evans City, 383 F.3d 110, 120-21 (3d Cir. 2004).

The plaintiffs' principal evidence of actual confusion is found in the surveys conducted by the plaintiffs' expert McDonald and the accompanying expert reports.26 McDonald's surveys found high rates of reverse confusion. In particular, both surveys concluded that between 47% and 59% of survey takers incorrectly attributed the source, sponsorship, or affiliation of plaintiffs' ibooks27 products to Apple. In light of serious flaws in McDonald's methodology, however, McDonald's survey results are not probative of the questions that are relevant to likelihood of confusion. Accordingly, Federal Rules of Evidence 702 and 403 compel exclusion of McDonald's surveys.

When a party proffers expert testimony, a court — as "gatekeeper" — has the responsibility to ensure that the expert is qualified to give such testimony and that the expert's opinions emanate from reliable methods, reliably applied. See Daubert v. Merrel Dow Pharmaceuticals, Inc., 509 U.S. 579, 592-93 (1993). In determining "whether the proffered testimony has a sufficiently reliable foundation to permit it to be considered," courts consider certain indicia of reliability, including "(1) that the testimony is grounded on sufficient facts or data; (2) that the testimony is the product of reliable principles and methods; and (3) that the witness has applied the principles and methods reliably to the facts of the case." Amorgianos v. National R.R. Passenger Corp., 303 F.3d 256, 265 (2d Cir. 2002) (citation omitted); Fed. R. Civ. P. 702.

Nimley v. City of New York, 414 F.3d 381, 397 (2d Cir. 2005) (citation omitted).

In order to understand the deficiencies in McDonald's reports and surveys, it is necessary to describe the surveys in some detail. Both of McDonald's surveys were conducted online. Consumers were deemed eligible for the survey if they

Those who were eligible to take the survey were divided into two groups — the Test Arm and the Control Arm. The Test Arm answered questions with respect to the iBooks/ibooks mark, while the Control Arm answered questions about a hypothetical eBooks/ebooks mark. The survey did not present survey takers with a visual representation of either the plaintiffs' or defendant's mark. Instead, the first question provided:

(Emphasis supplied.) In McDonald's first survey the Test Arm was told to envision a page with the word iBooks on it, while the Control Arm was told to envision a page containing the word eBooks. McDonald's second survey was the same as her first study, except that the Test Arm used the word ibooks and the Control Arm used the word ebooks.28

There is general agreement that, to be probative of actual confusion, a survey should make some attempt to replicate market conditions. See, e.g., Leelanau Wine Cellars Ltd. v. Black & Ref., Inc., 502 F.3d 504, 518 (6th Cir. 2007); Spraying Sys. Co. v. Delavan, Inc., 975 F.2d 387, 396 (7th Cir. 1992); Calvin Klein Cosmetics Corp. v. Lenox Labs., 815 F.2d 500, 504 (8th Cir. 1987); Am. Footwear Corp. v. Gen. Footwear Co. Ltd., 609 F.2d 655, 660 n.4 (2d Cir. 1979). This standard is consistent with the goals of the Lanham Act. The Lanham Act does not protect against confusion in the abstract, instead, it protects consumers from confusion in the marketplace. See Virgin Enters., 335 F.3d at 147. Accordingly, courts have repeatedly emphasized the importance of looking to actual market conditions in assessing likelihood of confusion. See Malletier, 426 F.3d at 534; Universal City Studios, Inc. v. Nintendo Co. Ltd., 746 F.2d 112, 117 (2d Cir. 1984) ("Where . . . the two properties are so different . . . [a] claim cannot stand without some indication of actual confusion or a survey of consumer attitudes under actual market conditions." (citation omitted)). Likewise, the contextual clues that can serve to dispel otherwise existent confusion are widely cited as relevant in assessing the likelihood that an appreciable number of consumers will be confused. See, e.g., Jim Beam Brands Co. v. Beamish & Crawford Ltd., 937 F.2d 729, 735 (2d Cir. 1991) (collecting cases); Vitarroz Corp v. Borden, Inc., 644 F.2d 960, 968 (2d Cir. 1981).

McDonald's surveys, however, made no attempt to replicate market conditions and deprived the survey takers of every contextual clue they would encounter when looking at the plaintiffs' products.29 As described above, there are a number of stylistic features that differentiate the plaintiffs' and defendant's marks. A consumer would be able to see these differences under actual market conditions, but McDonald's survey takers would not.

The plaintiffs offer roughly three reasons why McDonald's "conceptual stimulus" surveys were appropriate under the circumstances. First, the plaintiffs contend that McDonald's survey offered sufficient "context clues" because it instructed the survey takers to envision certain contextual information "such as the date of publication, the publisher, the Library of Congress number, etc." Even if survey takers have vivid imaginations, there is no reason to believe that the respondent's imagined pages come close to replicating what a consumer sees when they look at a page in a book. Furthermore, it is entirely possible that each survey taker envisioned a slightly different page, with a slightly different depiction of the iBooks/ibooks mark. What these survey takers saw in their minds is unknown and unknowable. Lastly, even if survey takers were able to imagine all of the contextual information they were instructed to envision, McDonald's surveys made no mention of plaintiffs' distinctive light bulb logo. The ibooks Logo, which by all accounts has never been used in connection with Apple's iBooks product, would be expected to dispel some confusion that consumers might otherwise experience. In sum, McDonald's survey results may be probative of the respondents' word associations, but little more.

Next, the plaintiffs argue that because a senior user of a trademark is "free, within certain parameters, to present their mark in whatever form they choose," a single visual representation of the plaintiffs' mark or products would have been unduly restrictive. This argument fails for at least two reasons. It is a fundamental precept of trademark law that a mark's entitlement to trademark protection stems from the owner's use of the mark in commerce such that consumers come to associate the mark with a single source. The Lanham Act is not meant to guard against hypothetical confusion that could exist if a trademark user altered his mark; it seeks to prevent actual confusion among consumers with respect to the source, affiliation, and sponsorship of consumer goods. It is true that a trademark owner can make non-material alterations to his mark without risking abandonment of the mark, see Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947, 955 (7th Cir. 1992), but this hardly justifies conducting a consumer study that uses no visual representation of the mark at all.30

The plaintiffs' argument is woefully disconnected from the facts of this case. The plaintiffs' principal admits that the ibooks mark has been depicted consistently with all lowercase letters from 1999 until 2011. Colby testified that he could think of no other way that the mark had ever been depicted. He further stated that the ibooks Logo, with the lower case "i" inside of a light bulb, appears with the ibooks mark on every physical book published by the plaintiffs. He also testified that the light bulb image is intended to appear on every ebook as well.31 Accordingly, the plaintiffs' proposition that no single visual representation of their mark could possibly capture the manifold variations of the plaintiffs' mark is mistaken at best.

Lastly, the plaintiffs argue that because books are different from other consumer goods, showing an actual book or ebook to survey takers would not adequately simulate the manner in which readers experience books and become aware of publishing imprints. In particular, the plaintiffs explain that "[i]t is only after a book has been read and experienced that the reader may be drawn to learn more about the book, and that certain subtleties, such as the imprint, may become relevant enough to command the reader's attention and reflection." Accordingly, McDonald's surveys did not attempt to replicate the book-shopping experience, but instead sought to "pick the moment when a customer becomes aware that there is something in a book that identifies it, in a digital book in particular, that identifies it as iBooks."

This fails to explain, however, why a survey that instructs respondents to envision a nondescript page in an unnamed book comes closer to simulating the experience of a reader with a book than a more traditional consumer confusion survey. The plaintiffs may be right that consumers do not experience books in the same way they experience other products. But however a consumer experiences a book, that experience occurs with respect to an actual book, not an imagined one. And once again, it is necessary to remember the context in which McDonald's surveys are offered. McDonald's expert testimony and the results of her surveys are useful only to the extent they address relevant questions. See Sterling Drug, 14 F.3d at 741. Here, where reverse confusion is alleged, the relevant question is whether there is a probability, not just a possibility, that an appreciable number of plaintiffs' potential customers would, upon seeing the plaintiffs' ibooks mark on one of plaintiffs' products, be confused about the source, affiliation, or sponsorship of plaintiffs' products. McDonald's surveys offer little assistance in answering this question. Their minimal probative value is thus substantially outweighed by their tendency to confuse the issues and mislead a jury.

The failure of the surveys to take market conditions into account is reason enough for the surveys to be excluded. But, the surveys also contain another serious flaw. In order to offer sound results, most surveys must employ an adequate control. By using a control, a consumer confusion survey is able to account for or rule out confusion that is caused by factors other than the defendant's infringing conduct. "In order to prove actual confusion, the confusion must stem from the mark in question." Gen. Motors Corp v. Lanard Toys, Inc., 468 F.3d 405, 414 (6th Cir. 2006).

In the present case, the possibility that consumers would be confused about the source, sponsorship, or affiliation of plaintiffs' products for reasons other than Apple's use of the iBooks mark is especially high. Apple has a well known family of "i" marks, which includes iPad, iPhone, iPod, and iTunes. But the plaintiffs' claims do not challenge the defendant's use of any mark other than "iBooks." By using the word eBooks/ebooks as the control, McDonald's surveys failed to account for those consumers who associated the mark ibooks/iBooks with Apple not because of Apple's use of the iBooks mark, but because of Apple's other unchallenged "i" marks. It is possible, for instance, that some consumers who have never heard of Apple's iBooks software would nonetheless associate the word ibooks with Apple simply because Apple promotes a number of "i" marks. Put another way, there are probably consumers who would think that products labeled iNovel or iLit are associated with Apple, even though Apple does not use these marks.

Confirming this expectation, some survey takers made the connection between ibooks/iBooks and Apple principally because of Apple's other marks. The following are four examples of reasons survey takers gave for attributing the ibooks/iBooks mark to Apple:

This type of confusion could exist even if Apple had never adopted the iBooks mark. Accordingly, in order to account properly for this background level of confusion, an appropriate control would have included an "i" prefix.

Apple argues that McDonald's surveys suffered from a variety of other shortcomings as well. In particular, the defendant argues that the surveys failed to ask appropriate questions to test confusion, posed questions that primed respondents to think of Apple, failed to test the proper universe of consumers,32 did not adequately screen respondents or take steps to prevent guessing, and were not validated. Some of these errors, on their own, may not have been fundamental enough to justify the exclusion of McDonald's reports and survey results. Taken together with the serious flaws described above, however, they confirm the conclusion that the plaintiffs' expert reports and surveys are inadmissible under Rules 702 and 403.

6. Bad Faith

As indirect evidence of likelihood of confusion, courts consider whether the defendant adopted the similar mark with the intent to capitalize on the plaintiff's reputation and goodwill or to foment confusion between the two marks. See, e.g., Star Indus., 412 F.3d at 388; Lang, 949 F.2d at 583. The burden of proving bad faith rests with the party claiming infringement. Star Indus., 412 F.3d at 388.

A defendant may also choose to offer evidence of its good faith, such as the fact that the defendant selected a mark which reflected the product's characteristics, requested a trademark search prior to the mark's selection, or relied on the advice of counsel in adopting the mark. Lang, 949 F.2d at 583. The failure of a defendant to conduct a trademark search, however, is not sufficient to demonstrate bad faith. Star Indus., 412 F.3d at 388. Indeed, even "actual knowledge of another's prior registration of a very similar mark may be consistent with good faith." Lang, 949 F.2d at 584.

The plaintiffs have offered no evidence indicating that Apple adopted the iBooks mark in bad faith.33 This absence of evidence is not surprising given the plaintiffs' small footprint in the marketplace.

The defendant, on the other hand, has offered substantial evidence of its good faith. First, the defendant selected a mark that describes characteristics of its product. The iBooks software allows users to download ebooks over the Internet and read those books. The mark is also related to marks Apple has adopted for other products and services it offers. The defendant also employed counsel to conduct a trademark search that uncovered no use of the ibooks mark by the plaintiffs. Where Apple did identify an owner of a relevant trademark — Family Systems — it negotiated a purchase of rights before announcing the iBooks mark.

The plaintiffs do not dispute as a factual matter that the defendant's trademark search did not reveal the existence of the plaintiffs or their use of the ibooks mark. The plaintiffs do not own a trademark registration for their ibooks mark. In addition, while the plaintiffs' predecessor did file a trademark application that was subsequently abandoned, the plaintiffs themselves have never filed such an application.

The parties agree that the trademark search did reveal Preiss' abandoned trademark application for ibooks and the fact that ibooks, inc. had filed for bankruptcy. With respect to Preiss's trademark application, the defendant's 30(b)(6) witness testified that

The plaintiffs object to the evidence of defendant's trademark search on two grounds. First, the plaintiffs argue that the inadequacy of the defendant's trademark search is actually evidence of the defendant's bad faith. They note, for example, that upon learning that Preiss' company ibooks, inc. had filed for bankruptcy, the defendant did not review any documents filed in the bankruptcy action and did not attempt to learn whether the assets of ibooks, inc. had been purchased by anyone. But, even if a more extensive investigation of the bankruptcy proceeding would have uncovered the plaintiffs' purchase of the assets, the defendant's failure to so expand its investigation is not evidence of bad faith. See Star Indus., 412 F.3d at 388.

Second, the plaintiffs argue that the defendant is barred from offering evidence of its trademark search because the defendant relied on its attorney-client privilege to bar discovery of Apple's communications with its attorneys concerning the search and clearance process. Apple has disclosed the results of its trademark search, and the plaintiffs have deposed two of Apple's attorneys who had roles in the trademark search. The defendant did not restrict the plaintiffs' inquiry into "the fact of what searches" were conducted or what was found in those searches. Under the circumstances, the defendant is entitled to offer evidence that the plaintiffs' use of the ibooks mark was not revealed by the trademark search.34

Although a defendant's reliance on a trademark search and advice of counsel are related, they are distinct bases for a finding of good faith. W.W.W. Pharm., 984 F.2d at 575 ("Good faith can be found if a defendant has selected a mark which reflects the product's characteristics, has requested a trademark search or has relied on the advice of counsel (emphasis supplied)). A defendant does not waive its attorney-client privilege by relying on evidence that it conducted a trademark search or by describing its investigation of the search results. There has been no improper effort by Apple to rely on the privilege as both a sword and a shield. See In re Sims, 534 F.3d 117, 132 (2d Cir. 2008).

Finally, the plaintiffs claim that Apple's bad faith can be inferred from the fact that Apple persisted in its decision to use the iBooks mark after receiving Colby's January 29 Email. Colby's email was sent two days after Apple publicly announced its iBooks e-reader software. Without evidence that Apple had prior knowledge of the plaintiffs' ibooks mark, no reasonable jury could conclude that Apple adopted the iBooks mark with the intent to trade on the plaintiffs' goodwill or to create confusion between the two IBOOKS marks. Even with evidence of Apple's prior knowledge — which plaintiffs do not have — the plaintiffs would have to show more to suggest bad faith. It is in any event worth noting that Colby's email cannot be fairly characterized as a cease and desist letter. The email does not contain any allegation that Apple was infringing the plaintiffs' rights or any request that Apple cease using the iBooks mark.

7. Quality of the Products

The Second Circuit has clarified that while "there are two issues with regard to quality . . . only one has relevance to determining the likelihood of confusion." Savin Corp., 391 F.3d at 461. The fact that a junior user's product is of inferior quality to the senior user's product may tend to make injury to the senior user's reputation more likely, but

Id. In the context of reverse confusion, a buyer will be less likely to assume that the better-known junior user has produced the senior user's products if the two products are of noticeably different quality. The parties have submitted little to no evidence on the relative quality of their products. The plaintiffs do not argue that this factor assists their claim of confusion.

8. Sophistication of the Consumer

A sophisticated consumer is less likely to be confused by similar marks than is a casual shopper. As in Merriam-Webster, Inc v. Random House, Inc., 35 F.3d 65 (2d Cir. 1994), "there are two pertinent classes of potential consumers: retail book sellers and individuals." Id. at 72. Retail sellers of books are generally "assumed to be sophisticated buyers." Id. It is extremely unlikely that the retail book sellers will mistakenly believe that Apple is the source of plaintiffs' books or is otherwise affiliated with the plaintiffs' products. Furthermore, the parties agree that readers of plaintiffs' books are sophisticated as well. Thus, this final factor in the Polaroid analysis does not suggest a likelihood of confusion.

9. Aggregate Assessment of Likelihood of Confusion

Taking all of the Polaroid factors into account, and drawing all inferences in the plaintiffs' favor, the plaintiffs have failed to raise a genuine issue of material fact with respect to likelihood of confusion. On balance, the Polaroid factors weigh heavily against a finding that an appreciable number of ordinary prudent consumers are likely to be confused. Indeed, apart from the fact that both parties use marks presenting a variation of the word IBOOKS, there is little in the record to suggest that consumers will mistakenly believe plaintiffs' books originate with, are sponsored by, or are affiliated with Apple. Accordingly, for this additional reason, the defendant is entitled to summary judgment on the plaintiff's Lanham Act claim.

II. State Law Claims

The plaintiffs also assert state law claims for infringement of common law trademark and unfair competition, wrongful misappropriation by unfair competition, and unjust enrichment under New York State law.35 The plaintiffs' infringement of common law trademark and unfair competition claims "share[] many common elements with the Lanham Act claims of false designation of origin and trademark infringement," including likelihood of confusion. W.W.W. Pharm., 984 F.2d at 576. Under New York common law, "the essence of unfair competition is the bad faith misappropriation of the labors and expenditures of another, likely to cause confusion or to deceive purchasers as to the origin of the goods." Jeffrey Milstein, Inc. v. Greger, Lawlor, Roth, Inc., 58 F.3d 27, 34 (2d Cir. 1995) (citation omitted). Accordingly, to prevail on such claims, the plaintiff must also establish the defendant's bad faith. See id. at 35. Because the plaintiff has failed to raise a genuine issue of material fact with respect to likelihood of confusion and bad faith, the plaintiffs' infringement of common law trademark and unfair competition and wrongful misappropriation claims fail.

The plaintiffs' claim of unjust enrichment must also fail. To prevail on a claim for unjust enrichment under New York law the plaintiff must demonstrate that "(1) defendant was enriched;

(2) at the plaintiff's expense, and (3) equity and good conscience militate against permitting defendant to retain what plaintiff is seeking to recover." Diesel Props S.r.l. v. Greystone Bus. Credit II LLC, 631 F.3d 42, 55 (2d Cir. 2001). In the present case, the plaintiffs' claim of unjust enrichment is founded on the alleged infringement of the plaintiffs' ibooks mark. The plaintiffs have not presented evidence from which a jury could reasonably find that the defendant was enriched at plaintiffs' expense under circumstances that, "in equity and good conscience" require the defendant to return the benefit to the plaintiffs. See BigStar Entm't, Inc. v. Next Big Star, Inc., 105 F.Supp.2d 185, 217 (S.D.N.Y. 2000).

CONCLUSION

The defendant's December 21 motion is granted and the plaintiffs' December 21 motion is denied. The Clerk of Court shall enter judgment for the defendant. Dated: New York, New York