IMIG, INC. v. ELECTROLUX HOME CARE PRODUCTS, LTD.

Full title: IMIG, INC., Plaintiff, v. ELECTROLUX HOME CARE PRODUCTS, LTD., Defendant

Court: United States District Court, E.D. New York

Date published: Mar 31, 2008

CV 05-0529 (JO) (E.D.N.Y. Mar. 31, 2008)

Opinion

CV 05-0529 (JO).

March 31, 2008

FINDINGS OF FACT AND CONCLUSIONS OF LAW

JAMES ORENSTEIN, Magistrate Judge

Plaintiff Imig, Inc. ("Imig"), a company that sells commercial vacuum cleaners, initiated this case by suing defendant Electrolux Home Care Products, Ltd. ("Electrolux") for interfering with its business relationships with prospective customers. Docket Entry ("DE") 1. Electrolux responded with counterclaims asserting that Imig, together with its distributor, Nationwide Sales Services, Inc. ("Nationwide") (collectively, "I N"), copied the protected trade dress of certain of its vacuum cleaners as well as the protected contents of its user manuals and that they also made false claims about the respective vacuum cleaners in their advertisements. DE 5. With the parties' consent pursuant to 28 U.S.C. § 636(c), I resolved their cross-motions for summary judgment by dismissing all of Imig's claims, finding that Electrolux had established I N's liability for copyright infringement and false advertising, and concluding that the remainder of Electrolux's counterclaims should proceed to trial. Imig, Inc. v. Electrolux Home Care Products, Ltd., 2007 WL 900310 (E.D.N.Y. Mar. 22, 2007). The parties litigated the remaining issues at a bench trial before me and later filed memoranda of law. Having reviewed the parties' evidence and arguments, I now award Electrolux a total of $281,153 on its false advertising and copyright counterclaims but find that it has not established its counterclaim of trade dress infringement.

I. Findings of Fact

The claims originally asserted in Imig's complaint — in which Imig is the plaintiff, Electrolux the defendant, and Nationwide a non-party — have all been resolved by virtue of my earlier rulings. Accordingly, following the usage adopted at the bench trial, and in recognition of the fact that the only causes of action that remain unresolved are the counterclaims asserted by Electrolux against Imig and Nationwide, I will refer to Electrolux as the "plaintiff" and to Imig and Nationwide as the "defendants." In making my factual findings, I rely on the parties' Stipulated Facts, DE 71 ("SF"); the trial testimony set forth in the Transcript of Trial dated September 10-11, 2007 ("Tr."); the parties' respective exhibits ("PX" for the exhibits with numerical designations submitted by plaintiff Electrolux and "DX" for those with letter designations submitted by defendants I N); the deposition testimony of certain witnesses submitted by the respective parties (each referred to as "[Witness's Last Name] Dep."); and a post-trial stipulation concerning I N's sales, costs, and profits, DE 89. I also refer to the Proposed Findings of Fact and Conclusions of Law submitted by each party. See DE 92 ("I N Prop."); DE 93 ("Electrolux Prop."). I include in this section findings of historical facts about the parties and their conduct. I postpone until the explanation of my conclusions of law certain additional factual findings — such as the extent to which the Sanitaire trade dress on which Electrolux bases its Lanham Act claim should be considered functional — that are inextricably intertwined with the legal determinations I am called upon to make.

A. The Parties

Plaintiff Electrolux is a Texas limited partnership, and the successor of the Eureka Company. It manufactures cleaning products, including vacuums and replacement parts. In addition to vacuum cleaners bearing the Electrolux brand, the company also designs and manufacturers "private label" vacuums that are sold under the brand names of various third parties. Among the commercial vacuums that Electrolux manufactures and sells under its own brand name are certain models known collectively as the "Sanitaires." Two Sanitaires in particular — specifically, models 886 and 899 — are the subjects of this litigation. See SF ¶¶ 1-7, Tr. 39; PX 1-2.¹

1 I invite the reader to "[s]ee" the cited exhibits in only a limited sense. Those exhibits are vacuum cleaners, and have been returned to Electrolux for safe keeping. They were among some 25 such machines that crowded the courtroom (which was, unsurprisingly, left spotless) upon being received in evidence during the bench trial. PX 1-23; DX YY-ZZ. Because the Clerk's office, like nature, abhors a vacuum (or at least 25 of them submitted as exhibits), photographs of the exhibits have been substituted for the actual machines in the court's records.

Defendants Imig and Nationwide are both corporations organized under New York law and co-owned by two brothers — Mark and Scott Genoa (collectively, the "Genoas") — who also serve as officers of both companies. Nationwide has distributed vacuums and other cleaning products since 1996. From 1997 through 2005, Nationwide was a distributor for Electrolux. Imig, formed after Nationwide, imports and sells vacuums and replacement parts; in doing so it employs the services of various distributors, including but not limited to Nationwide. As discussed further below, Imig developed — and with Nationwide's help has advertised and sold — a line of commercial vacuum cleaners known as the "Perfects" to compete with the Sanitaires. Two Perfects in particular — specifically model P101 (designed to compete with the Sanitaire 886) and model P102 (designed to compete with the Sanitaire 899) — are likewise at issue in this litigation. See SF ¶¶ 11-17; Tr. 117; PX 3-4.

B. The Vacuum Cleaners

At the heart of the trade dress infringement claims before me are the physical characteristics of the parties' respective vacuum cleaners. The evidence demonstrated several similarities between those machines as well as certain differences. Both are upright commercial vacuum cleaners, and both display the words "HEAVY DUTY COMMERCIAL" in white lettering (with similar fonts) in the same position on a hanging bag. The Sanitaires have red bags, while the Perfects have bags that are black (or, in newer models, grey). Likewise, the two brands have base assemblies that appear similar but for coloring: the base assembly on the Sanitaires is red with a chrome hood, into which motor vents are cut, while the assembly on the Perfects is similarly configured but black or grey in color. Both brands have a round knob affixed to a black plate at the front of the base assembly, with lettering on the knob and the plate — with one brand using the legend "Sanitaire" in white lettering and the other using "PERFECT" in black lettering near a small United States flag. The base assemblies of both machines have similar configurations of knobs for adjusting the height of the vacuum and power switches, and in both the chrome hood is raised and set back a few inches from the front of the machine; however, the hood on the Perfects' base assembly is slightly lower toward the front and has a raised portion in the middle of the rear.

The incidence of physical similarities between the Perfects and the Sanitaires is no accident. The Genoas developed the Perfects for Imig in late 2002, as a result of their concern that their other company, Nationwide, would lose its position as a distributor for Electrolux. After abortive initial discussions with one manufacturer, the Genoas began working to develop their own product with a man located in China named Xu Shihui ("Xu"), who in turn dealt with Chinese manufacturers on their behalf. Rather than draw up design documents on their own, the Genoas simply directed the Chinese manufacturer to refer to the specifications of the Sanitaires. In designing their products, the Genoas specifically intended to create a vacuum the parts of which would be interchangeable with Sanitaire parts — and with one exception (a "brush roll" that is not interchangeable with the corresponding part of the Sanitaires), they accomplished that goal. The Genoas also incorporated into the Perfects certain elements not found in the Sanitaires that they admired in other vacuum models. See Tr. 126, 141-45, 163-67, 171, 174, 364, 398; SF ¶¶ 26-27; PX 11, 12, 15.

Further bolstering the inference of intentional copying is evidence of the steps the Genoas took to avoid any claim of patent infringement. In 2003, Mark Genoa first met with attorney Len Holtz ("Holtz") to determine if the Perfects would infringe on any Electrolux patents. Holtz later sent a copy of the Owner's Guide for the Sanitaires to a patent research company along with a cover letter in which he wrote of his client's desire "to make a private label vacuum cleaner that is virtually identical in appearance (and perhaps also mechanically identical) to the Eureka vacuum cleaners shown the enclosures." SF ¶ 29. During the next year, Mark Genoa met with Holtz again to discuss possible patent infringement problems with the Perfects. See Tr. 126-28.

C. The Owner's Guides

The Genoas first sold the Perfects in October 2004. Each of the first 425 such machines was sold with an Owner's Guide that was a copy of the corresponding Owner's Guide for the Sanitaires — for which Electrolux holds the pertinent copyrights. See SF ¶¶ 14, 33-35; Imig, Inc., 2007 WL 900310, at *15-17 (finding I N in violation of the Copyright Act of 1974). Electrolux spent approximately $9,000 to develop the copyrighted work for its own use. See Tr. 89-91.

D. I N's False Advertising

The Genoas ran advertisements for the Perfects for a period of four months beginning on April 29, 2005, that touted the Perfects as having three times as much motor life and 30 percent more strength compared to their "nearest competitors" and that also asserted that the Perfects had a motor strength of 9.5 amperes. PX 27, 39; SF ¶¶ 50-52. The advertisements bore a legend indicating the Perfects had been tested by ETL, a company that performs tests on machinery to determine compliance with the relevant industry standards. SF ¶¶ 56, 58. A later advertisement, that ran in the August issue of the Vacuum Sewing Dealers Trade Association magazine, explicitly compared the Perfects to the Sanitaires with the same claims of motor strength, amperage, and ETL testing. Tr. 189; PX 43. As I have previously determined, these advertising claims were literally false. See Imig, Inc., 2007 WL 900310, at *13-15 (granting Electrolux's motion for summary judgment to the extent of finding I N liable for violating the Lanham Act's prohibition against false advertising).

The trial testimony fleshed out the extent to which the advertising was false. The Chinese manufacturer of the Perfects conducted various tests on the vacuums and worked with ETL to determine the machines' specifications. Tr. 189-209. ETL did not test motor life nor did it compare the Perfects' performance to the Sanitaires'. Rather, the Chinese manufacturer conducted some comparison tests. In the only such test that Scott Genoa observed, four Sanitaires were turned on and allowed to run continuously; one failed after 368 hours, and the three others lasted longer, with the longest-lasting Sanitaire motor failing after 510 hours. Scott Genoa claimed that a Perfect vacuum subjected to such testing had lasted 1,050 hours when the test was suspended (although there was no documentation of that claim). He therefore divided 1,050 (the assumed motor life of the Perfect) by 368 (the worst result of the four Sanitaires that were tested), and — notwithstanding the fact that the quotient resulting from such a patently unreliable method is only 2.85 — decided to claim that the Perfects have triple the motor life of the Sanitaires. Tr. 347-48, 394-97; SF ¶¶ 76; 77.

The record is less clear with respect to the advertised claim that ETL had tested the Perfects and determined them to have a motor strength of 9.5 amperes, but it nevertheless leads me to conclude that the claim was knowingly false. In September 2004, the Perfects' Chinese manufacturer sent documents to ETL reflecting that the vacuum had a motor strength of only 7.4 amperes. Tr. 206; PX 47-48. In a later letter, dated October 14, 2004, ETL asked Mark Genoa to provide certain "missing" information about the Perfects, explicitly including their amperage. PX 45 at 4; SF ¶ 80. The obvious inference I draw from that correspondence is that as of the date of the letter, ETL had not already determined the Perfects' amperage by performing its own test. See PX 45 at 4. In addition, ETL employee Karim Abu-Hashim ("Abu-Hashim") testified in a deposition that was submitted in evidence that to the extent ETL eventually reported an amperage rating of 9.5, it relied exclusively on information provided by Imig. See e.g. Abu-Hashim Dep. at 111; see also id. at 78, 86. As both Abu-Hashim and his ETL colleague Ashruf Matar explained, ETL sought data from the manufacturer and then performed tests to verify it. Id. at 88; see also Matar Dep. at 132. However, no evidence from ETL suggests that the company ever verified that the Perfect vacuums had a motor strength of 9.5 amperes. See Abu-Hashim Dep. at 89, 110; Matar Dep. at 63-66.

I N nevertheless contend that the advertised claims about ETL testing were true when made. In support of that proposition they rely on Mark Genoa's testimony about a telephone conversation that he claimed to have had with an ETL employee (whose name he did not offer, and for which no party's counsel asked) some time before October 14, 2004 — the date of the ETL letter described above. Mark Genoa swore that the employee had orally assured him during this conversation that ETL had tested the Perfect's motor strength and rated it at 9.5 amperes. Tr. 277-78; see also PX 34 ¶ 4 (Declaration of Mark Genoa) ("I decided to place the rating of 9.5 amps on the rate plate for the P 101 and P 102 machines after the rating of 9.5 amps was given to us by ETL as a result of their official evaluation of the P 101 and P 102 machines.").

I did not find Mark Genoa's testimony on this matter to be credible. It is at odds with the clear import of documents created at the time of the events he purported to recall. Moreover, the testimony that ETL would communicate its testing result solely by telephone, or that I N would rely on such information without seeking to obtain it in writing, strikes me as implausible. Finally, although I did not see Abu-Hashim testify and therefore cannot assess his demeanor, his apparent lack of any bias, combined with my observation of Mark Genoa's demeanor in testifying to his version of the relevant events, suffices to persuade me that the claim about ETL testing that I N included in its advertising was knowingly false.²

2 I note that Abu-Hashim acknowledged that ETL made a mistake by not objecting when Imig submitted to ETL for approval a "rating plate" (a plate affixed to each vacuum that lists such specifications as motor strength) that indicated a motor strength of 9.5 amperes. Abu-Hashim Dep. At 119-20. In light of the credible evidence that ETL never actually conducted its own test that confirmed such a rating, the fact that Imig managed to get ETL to approve a rate plate bearing unverified information that Imig itself had supplied supports neither Mark Genoa's assertion that the information on the rating plate was placed there on the basis of ETL's "evaluation" nor the more pertinent assertion that I N's advertising claims concerning the vacuum's motor strength were true.

E. I N's Profits

In calculating Electrolux's damages on certain claims as to which I previously found I N liable, it is necessary to determine I N's profits and costs. My findings as to those amounts, as set forth below, are based on certain reports that Electrolux submitted in evidence subject to certain adjustments to which the parties have stipulated. See PX 57, 64, 65, 70; DE 89.

1. The First 400 Perfects

The Genoas first sold the Perfects in October 2004. See Tr. 134; SF ¶¶ 14. In its claim for damages, Electrolux seeks the profits associated with only 400 of those first 425 units, all of which were Perfect P101 models, and none of which was sold to Nationwide. Electrolux Prop. at 41, 64-65; PX 30.³ I therefore focus on those first 400 sales in the calculations below.

3 Electrolux presumably seeks the profits of the first 400 Perfects, rather than the entire initial shipment of 425 vacuums based on the evidence that Imig did not apparently sell the entire shipment. See Tr. 136.

Although the evidence does not reveal the specific selling price for the first 400 vacuums, it does reveal that Imig's average selling price for the year before September 2005 (which included the first 400 sales in October 2004) was $98.53 per vacuum. Imig's costs during the same period totaled approximately $68.89 per vacuum, consisting of the purchase price of $56.89 per unit,⁴ operating costs of approximately $10.00 per unit, and tooling costs of approximately $2.00 per unit. PX 64 at Schedule 4, lines 1-4. Multiplying the unit price and costs by 400 produces the following results: total revenues from sales of $39,412, and total costs of $27,556.00, for a total profit (before returns) of $11,856.00. I find the latter amount to be the measure of Imig's profit from its sale of the first 400 Perfects (each of which was sold with an owner's guide that infringed Electrolux's copyright).⁵ As explained further below in the conclusions of law, however, while that amount measures the profit from the sale of the first 400 vacuums, it does not measure the profit from Imig's inclusion of the misappropriated Owner's Guide with each such sale. The latter profit is equivalent to the money that Electrolux spent producing the manual — an amount Imig did not have to spend (and thereby reduce its profit on the sales of vacuums) by virtue of its copyright infringement.

4 This is also referred to in Gallagher's reports as the "FOB cost" presumably referring to the term "free on board" meaning the seller bears the costs of transportation of the goods. See Black's Law Dictionary 690-91 (8th ed. 2004) (defining "free on board").

5 The profit level that I find accords with the parties' respective submissions. I note, however, that the parties agree that the first 425 vacuums that Imig sold suffered from significant quality problems, as a result of which Imig was obliged to accepted returns at a rate of 40 percent. SF ¶¶ 30-31. I assume — but the record does not establish — that Imig reimbursed to its customers the purchase price for all such defective vacuums. If that is the case, then a more accurate measure of Imig's profits for this period would be to subtract from the amount calculated above the money that it had to reimburse for such returns. A 40 percent return rate means that 160 of the first 400 sales resulted in returns; multiplying the sales price of $98.53 by that number produces a reimbursement amount of $15,764. Subtracting that amount from the amount calculated above results in a net loss of $3,908.80 associated with the first 400 sales. If the record established a factual basis for this calculation — that is, if it demonstrated that Imig refunded the purchase price of defective vacuums — and if Imig had argued the matter, I would have a basis to find that Imig had no profits from its first 400 Perfect sales. Under the circumstances, however, with neither party having addressed the impact of the returns on the issue of profits for this period, and because it is I N's burden "to prove [its] deductible expenses," 17 U.S.C. § 504(b), I defer to the parties' stipulation on the matter.

2. April 29, 2005 through August 31, 2005

The parties have stipulated to profit calculations that break down I N's profits for the following discrete periods: October 1, 2004 — September 30, 2005; October 1, 2004 — February 28, 2006; October 1, 2004 — June 30, 2007; and the calendar years of 2004 and 2005. See PX 57, 64, 65, 70. Attached to the First Supplemental Expert Report of Mark Gallagher are the raw sales and cost data for the P101 and P102 models sold between April 29, 2005 and August 31, 2005. See PX 65. Having analyzed the data applying the methods used in those reports, I make the following findings with respect to the defendants' profits during the period in which the false advertisements were published. Nationwide had net sales during the period of $90,242, and net costs of $60,920, resulting in an unadjusted profit of $29,952. During the same period, Imig had net sales of $831,165 and net costs of $560,767, resulting in an unadjusted profit of $270,398. The unadjusted combined profit for the two companies during the period of the false advertisements is thus $300,350. In accord with the parties' agreement, I have adjusted that combined profit by subtracting both companies' overhead expenses and as well as an amount based on Imig's amortization of tooling costs.⁶ As a result of that calculation, I find that I N's total profit during the months in which the false advertisements were published was $272,153.

6 The stipulation provides that $7,139.02 should be deducted for each calendar quarter starting October 1, 2004. Adjusting that rate for a four month period results in a reduction of Imig's net sales by $9,518.69.

II. Conclusions of Law

A. Trade Dress Infringement

Section 43(a) of the Lanham Act prohibits the use of "any word, term, name, symbol, or device, or any combination thereof" in a manner that is "likely to cause confusion . . . as to the origin, sponsorship, or approval" of goods. 15 U.S.C. § 1125(a). The Lanham Act's regime of protection for trademarks and trade dress serves an insulating function, protecting manufacturers of socially desirable products from the kind of imitation that "capitaliz[es] on a consumer's inability quickly to evaluate the quality of an item offered for sale." Qualitex Co. v. Jacobson Prods. Co., Inc., 514 U.S. 159, 163 (1995). This aim must be balanced against the social benefits that can flow from some forms of imitation — for example, the sort that operates to prevent an innovator from obtaining a monopoly over features that improve the product for practical reasons — and thus, such protection should be granted only where it will promote competition. TrafFix Devices, Inc. v. Marketing Displays, Inc., 532 U.S. 23, 28-29 (2001). Invitations to extend trade dress protection to the specific configuration of a product's features must be viewed with special caution because of the risk presented: protecting ordinary product designs would be tantamount to creating a monopoly in the actual goods. Wal-Mart Stores, Inc. v. Samara Bros., 529 U.S. 205, 213 (2000).

To prevail on a claim of trade dress protection of a product's design, a party must prove first "that the mark is distinctive as to the source of the good," and second "that there is a likelihood of confusion between its good and the defendant's." Yurman Design, Inc. v. PAJ, Inc., 262 F.3d 101, 115 (2d Cir. 2001). To be deemed differentiable from goods produced by other manufacturers, and therefore "distinctive," a product design must be nonfunctional and must contain characteristics that, over time, come "to identify and distinguish the goods — i.e., to indicate their source[.]" Qualitex Co., 514 U.S. at 163 (internal quotation marks omitted). Functionality is statutorily presumed; a party seeking trade dress protection under the Lanham Act thus bears "the burden of proving that the matter sought to be protected is not functional." 15 U.S.C. § 1123(a)(3). To do so, the party must show that the elements that purportedly comprise the trade dress are not "essential to the use or purpose of the article" and do not "affect the cost or quality of the article[.]" TrafFix Devices, Inc., 532 U.S. at 32-33 (quoting Qualitex, 514 U.S. at 165); see Inwood Labs., Inc. v. Ives Labes., Inc., 456 U.S. 844, 850 n. 10 (1982). Moreover, to the extent that the primary purpose of any feature is aesthetic, the design nonetheless will be deemed functional "if the right to use it exclusively would put competitors at a significant non-reputation-related disadvantage." Yurman, 262 F.3d at 116 (quoting TrafFix Devices, Inc., 532 U.S. at 32-33) (internal quotation marks omitted); see Qualitex, 514 U.S. at 170; Eliya, Inc. v. Kohl's Dept. Stores, 2006 WL 2645196, at *3 (S.D.N.Y. Sept. 13, 2006).

The behavior of other participants in the pertinent market is sometimes instructive; the existence of any alternative product designs is relevant to assessing whether a feature affects cost or quality, although decidedly not relevant after any finding of functionality. See, e.g., Major League Baseball Props., Inc. v. Salvino, Inc., 420 F. Supp. 2d 212, 222 (S.D.N.Y. 2006). Once deemed functional, a product's design is not eligible for trade dress protection and any secondary meaning the design might have acquired is irrelevant. TrafFix Devices, Inc., 532 U.S. at 33. If the court deems a product's design to be nonfunctional, however, its inquiry turns to assessing whether the design possesses "secondary meaning" — in other words, whether the particular configuration of features has come, over time, to identify the source of the product in the minds of consumers. Qualitex, 514 U.S. at 163 (citing Inwood Labs. Inc., 456 U.S. at 856 n. 11).

In addition to establishing the distinctiveness of a product design, a party asserting a trade dress claim under the Lanham Act must also demonstrate that the allegedly infringing product is likely to confuse consumers about its source or identity. Nora Beverages, Inc. v. Perrier Group of America, Inc., 269 F.3d 114, 118-19 (2d Cir. 2001). The claimant's burden in this regard is to demonstrate a "probability" — not merely a "possibility" — of confusion. Cadbury Beverages, Inc. v. Cott Corp., 73 F.3d 474, 482 (2d Cir. 1996). Critical to such an analysis are the eight so-called "Polaroid factors." See Polaroid Corp. v. Polaroid Elecs. Corp., 287 F.2d 492, 495 (2d Cir. 1961). Those factors are

(1) strength of the plaintiff's trade dress; (2) similarity of the trade dresses; (3) proximity of the products in the marketplace; (4) likelihood that the plaintiff will bridge the gap between the products (enter a market related to that in which the defendant sells its product); (5) evidence of actual confusion; (6) the defendant's bad faith; (7) quality of the defendant's product; and (8) sophistication of the relevant consumer group.

Natural Organics, Inc. v. Nutraceutical Corp., 426 F.3d 576, 578 (2d Cir. 2005) (quoting Playtex Prods., Inc. v. Georgia-Pacific Corp., 390 F.3d 158, 162 (2d Cir. 2004)). The court's task is to weigh these enumerated factors with an eye to the ultimate question: whether, taken as a whole, there is a likelihood of consumer confusion with respect to the protected product and the product charged with unlawful infringement. Nora Beverages, Inc., 269 F.3d at 119.

For the reasons set forth below, I conclude that Electrolux has failed to establish its claim for trade dress infringement with respect to the design of the Sanitaires. Specifically, I find that Electrolux has not proved either that the trade dress it seeks to protect is nonfunctional, that it has acquired secondary meaning, or that the accused Perfects are so similar to the Sanitaires as to create the probability of confusion.

1. Distinctiveness Of The Product Design

At the heart of any trade dress dispute is the specific design that the claimant seeks to protect. I therefore begin my analysis by taking note of the specific design at issue here:

Electrolux alleges that its trade dress for [each Sanitaire] vacuum [at issue in this case] consists of (i) the shape of the base assembly; (ii) the use of the colors red, black and white on the base assembly and the bag; (iii) the shape, color and location of the on/off switch; (iv) the location, color scheme, and dial position settings for the height adjuster; (v) the shape, size, and chrome finish of the hood and the location thereon of the motor vents; and (vi) the font, writing, and position of the writing on the bag.

SF ¶ 19; see id. ¶ 20.⁷ It is this specific trade dress, consisting of all six specified elements, that Electrolux must prove to be distinctive. TrafFix Devices, Inc., 532 U.S. at 33 (citing Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 774 (1992)).

7 Electrolux uses identical language to describe its claimed trade dress as to each of the Sanitaires at issue. SF ¶¶ 19-20. The parties agree that with respect to trade dress, the two Sanitaire models involved in this case "are substantially identical, the principal difference resulting from the fact that" one model has a wider brush than the other. Id. ¶ 21.

a. Functionality

There are two ways in which the evidence can show that a design is functional and therefore not sufficiently distinctive to warrant protection under the Lanham Act. First, the evidence may demonstrate that a product's design is essential to the product's use or purpose; second, the evidence may reveal that the design itself affects the product's cost or quality. TrafFix Devices, Inc., 532 U.S. at 32-33. Because the parties agree that the trade dress Electrolux describes for the Sanitaires is not essential to the vacuums' use or purpose, SF 22, I confine my analysis of functionality to the question of cost and quality. In doing so, I consider the design as a whole, and not whether any specific element of the claimed trade dress might affect the Sanitaires' cost or quality. TrafFix Devices, Inc., 532 U.S. at 32-33.

The eligibility of the Sanitaires' design for Lanham Act protection thus turns on whether Electrolux has demonstrated that the particular combination of six features has no measurable effect on the cost or quality of the vacuums. In seeking to do so, Electrolux relied on the testimony of its long-time employee Brad Hoare ("Hoare"). See Tr. 28-112. Hoare testified that each of the six features of the claimed trade dress has no effect on the cost or quality of the Sanitaires. Tr. 58 (base assembly); Tr. 61-62 (colors); Tr. 63-66 (on/off switch); Tr. 67-69 (height adjuster); Tr. 70-73 (hood and vents); Tr. 74-75 (writing).

Beyond general statements that the features comprising the trade dress individually have no effect on the cost or quality of the Sanitaires, Hoare relied on the existence of alternative designs to bolster his conclusions. Evidence of alternative designs may, in some limited circumstances, be relevant to the issue of functionality, but such evidence alone generally does not suffice to overcome a finding of functionality that would otherwise be warranted. See General Motors Corp. v. Lanard Toys, Inc., 468 F.3d 405, 416 (6th Cir. 2006) ("If a trade dress is found to be functional, the mere fact that there are other non-infringing designs which would serve the same functional purpose is no defense to functionality."). In making such an argument, Electrolux — as the party seeking Lanham Act protection — bears the burden of demonstrating that its product design has a source-identifying function. See Maharishi Hardy Blechman Ltd. v. Abercrombie Fitch Co., 292 F. Supp. 2d 535, 549 (S.D.N.Y. 2003) (citing Knitwaves, Inc. v. Lollytogs Ltd. (Inc.), 7 F.3d 996, 1008-09 (2d Cir. 1995)).

In adducing evidence of alternative designs, Electrolux demonstrated that it could have chosen a design that would produce a cheaper and more efficient vacuum. See Tr. 58-59, 62, 66, 68, 71. For example, instead of using a chrome hood on the base assembly, Electrolux could have produced a more durable and less expensive product by using a plastic hood. Tr. 59. I can fairly draw an inference from Electrolux's choice to use chrome for the hood that it viewed as worthwhile the trade-off of cost and durability for what it deemed to be a more distinctive product design. But such an inference does not inevitably lead to the conclusion that the design of the Sanitaire hood, when combined with the other elements, is non-functional.

The record before me suffices to support a finding that the Sanitaire trade dress does affect the quality and cost of the vacuums. For example, the shape of the base assembly hood has a low profile that allows the vacuum to reach under furniture. Tr. 100; see also Hoare Dep. at 81-82 (explaining that the low hood is a "desirable" feature although not necessarily "the most desirable available in other products"). The prominent display on the bag of the term "heavy duty commercial," which is used commonly throughout the industry, suggests a descriptive or identifying function. Tr. 256, 372; PX 13-22. The use of a dial for the height adjuster is less expensive than the alternative of using a slide lever, Tr. 69, and the location in the front of the base is viewed as a "useful" feature. Hoare Dep. at 82-83. Moreover, Electrolux itself has previously advertised various features of its trade dress as having utility for its prospective customers. See Tr. 98-101; DX M-R. In short, Electrolux — which bears the burden of disproving functionality — has failed to persuade me by a preponderance of the evidence that its trade dress is non-functional. That finding alone suffices to defeat the trade dress claim.

b. Secondary Meaning

I further conclude that Electrolux has failed to establish its Lanham Act claim, regardless of the functionality of the Sanitaire design, because it has not sufficiently demonstrated that the trade dress it seeks to protect has acquired a secondary meaning that over time has come to identify the source of its vacuums in the minds of consumers. Secondary meaning turns on whether "the public is moved to consume a product because of its source." PaperCutter, Inc. v. Fay's Drug Co., 900 F.2d 558, 564 (2d Cir. 1990). A court considering secondary meaning should take into account six factors that bear on the extent to which the "consuming public for a particular product" associates that product with its source on the basis of its trade dress: "(1) advertising expenditures, (2) consumer studies linking the mark to a source, (3) unsolicited media coverage of the product, (4) sales success, (5) attempts to plagiarize the mark, and (6) length and exclusivity of the mark's use." Mana Prods., Inc. v. Columbia Cosmetics Mfg., Inc., 65 F.3d 1063, 1071 (2d Cir. 1995) (quoting Centaur Comms., Ltd. v. A/S/M Comms., Inc., 830 F.2d 1217, 1222 (2d Cir. 1987) (internal quotation marks omitted)). This is an "essentially factual determination, proof of which entails vigorous evidentiary requirements." Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1041 (2d Cir. 1992) (internal quotation marks and citation omitted).

In seeking to establish secondary meaning, Electrolux focused its efforts almost exclusively on the second factor listed above. Such consumer surveys are a probative source of evidence as to secondary meaning, but they are not conclusive. Harlequin Enterprises Ltd. v. Gulf Western Corp., 644 F.2d 946, 949-50 n. 2 (2d Cir. 1981) (finding evidence of secondary meaning in a 500-person survey in which 50 percent of respondents identified product source based on trade dress). And for the reasons explained below, I conclude that Electrolux's reliance on such evidence in this case is misplaced.

As it did at the summary judgment stage, Electrolux submitted the results of a survey of 157 participants — supplemented at trial with the opinion testimony of Hal Poret ("Poret") — to support its contention that the Sanitaire trade dress has come to possess secondary meaning. Tr. 400-458; PX 80 (Report); PX 81 (Survey). Poret, an expert in "consumer perception, research in general and trademark," Tr. 404, conducted the survey and concluded the Sanitaire had secondary meaning, based on the fact that 75.2 percent of participants in the survey's test group correctly identified the image of the upright vacuum they were shown as a Sanitaire, compared to 32.9 percent of a control group who incorrectly identified as a Sanitaire the image of a rather different-looking Hoover that they were shown. Using the control group results as an indication of "noise," Poret calculated the difference in percentages and concluded that 42.3 percent of the test group respondents based their correct identifications on the Sanitaire trade dress. Tr. 421-22.

I did not find Poret's testimony or the results of his survey persuasive on the issue of secondary meaning. Despite the various controls used to reduce "noise" in the survey results, any reliable association between the number of correct Sanitaire identifications among survey participants and the specific trade dress that Electrolux seeks to protect in this litigation is a matter of speculation. For example, of the 118 respondents who correctly identified an image as being that of a Sanitaire or Electrolux, 114 also identified at least one design feature that they believed supported the identification. Tr. 449. Between them, those 114 respondents mentioned all six of the features on which Electrolux relies, but not one mentioned all six — in other words, not a single respondent explained his identification of the Sanitaire by reference to the specific trade dress that Electrolux seeks to protect. Further, approximately 75 respondents relied solely on one or more trade dress features to identify the vacuum, meaning 39 respondents identified the vacuum by a trade dress feature in addition to a non-trade dress related characteristic. Tr. 449. The fact that many respondents relied on characteristics other than the trade dress at issue further decreases the study's persuasive value for me as a fact-finder. As for the features identified, more than two thirds of the 114 respondents who specified an identifiable feature of what they recognized to be a Sanitaire (or Electrolux) mentioned its red color,⁸ another 36 percent specified the chrome hood, and smaller percentages mentioned the remaining four features. Tr. 424-25; PX 80 at 17 (Report).

8 Although not pertinent for purposes of assessing secondary meaning, I note that this identification of the Sanitaire with its red coloring — the most prevalent reason given in the survey for identifying the vacuum's brand — is also informative with respect to the likelihood of confusion in light of the fact that the accused Perfect vacuums are not red. Of greater relevance to the question of secondary meaning is the fact that, as Hoare acknowledged, Electrolux refers (at least internally) to the Sanitaires it now seeks to protect as the "Red Sanitaire." Tr. 105; Miller Dep. at 10 (testimony that the term "Red Sanitaire" is used within Electrolux, by its customers, and by "the trade" to refer to the 886 and 899 models). There is simply insufficient evidence for me to determine whether a significant number of the respondents who correctly identified the Sanitaire images in Poret's study did so simply because they saw a red upright vacuum cleaner, or because they saw a red vacuum with some additional distinctive features that are not part of the trade dress for which Electrolux seeks protection, or because — as Electrolux seeks to establish but has failed persuasively to do — the specific trade dress features that Electrolux has defined have become distinctive as to source in the minds of consumers.

These numbers persuasively demonstrate that some people reliably recognize two dominant features and attribute them to Sanitaire or Electrolux. But the data do not go further to persuade me, in my capacity as a fact-finder, that the combination of six features for which Electrolux seeks protection has acquired a secondary meaning. The latter proposition may be more difficult to establish by means of a consumer survey, but such difficulty does not relieve Electrolux of its burden of persuasion.⁹ Nor does it do anything to undermine the availability of other sources of proof — including such things as advertising expenditures, and unsolicited media coverage of the product — that Electrolux could have sought to adduce but did not.

9 In the latter regard, I note that there were methods that Electrolux could have employed in conducting its survey that might have more persuasively suggested that positive responses were proof that the trade dress at issue has acquired secondary meaning. For example, as an alternative or additional control, Poret could have shown survey participants any of several vacuums that look more like the Sanitaire than the very dissimilar Hoover, and yet are not accused of infringing the Sanitaire trade dress. Such a comparison might provide useful information about the extent to which those in the test group recognized the trade dress itself, rather than some other combination of design features for which Electrolux does not claim statutory protection. Also, rather than relying on photographs and written questions to conduct the test, Poret could — undoubtedly at greater expense to his client, but with greater likelihood of producing a persuasive result — have shown his test subjects physical specimens and conducted oral interviews to more precisely gauge the basis for their responses. See Tr. 442-47. I hasten to make clear that I do not view any such alternative testing techniques to be legally required; I merely note that the data on which Electrolux relied did not persuade me as a factual matter, and that other techniques might have produced more compelling results in the context of this case.

To be sure, although Electrolux relied heavily on its consumer survey, it did not entirely ignore other methods of proving secondary meaning such as sales success, attempts to plagiarize, and the length and exclusivity of its use of the trade dress it seeks to protect. The Sanitaire sales figures to which the parties have stipulated, SF ¶¶ 8-10, are informative but do little to persuade me — particularly in the absence of any information to put the evidence in context — that the Sanitaire trade dress has acquired secondary meaning. For example, the sales evidence relates to a period from 2000 through the middle of 2007, during which time the sales levels remained largely static. But Electrolux claims (without, I note, citation to any supporting evidence in the record) that "since its introduction in 1984, Electrolux has used the trade dress exclusively." Electrolux Prop. at 52. Assuming that to be the case, and assuming that the trade dress has acquired a secondary meaning, I would find it persuasive if Electrolux had demonstrated that Sanitaire sales had increased over the decades that its trade dress has filtered into the public consciousness. But in the absence of such evidence, the sales figures in the record convey little of value.

Finally, the evidence that both Imig and another company called Nilfisk created products that are similar in appearance to the Sanitaire has some — albeit limited — value. The evidence easily establishes that Imig not only created a similar-looking product, but that it did so intentionally. See SF ¶ 29 (letter from I N's counsel explaining that his "client wishes to make a private label vacuum cleaner that is virtually identical in appearance" to the Sanitaire). That evidence supports an inference that the Genoas, at least, considered the design features that they copied to have value in the market. With respect to Nilfisk, the evidence shows only that the company created a product, called the "Relia Vac," that is somewhat similar in appearance to the Sanitaire — but whether it did so intentionally, or for what purpose, is a matter of conjecture based on the record before me. See Tr. 50. To the extent that the test of secondary meaning is an association that "the public" makes between a set of design features and a product's source, see PaperCutter, Inc., 900 F.2d at 564; Mana Prods., Inc., 65 F.3d at 1071, the fact that one competitor can be viewed as making such an association by virtue of its attempt to plagiarize (or perhaps that two competitors can be so viewed) is something that unquestionably supports Electrolux's claim, but only to a point. Moreover, as discussed below in the context of assessing the likelihood of confusion, both the Perfects and the Relia Vac share some physical features with the Sanitaires, but so too do they resemble — and to my eye, more closely — other "private label" brands of vacuum that Electrolux produces but for which it does not seek trade dress protection. See PX 3-4 (Perfects), DX YY (Relia Vac); PX 16-22 (various private label models labeled Powr-Flite, Triple S, Tornado, and Nobles/Castex). What that tells me is that while the Genoas were undoubtedly trying to copy the overall look of a presumably successful competitor, it is not clear that the look they were trying to copy was one uniquely associated — either in their minds or the public's — with the Sanitaire trade dress at issue in this case. Briefly stated, because both Imig and Nilfisk made vacuums that could just as easily be described as knock-offs of the private label Electroluxes as of the Sanitaires, I do not conclude as a factual matter that such copying persuasively demonstrates that the Sanitaire trade dress at issue here — as opposed to the trade dress of the private label Electroluxes — has acquired secondary meaning.

It is entirely plausible that the Sanitaire trade dress has acquired a secondary meaning, and that Electrolux will be able to establish that fact in litigating similar claims in the future. But based on my review of the record in this case and the findings of fact that I make on the basis of that evidence, I cannot conclude that the trade dress at issue here has acquired secondary meaning. Accordingly, I must conclude that the Sanitaires are not distinctive and therefore not subject to protection under Section 43(a) of the Lanham Act.

2. Likelihood Of Confusion

In the interest of a complete record, I now turn to the second element of the Lanham Act claim despite the fact that Electrolux has failed to establish the first. As noted earlier, analysis of the likelihood of confusion requires a consideration of the eight Polaroid factors, none of which need be dispositive but any one of which may be. See Arrow Fastener Co. v. Stanley Works, 59 F.3d 384, 400 (2d Cir. 1995). The organizing principle for such an inquiry is whether the party requesting protection has shown a probability that consumers will be confused as to the source of the purportedly infringing product. Nora Beverages, Inc., 269 F.3d at 121. I discuss each factor in turn recognizing that the boundaries between them are somewhat porous, and that in the circumstances of a particular case the same facts may be pertinent to several different factors.

a. Trade Dress Strength

The "strength of a particular [trade dress] encompasses the totality of the product in the relevant commercial context." Nora Beverages, Inc., 269 F.3d at 122. A trade dress is considered strong if it possesses characteristics that closely associate the product with a specific source in the eyes of the public. Sports Auth. v. Prime Hospitality Corp., 89 F.3d 955, 960-61 (2d Cir. 1996). However, a product's manufacturer can dissipate the strength of its trade dress by authorizing others to use it. Nora Beverages, Inc., 269 F.3d at 123 (citing Versa Prods. Co. v. Bifold Co. (Mfg.) Ltd., 50 F.3d 189, 216 (3d Cir. 1995) (private labeling undermines strength of mark)). That principle is of significant weight under the circumstances of this case, as the evidence showed that Electrolux has authorized a number of companies to sell private label vacuums that incorporate several, but not all, of the trade dress features that comprise the Sanitaire's design.

During the trial of this case, the well of the courtroom was filled with dozens of stand-up vacuum cleaners, including several private label brands manufactured by Electrolux that, with the latter's permission, incorporated many design characteristics similar to the Sanitaires. PX 16-22. Nearly all of the private label models has a bag assembly similar in shape to the Sanitaire's and corresponding to the first of the six elements of the Sanitaire trade dress that Electrolux now seeks to protect.¹⁰ One replicates the third element of the claimed trade dress in that it has an on/off switch of the same shape and location as the Sanitaire.¹¹ All of them replicate the fourth element by incorporating a height adjustment dial in the same location as the Sanitaire,¹² several replicate the fifth element to some degree by incorporating a base assembly hood that is either chrome plated or of a similar shape to the Sanitaire,¹³ and half of the private labels replicate the sixth element with identical or substantially similar printing on the vacuum bag.¹⁴ Indeed — and most likely not coincidentally — the sole trade dress element that Electrolux ascribes to its Sanitaire models that it does not permit any of the private label models to incorporate is the striking red color.

10 All of the private label models except the Powr-Flite PF 70DC and Powr-Flite PF 50 use an upright, narrow bag topped with a metallic trim. The former incorporates a "dirt cup" rather than a full bag while the latter has a fuller bag but no metal band at the top. See PX 17, 18.

11 The Powr-Flite PF 757 has the same type of metal button as the Sanitaire, and in the same location. PX 16.

12 All seven private label models use a round dial to adjust height located at the front, center of the base assembly. See PX 16-22.

13 The Powr-Flite PF 757, PX 16, has a chrome hood of a nearly identical shape to the Sanitaires. While none of the other private labels in evidence incorporate the chrome hood, the rest of the Powr-Flite series, the Triple S HD 100, and the Nobles/Castex TV 1200 feature a raised section above the dial similar to the Sanitaires, as opposed to the more bulbous shape of the Tornado hood. See PX 17-22.

14 The Powr-Flite PF 70DC, PF 50, and Tornado Duo Carpet Keeper feature "HEAVY DUTY COMMERCIAL" in font and location on the bag that is identical to the Sanitaires. PX 17; 18; 21. The Nobles/Castex TV 1200 has an identical "COMMERCIAL" label without the descriptive "HEAVY DUTY." PX 22.

After reviewing the evidence, I was left with the firm impression that Electrolux has populated the market with several different brands of vacuum that bear a strong resemblance to one another. In doing so — and in particular by authorizing a variety of combinations of its claimed trade dress elements aside from the color red — Electrolux has significantly strengthened the association of the red color with the Sanitaire brand but at the same time significantly weakened the association between that brand and the specific combination of design elements for which it seeks trade dress protection. In other words, if not for the mix-and-match approach that Electrolux has taken in designing the look of the various private label vacuums, I would find its argument on the "strength" factor more persuasive. Under all of the circumstances, the best that Electrolux can show as to this factor is that it is in equipoise, and neither helps nor harms its contention that the Perfects create a likelihood of confusion.

b. Similarity

In considering the similarity between the Sanitaire trade dress and that of the Perfect models, I "must analyze the mark's overall impression on a consumer, considering the context in which the marks are displayed and the totality of factors that could cause confusion among prospective purchasers." Louis Vuitton Malletier v. Dooney Bourke, Inc., 454 F.3d 108, 117 (2d Cir. 2006) (internal quotation marks and citations omitted). Side-by-side comparison of the trade dresses is permissible, but not to the exclusion of considering the "overall impression" that consumers are likely to have of the products "when they are viewed sequentially, and in different settings, rather than simultaneously." Louis Vuitton Malletier v. Burlington Coat Factory, 426 F.3d 532, 539 (2d Cir. 2005). Prominently displayed source-identification, such as a label, can negate any confusion as to the source of marks that might otherwise be similar. Nabisco, Inc. v. Warner-Lambert Co., 220 F.3d 43, 46-47 (2d Cir. 2000) (collecting cases).

There are several similarities between the Perfects and the Sanitaires: the shape and use of chrome hoods; the location of the power switch and dial; and the appearance of and text on the bag. However, each clearly presents its brand name with face plates at the front of the base assembly reading, respectively, "Sanitaire" and "PERFECT" — with the words "Sanitaire" in lower case and "PERFECT" in upper-case. PX 1-4. Whether viewed side-by-side or sequentially in different conditions, this clear labeling prevents the two from being confusingly similar.

Electrolux argues that the name "Perfect" is relatively unknown — a factual contention that may be accurate but that was not the subject of any proof in the record aside from Hoare's conclusory and self-serving assertion that the Perfect is "not a known brand," Tr. 104 — and that therefore the "PERFECT" nameplate does not effectively identify the source of the accused vacuum as being other than Electrolux. Electrolux Prop. at 55. In that regard, Electrolux emphasizes that Imig uses the "Perfect" name to identify a line of private label scrubbers marketed under the Kent private label. Electrolux Prop. at 55; Tr. 309.

The argument harms Electrolux more than it helps. At most, the argument supports the inference that the name "Perfect" may foment confusion between Imig's product and a private label. And indeed, that kind of confusion seems entirely plausible under the facts of this case. In the letter undertaking his patent inquiry, I N's counsel explicitly expressed his client's intention to produce a "private label" vacuum. SF ¶ 29. Moreover, like each of the private labels that Electrolux authorizes, the Perfect models share some of the six elements of the Sanitaire trade dress, but not all of them, and not the color red. And yet Electrolux has argued that the entire point of ensuring that the private labels do not incorporate all six of the elements of its claimed Sanitaire trade dress is to avoid confusion and thereby preserve the strength of its mark. Electrolux Prop. at 10 (citing Tr. 76-77, 80). Simply stated, the design of the Perfect is more similar to that of the various private labels that Electrolux authorizes — without fear of confusion — than it is to the Sanitaire itself. As a result, this factor does not weigh in favor of Electrolux's claim that I N have infringed on the trade dress of the Sanitaires.

c. Product Proximity In The Marketplace

Evaluation of the products' proximity in the marketplace requires assessment of "whether and to what extent the two products compete with each other and the nature of the products themselves and the structure of the relevant market." Morningside Group Ltd. v. Morningside Capital Group, LLC, 182 F.3d 133, 140 (2d Cir. 1999). Although I highlighted in my earlier ruling a need for more information on this factor, Imig, Inc., 2007 WL 900310, at *10, Electrolux adduced virtually none: it elicited from Scott Genoa that Nationwide sells both the Sanitaires and the Perfects, Tr. 306, and in the course of proving its false advertising claim it introduced exhibits in which I N made references to "its nearest competitors" that appear to have been intended to refer to the Sanitaires. PX 27. Aside from that, Electrolux offers only the argument, which is not evidence, that the Perfects are a "knock off" of the Sanitaires and therefore in competition. Electrolux Prop. at 56. I assume of course that two models of a commercial vacuum naturally compete in the same product market; but beyond that the record discloses virtually nothing about the geographic boundaries of the relevant market, the number and relative strength of other competitors, and other potentially relevant characteristics. Because the record reflects so little information about the nature of the competition between the relevant products, I view this factor as one that weighs in favor of Electrolux, but only slightly.

d. "Bridging The Gap"

The fourth Polaroid factor assesses the "likelihood that the plaintiff will bridge the gap between the products" by "enter[ing] a market related to that in which the defendant sells its product[.]" Natural Organics, Inc., 426 F.3d at 578 (quoting Playtex Prods., Inc., 390 F.3d at 162). Because, as noted above, all of the parties here compete in the same product market for commercial vacuums, this fourth factor is inapposite to this case.

e. Actual Confusion

It is of course relevant to consider evidence, even if only anecdotal, of actual consumer confusion, or the lack thereof, between the two products at issue. See Nora Beverages, Inc., 269 F.3d at 124. Despite the fact that I previously identified this factor as one I would find highly probative under the circumstances of this case, Imig, Inc., 2007 WL 900310, at *11, neither party submitted any evidence on this point. As even Electrolux concedes, this factor weighs in favor of I N. Electrolux Prop. at 56. Electrolux is of course correct to point out that evidence of actual confusion is not required. But in light of the nature of the similarities and differences between the models at issue — and also in light of the fact that Electrolux had the benefit of testimony from a witness (Hoare) who had extensive experience as a distributor and could provide specific anecdotal evidence on other matters — I find it significant that Electrolux failed to adduce a single example of a purchaser being confused by the design of the Perfect models.

f. Good Faith

Consistent with the overall purpose of trade dress protection, analysis under this factor centers on whether the defendant adopted the challenged trade dress deliberately to exploit the plaintiff's goodwill and reputation. Nora Beverages, Inc., 269 F.3d at 124-25 (quoting Lang v. Ret. Living Pub. Co., Inc., 949 F.2d 576, 583 (2d Cir. 1991)). Intentional imitation does not necessarily indicate bad faith, except to the extent that it involves a "confusingly similar" trade dress or other behavior suggesting purposeful deception. Id.

It is undisputed that Imig relied on the Sanitaires when developing the Perfects and intentionally copied certain elements. SF ¶¶ 25, 29. The parties further agree that Imig intentionally incorporated several features of the Sanitaires into its product by directing the Chinese manufacturer to refer to the latter product's specifications when designing the Perfects. SF ¶ 26. Such copying does not, in itself, demonstrate bad faith.

Both of the Genoa brothers denied an intention to make a complete copy of the Sanitaires. Tr. 162, 335. Moreover, both Genoas testified that their aim was to improve on the Sanitaires by keeping some elements while changing others, and thereby marketing a different, competitive product. Tr. 254-55, 373, 375. If the credibility of such testimony could itself constitute affirmative evidence on which Electrolux could rely to establish bad faith, this factor would present a closer question. As a general matter I found it difficult to accept as fully credible the testimony of either of the Genoas.¹⁵ Moreover, their conduct — and in particular that of Scott Genoa, in creating advertisements for the Perfects that included false comparisons to the Sanitaires — bespeaks an intention to compete unfairly against Electrolux, in part by deceiving consumers, at least as to the relative merits of the two products.

15 I note that Mark Genoa also testified that when he first contacted a man named Park to explore the possibility of developing his own commercial vacuum, he turned down Park's proposal to build a machine that was "exactly like the Sanitaire and extremely high in price." Tr. 142-43. Even if I were inclined to accept Mark Genoa's testimony at face value — which, on the whole, I am not, having observed it carefully — his description of the interaction with Mr. Park is equally amenable to a finding of good faith (he rejected the design because it copied the Sanitaire too closely) or bad (he rejected the design because it was not a cheap enough knock-off).

But to the extent that I discount the Genoas' credibility, it merely detracts from evidence that might constitute a defense; such impeachment does nothing to establish an affirmative case. Moreover, while purposeful deception can constitute evidence of bad faith, see Nora Beverages, Inc., 269 F.3d at 124-25, the nature of the deceptive conduct at issue here is incompatible with the proposition that the Perfects are likely to be confused with the Sanitaires. That is, Electrolux has demonstrated that I N engaged in false advertising by making baseless claims about the superiority of the Perfects to the Sanitaires. Such deception, while amenable to sanction on other legal grounds, not only fails to establish a likelihood of confusion for purposes of a trade dress claim, it affirmatively serves to undermine such a claim. Inherent in the deception that the Genoas sought to practice was an assumption that consumers would differentiate between the Perfects and the Sanitaires — not that I N could profit by confusing consumers into paying for Perfects while thinking they were obtaining Sanitaires. Thus, while I N may have competed in less than good faith, they did not do so in a manner likely to produce confusion in the market as to the provenance of their product. This factor accordingly — if admittedly somewhat perversely — weighs against Electrolux.

g. Quality

When an infringer markets a product of inferior quality that is likely to be confused with the plaintiff's product, it creates a risk that consumers who purchase the infringing product will mistakenly, and unfairly, blame the plaintiff for its flaws, thereby harming the plaintiff's reputation in the market. Morningside Group Ltd., 182 F.3d at 142. Here again, the record contains little evidence, and what it does contain does not advance Electrolux's cause. The evidence establishes only that the initial batch of Perfects suffered from quality control problems leading to a 40 percent return rate. SF ¶¶ 30-31; Tr. 285-86. As I explained in resolving the summary judgment motions, Imig, Inc., 2007 WL 900310, at *12, such evidence demonstrates that the Perfects were flawed but does nothing to show that any consumers were even remotely confused about the source of those defective vacuums. As a result, this factor does not weigh in Electrolux's favor.

h. Consumer Sophistication

Except in circumstances of very pronounced similarity between the trade dress features of competing products, the greater the sophistication of a product's consumers the less likely it is that those consumers will be confused about the product's source. Morningside Group Ltd., 182 F.3d at 142-43. In resolving the motions for summary judgment, the record then before me did not suffice to reach any conclusion about the level of consumer sophistication in the market for commercial vacuums, and I declined to speculate on the matter. Imig, Inc., 2007 WL 900310, at *12. To a large extent, that remains the case: Electrolux has offered neither evidence nor argument that would directly support a finding that this factor weighs in its favor.

There is one respect in which the evidence that Electrolux offered actually undermines its position on this factor. In addressing the significance of its private label models — which, as noted, include various combinations of some but not all of the components of the Sanitaire design — Electrolux attempted to walk a very fine line. On one hand, it emphasized that while it does permit the private label vacuums to incorporate elements of the Sanitaire design, it allowed them to do so only under distinct brand names and without authorizing any product that entirely replicated the Sanitaire trade dress. On the other hand, Hoare's testimony suggested that Electrolux tacitly fostered an understanding in the marketplace that it was in fact manufacturing the vacuums sold under those private labels. By doing so he appeared to contend that the Perfects' visual similarity to the trade dress of the private label models engendered a risk of confusion that would harm Electrolux. Specifically, Hoare noted the similarities between the private labels and the Sanitaires and then agreed with Electrolux's counsel that "customers associate the private label vacuums . . . with Sanitaires[.]" Tr. 82. Asked to provide an explanation for that association, Hoare appeared to assume that the buyers of commercial vacuums are sophisticated consumers who can carefully discern a vacuum's true source:

Well, there are several reasons. Even though we have differences there are still a lot of similarities. The internal components, some of these internal components, actually have our part number stamped on them. We actually have logos stamped on some of the parts so they would recognize it as ours.

Tr. 82.

Electrolux cannot have it both ways. If its customers are sophisticated enough to recognize private label models as being made by Electrolux by observing the Electrolux logo on internal components, they are sophisticated enough to understand that the absence of such logos on the Perfect internal components — combined with a nameplate that does not appear on the private label models that do have Electrolux components — means that the Perfects are not made by Electrolux. If such customers are not so sophisticated as to associate the private labels with Electrolux manufacture, then the similarity between the Perfects and the private label models does not raise the specter of confusion between the Perfect and the Sanitaire.

3. Conclusion

Electrolux bears the burden of demonstrating by a preponderance of the evidence that the trade dress for which it seeks protection is distinctive — meaning that it is non-functional and has acquired secondary meaning — and that there is a likelihood of confusion between the Sanitaires and the Perfects. For the reasons explained above, I conclude it did not establish either prong of the distinctiveness element, and that all of the factors that weigh in the balance in assessing the likelihood of confusion either undermine its claim or support it only slightly. I therefore conclude that Electrolux has failed to establish a claim of trade dress infringement against I N.

B. False Advertising

In ruling on Electrolux's earlier motion for summary judgment, I found I N liable for a false advertising violation of the Lanham Act and reserved judgment on the issue of appropriate relief. See Imig, Inc., 2007 WL 900310, at *13-15. Now that the bench trial is concluded, Electrolux seeks an injunction as well as monetary damages based on its own claimed loss of sales as well as damages reflecting I N's profits. I discuss each request for relief in turn.

1. Injunctive Relief

Electrolux seeks to enjoin I N from repeating in any advertising the claims I previously to be false. See Imig, Inc., 2007 WL 900310, at *15. It also asks me to enjoin I N from making any other claims about the performance of its vacuums "unless those claims are truthful and supported by industry recognized testing." Electrolux Prop. at 63. As a general matter, an injunction is a form of relief that is available for a violation of Section 43(a) the Lanham Act. See 15 U.S.C. § 1125(c)(5). In such circumstances, a court has the "power to grant injunctions, according to the principles of equity and upon such terms as the court may deem reasonable." 15 U.S.C. § 1116.

To demonstrate that it is entitled to such relief, Electrolux must demonstrate that I N "(1) ha[ve] made material misrepresentations about the . . . characteristics [of the relevant products]; (2) ha[ve] used false or misleading representations `in commerce;' (3) ha[ve] made the representations in the context of commercial advertising . . .; and (4) ha[ve] made [Electrolux] believe that [it] is likely to be damaged by the misrepresentations." Mobius Management Systems, Inc. v. Fourth Dimension Software, Inc., 880 F. Supp. 1005, 1019 (S.D.N.Y. 1994) (internal citations omitted). If the misrepresentation is shown to be literally false, the court may disregard the impact of the advertisement on consumers. Id. at 1022 (quoting Coca-Cola Co. v. Tropicana Products, Inc., 690 F.2d 312, 317 (2d Cir. 1982)). Further, if the advertisement specifically compared the plaintiff's and defendant's products in a misleading manner, "irreparable harm warranting injunctive relief will be presumed." Id. (internal citations omitted). Typically, in false advertising claims, the plaintiff must show "the likelihood that purchasers of the product may be misled in the future by the false advertising." Burndy Corp. v. Teledyne Industries, Inc., 748 F.2d 767, 772 (2d Cir. 1984) (emphasis added). However, when the court finds the defendant to have acted in bad faith, such willful infringement casts doubt on the defendant's future behavior, warranting injunctive relief.

I have previously determined that I N made material misrepresentations about the performance of the Perfects and they used those statements in commercial advertising that were in commerce. Imig, Inc., 2007 WL 900310, at *15. Electrolux has therefore met its burden with respect to the first three elements listed above, and the fact that the claims were literally false relieves Electrolux of any need to establish the fourth. Id. With respect to the risk of future harm, I note that in 2005, I N ceased using the advertising I have determined to be false. I N Prop. at 8. That fact alone does not obviate the need for injunctive relief, however, in light of the apparent willfulness with which I N acted in publishing its false advertisements. At any rate, I N have no objection to an order that permanently enjoins the specific advertising claims I have previously determined to be false. See Imig, Inc., 2007 WL 900310, at *15. As to the request that I enjoin the defendants from making "any other advertising claims . . . unless those claims are truthful," Electrolux Prop. at 63, such prior restraint on advertising seems neither necessary nor equitable; it also seems likely to produce needless litigation about the vetting process such a regime would require. I decline to grant such open-ended injunctive relief.

2. Monetary Relief

In addition to injunctive relief, the Lanham Act authorizes Electrolux to seek various forms of monetary relief, including "(1) defendant's profits, (2) any damages sustained by the plaintiff, and (3) the costs of the action." 15 U.S.C. § 1117(a). I examine each of the first two below, and return to the issue of costs — which arises with respect to the copyright claim as well — in a separate section at the end of this discussion.

a. Electrolux's Damages

Electrolux has failed to provide any reliable measure of its actual damages. Indeed, in describing the state of the record in this regard, Electrolux says only the following: "Electrolux failed to meet its budget projections. In 2005, Electrolux's sales of the [Sanitaires] were off by $1 million. In 2006, Electrolux's sales were off by $600,000. Electrolux, however, hit its budget projections for its other products during this time." Electrolux Prop. at 39 (citing Tr. 88). Even taken at face value, such argument fails to make a sufficient causal connection between I N's misconduct and the cited harm. But the assertions cannot be taken at face value: they rest on nothing more than conclusory allegations made by Hoare without any attempt to show that Electrolux's budget projections were reasonable, without any data supporting the assertion of sales being "off," and without any exploration of other factors that might have caused the loss of sales. Such projections, without any supporting evidence to provide context or reliable calculation methods, are plainly insufficient. See, e.g., Tambrands, Inc. v. Warner-Lambert Co., 673 F. Supp. 1190, 1197 (S.D.N.Y. 1987) ("Tambrands offered sparse evidence to show that its projections were reliable, were based upon nonerroneous assumptions, were calculated using a methodology that is appropriate, or were based on anything more than pure speculation.") (citing Foxtrap, Inc. v. Foxtrap, Inc., 671 F.2d 636, 642 (D.C. Cir. 1982)). I therefore decline to award any damages on the basis of Electrolux's purportedly lost sales.

b. I N's Profits

As an alternative to actual damages, Electrolux can seek an award for I N's false advertising violation of the Lanham Act based on I N's profits. A court granting such relief retains significant discretion to modify it as fairness may require. 15 U.S.C. § 1117(a) ("If the court shall find that the amount of the recovery based on profits is either inadequate or excessive the court may in its discretion enter judgment for such sum as the court shall find to be just, according to the circumstances of the case.").

In considering such relief, I reject I N's argument it should be responsible only for those damages caused by its Lanham Act violation. I N Prop. at 23. That argument is inconsistent with the statute's purpose. See Riggs Investment Management Corp. v. Columbia Partners, LLC, 996 F. Supp. 1250, 1271 (D.C. Cir. 1997) ("The Lanham Act does not prohibit false advertising `until it confuses the public.' Rather it prohibits false advertising."). Moreover, while a plaintiff may normally not recover a defendant's profits related to a non-willful Lanham Act violation, see LifeServices, Inc. v. Natural Organics, Inc., 2007 WL 4437168 (S.D.N.Y. Dec. 17, 2007), a court may award a plaintiff its adversary's profits, without any showing of actual damages, where the evidence shows that the defendant acted in bad faith. Riggs Investment Management Corp., 996 F. Supp. at 1271 (quoting Foxtrap Inc. v. Foxtrap, Inc., 671 F.2d 636, 641-42 (D.C. Cir. 1982); see also Malletier v. Dooney Bourke, Inc., 500 F. Supp. 2d 276, 279 (S.D.N.Y. 2007) ("in order to justify an award of profits, a plaintiff must establish that the defendant engaged in willful deception") (quoting George Basch Co. v. Blue Coral, Inc., 968 F.2d 1532, 1540 (2d Cir. 1992)).

I N contend that they did not act in bad faith. I N Prop. at 21. Specifically, they assert that the Genoas, who were responsible for the advertisements, did not know that the information they conveyed was false, and in particular they point to Scott Genoa's testimony about having viewed performance tests in China in 2004 that convinced him of the advertisements' factual claims. Id. at 8-10. I disagree. I found Scott Genoa's testimony on this subject to lack credibility; but even taken at face value, it is clear that he cherry-picked the available evidence in a patently unfair way to make claims about motor life that were demonstrably stronger than even the evidence on which he purported to rely. Similarly, I find Mark Genoa's testimony about the ETL testing of the Perfects' motor strength to be incredible, and therefore find that the corresponding claims in the advertisements were knowingly false. I therefore find that the defendants — both of which are controlled by the Genoas — acted willfully in making the false claims that appeared in their advertisements.

I therefore can and do grant Electrolux's relief for an award of I N's profits. While there is a risk that such an award over-compensates Electrolux's actual harm, the nature of I N's violation renders it difficult to measure that injury more precisely. While I do not assume that every dollar of I N's profit was attributable to the false advertising, it is fair to assume that I N engaged in such conduct precisely for the purpose of realizing such profits. It is, therefore, the most reasonable measure of damages available under the circumstances. The parties have stipulated the amounts of I N's profits and costs, and I predicate my calculation of the relevant amount of profits on that stipulation and the reports on which it is based. See DE 89; PX 57, 64, 65, 70. Based on that information, as set forth above in the findings of fact, I award to Electrolux the amount of $272,153, corresponding to the amount of I N's profits for the period from April 29, 2005 through August 31, 2005.

C. Copyright Infringement

I have previously found I N liable for violation of the Copyright Act of 1974 based on its dissemination of an owner's guide, in conjunction with the sales of the first 425 Perfect vacuums, that replicated Electrolux's corresponding guide for the Sanitaires. See Imig, Inc., 2007 WL 900310, at *16-17. I reserved a decision as to damages on that claim pending the completion of the bench trial.

Electrolux seeks an award based on I N's profits from the sale of the first 400 Perfects. Electrolux Prop. at 64-65. Following a determination of liability for copyright infringement under the Copyright Act of 1974, the copyright owner may recover either actual damages and profits or statutory damages. 17 U.S.C. § 504(a). Statutory damages, which permit an increase up to $150,000 following a finding of willful infringement, may only be awarded at the election of the copyright owner, prior to the rendering of final judgment. Id. § 504(c)(1)-(2). Since Electrolux made no such election, it is entitled to actual damages and profits, including "actual damages suffered by [it] as a result of the infringement, and any profits of the infringer that are attributable to the infringement and are not taken into account in computing the actual damages." Id. § 504(b). In determining the amount of such damages, the plaintiff must "present proof only of the infringer's gross revenue" which the infringer may offset with proof of its "deductible expenses and elements of profit attributable to factors other than the copyrighted work." Id.

As explained above in the statement of facts, the parties have agreed on facts from which I calculate that Imig's profit on the first 400 Perfects that it sold (the sales for which Electrolux seeks profits on the basis of the copyright violation) was $11,856.00. There is of course a sense in which that calculation should end the matter: Imig has established expenses that reduce the gross sales revenue to that amount and nothing lower. But such methodology plainly overcompensates Electrolux. Not only has the latter shown no actual damages, but the profits it seeks to recover are not those associated with the sale of its copyrighted Owner's Guide as such, but rather the profits Imig realized by selling vacuums — and by including the Owner's Guides with those vacuums upon completing the sale. Simply put, no marketer sells — and no typical consumer seeks to buy — a user's manual with a vacuum thrown in; the subject of the transaction is the vacuum, and the inclusion of a manual is merely one of several incidents to such a transaction that market participants have come to expect.

If there were no way of accounting for that reality, I would not hesitate to make Imig bear the burden of overcompensating Electrolux by awarding the profits calculated above. But the record does include evidence that suffices to achieve a more nuanced assessment of the profits associated specifically with the copyright infringement: namely, the cost to Electrolux of producing the Owner's Guide. As Hoare explicitly testified on both direct and cross-examination (citing, on cross-examination, invoices that neither side offered in evidence), Electrolux spent approximately $9,000 to develop its Owner's Guide. Tr. 89-91. That amount therefore provides a comparatively reliable measure of what Imig would have had to spend to develop it's own manual for the first 400 vacuums it sold if it had not misappropriated Electrolux's copyright-protected work.¹⁶ In other words, if instead of spending nothing to produce an Owner's Guide and improperly copying Electrolux's, Imig had taken the same actions as previously taken by Electrolux and spent the same $9,000 to produce its own manual, the result would have been a profit on the first 400 vacuums of $2,856: $11,856 in overall profits minus $9,000 in increased costs attributable to paying to produce the owner's guide. I therefore conclude that the amount of Imig's profit attributable to the copyrighted work — as opposed to the profit attributable to non-infringing conduct — is $9,000, and I therefore award that amount to Electrolux on its copyright claim.¹⁷

16 I N assert that after Imig stopped copying the Electrolux Owner's Guide, Mark Genoa created a new manual at no cost — although he concedes that in doing so he used other manufacturers' manuals as a guide and rephrased them in his own words, and that his Chinese manufacturer then printed them at no charge to I N. Tr. 273-74. Even taking this testimony at face value — and ignoring the implicit suggestion that taking advantage of other manufacturers' (presumably copyrighted) manuals does not reflect a development cost to someone, if not I N — it establishes only that the cost of developing the replacement manual cannot easily be measured in dollars. It does not establish that it costs nothing (or should have cost nothing) to produce the copyrighted work that Imig misappropriated. The testimony in this regard therefore does not support a different remedy for the copyright violation.

17 To be sure, the $9,000 figure was an estimate. Had Imig subpoenaed and introduced the actual invoices — which Hoare estimated might actually reflect an amount slightly less than the round number he cited — it might have succeeded in proving that the amount of profits attributable to the copyrighted work was somewhat lower. But because it is Imig's burden to prove "elements of profit attributable to factors other than the copyrighted work," 17 U.S.C. § 504(b), it must bear the risk of failing to do so with greater precision.

D. Other Relief

1. Attorneys' Fees

In addition to its requests for injunctive relief and monetary damages, Electrolux also seeks reimbursement of its attorneys' fees. The Lanham Act does authorize an award of "reasonable attorney fees to the prevailing party," but only in "exceptional cases." 15 U.S.C. § 1117(a). "A finding of deliberate, wilful, or bad faith violations of the Lanham Act is sufficient to warrant the finding of an exceptional case." Mobius Management Systems, Inc., 880 F. Supp. at 1026 (internal citations omitted). Indeed, having found that I N's false advertising was willful, I would abuse my discretion by failing at least to consider an award of reasonable attorneys' fees. See id. (citing Goodheart Clothing v. Laura Goodman Enterprises, 962 F.2d 268, 272 (2d Cir. 1992)). I have therefore considered the application.

In light of all the circumstances, I conclude that no award of fees is warranted. First, although Electrolux has succeeded in establishing that I N committed a false advertising violation of the Lanham Act and did so willfully, that claim was by no means the primary issue in this litigation. The bulk of the trial, and of the attorneys' work leading up to it, focused on the trade dress claim — a claim that Electrolux thoroughly failed to establish. Accordingly, if any award of fees was warranted, it would have to be discounted significantly to reflect the extent to which Electrolux did not prevail on its primary claim. Second, although the award of monetary damages on the false advertising claim is appropriate for the reasons set forth above, it almost certainly overcompensates Electrolux for any harm it actually suffered as a result of the false statements that I N used to tout the Perfects. I N must bear the burden of such potential overcompensation under the circumstances, but to the extent I have the discretion to avoid compounding the windfall to Electrolux by denying its request for attorneys' fees, I believe it is appropriate to do so. See S.C. Johnson Son, Inc. v. Carter-Wallace, Inc., 781 F.2d 198, 201 (Fed. Cir. 1986) ("Even an exceptional case does not require in all circumstances the award of attorney fees. Many factors could affect this result. The trial judge is in the best position to weigh considerations such as the closeness of the case, the tactics of counsel, the conduct of the parties, and any other factors that may contribute to a fair allocation of the burdens of litigation as between winner and loser."). I therefore deny the request for an award of attorneys' fees.

2. Costs

Finally, Electrolux also seeks the taxation of its litigation costs. That request is premature at this stage: the applicable rule provides that "[w]ithin thirty (30) days after the entry of final judgment . . . any party seeking to recover costs shall file with the clerk a request to tax costs." Loc. Civ. R. 54.1 (emphasis added). I therefore deny the request for costs, without prejudice to renewal within 30 days after the entry of judgment.

III. Conclusion

For the reasons set forth above, as well as those set forth in my earlier decision on the cross-motions for summary judgment, I respectfully direct the Clerk to enter judgment as follows:

A. In favor of defendant Electrolux Home Care Products, Ltd. on all of the claims in the complaint of plaintiff Imig, Inc.;

B. In favor of counter-defendants Imig, Inc. and Nationwide Sales Services, Inc. on Count I of the counterclaim asserted by Electrolux Home Care Products, Ltd. (Federal Unfair Competition);

C. Jointly and severally against the counter-defendants in the amount of $272,153.00 on the claim for false advertising pleaded as a basis for liability under Count III of the counterclaim;

D. Jointly and severally against the counter-defendants in the amount of $9,000.00 on Count I of the counterclaim (Copyright Infringement); and

E. Permanently enjoining both counter-defendants from making any factual claim that I have previously determined to be literally false.

SO ORDERED.