Before HIGGINBOTHAM, SMITH, and DENNIS, Circuit Judges. JERRY E. SMITH, Circuit Judge: Future Proof Brands, L.L.C. (“Future Proof”) and Molson Coors (“Coors”) sell competing hard seltzer beverages. Future Proof named its seltzer “Brizzy.” Coors chose “Vizzy.” Future Proof sued Coors for trademark infringement, claiming that consumers would confuse Vizzy and Brizzy. The district court denied Future Proof’s motion for a preliminary injunction. Finding no abuse of discretion, we affirm. I. Hard seltzers are alcoholic beverages that contain carbonated water, alcohol, and—in most cases—fruit flavors, and that have enjoyed skyrocketing popularity in the United States.[1] Many alcoholic-beverage makers have rushed to capitalize on that trend, releasing their own lines of hard seltzers. Future Proof and Coors are two such competitors. Both named their hard seltzers with a variation of the word “fizzy.” Future Proof styles Brizzy as a “seltzer cocktail” and sells 12-packs at retail for $14.99 in four states. Future Proof registered its BRIZZY mark with the United States Patent and Trademark Office (“USPTO”) in 2019. Coors branded Vizzy by amalgamating its two “most prominent attributes: Vitamin C and fizzy.” Future Proof contends that Coors knew about Brizzy by the time it began marketing Vizzy and that a wholesaler once asked a Future Proof employee about Vizzy instead of Brizzy. Neither company was the first to have the inspiration to brand a carbonated product with a variant of “fizzy.” Other products include Malibu’s “FIZZY PINK LEMONADE,” Malibu’s “FIZZY MANGO” drink, “IZZE” sparkling juice, “FIZZY FOX” sparkling shrub, IGA’s “FIZZY” sparkling water, and Hubble’s “FIZZY JUICE” sparkling juice drink. The parties’ respective products look like this: The cans are dissimilar in a variety of ways. First, they are shaped differently. Second, Brizzy cans have “solid and dark backgrounds” along with a “‘digital’ appearance with bubbles and the vague shapes of drinking glasses.” In contrast, Vizzy cans have “white backgrounds” and “pictures of fruit.” Third, the text on each is different. Brizzy cans include the flavor and the words “SELTZER COCKTAIL” in small text. Vizzy cans feature the words “HARD SELTZER,” the flavor, and the bolded phrase “With Antioxidant Vitamin C.” Future Proof sued Coors for trademark infringement, seeking to preliminarily enjoin Coors from “selling and marketing products confusingly similar to” its BRIZZY mark. The district court declined to issue the injunction. Future Proof timely appeals. II. We review the denial of a preliminary injunction for abuse of discretion. MWK Recruiting Inc. v. Jowers, No. 19-51064, 2020 U.S. App. LEXIS 35286, at *3 (5th Cir. Nov. 6, 2020) (per curiam) (unpublished). “As to each element of the district court’s preliminary-injunction analysis, however, the district court’s findings of fact are subject to a clearly-erroneous standard of review, while conclusions of law are subject to broad review and will be reversed if incorrect.” Dennis Melancon, Inc. v. City of New Orleans, 703 F.3d 262, 267 (5th Cir. 2012) (cleaned up). For instance, the preliminary-injunction factor at issue—likelihood of confusion—”is a question of fact reviewed for clear error.”[2] Under the clearly-erroneous standard, we uphold factual findings that “are plausible in light of the record as a whole.” Moore v. Brown, 868 F.3d 398, 403 (5th Cir. 2017) (per curiam). A preliminary injunction is “an extraordinary remedy which should not be granted unless the party seeking it has clearly carried [its] burden of persuasion . . . .” PCI Transp., Inc. v. Fort Worth & W. R.R. Co., 418 F.3d 535, 545 (5th Cir. 2005) (cleaned up). “Only under extraordinary circumstances will we reverse the denial of a preliminary injunction.” Anderson v. Jackson, 556 F.3d 351, 355–56 (5th Cir. 2009) (cleaned up). Even if we disagree with the district court’s analysis in some places, “we may not simply . . . substitute our judgment for the trial court’s, else that court’s announced discretion would be meaningless.” White v. Carlucci, 862 F.2d 1209, 1211 (5th Cir. 1989) (cleaned up). III. A plaintiff must prove four factors to obtain a preliminary injunction.[3] Because Future Proof fails on the first factor—”a substantial likelihood of success on the merits”—we do not address the other three. Planned Parenthood, 692 F.3d at 348. To determine likelihood of success on a trademark-infringement claim, courts consider (1) the type of mark infringed, (2) the similarity between the marks, (3) the similarity of the products, (4) the identity of the retail outlets and purchasers, (5) the identity of the advertising media used, (6) the defendant’s intent, (7) evidence of actual confusion, and (8) the degree of care exercised by potential purchasers. Bd. of Supervisors for La. State Univ. Agric. & Mech. Coll. v. Smack Apparel Co., 550 F.3d 465, 478 (5th Cir. 2008). We label those “digits of confusion.” Viacom Int’l v. IJR Cap. Invests., L.L.C., 891 F.3d 178, 192 (5th Cir. 2018). Two of those digits possess particular prominence: The sixth—bad intent—is “not necessary” but “may alone be sufficient to justify an inference that there is a likelihood of confusion.” Streamline Prod. Sys., Inc. v. Streamline Mfg., Inc., 851 F.3d 440, 455 (5th Cir. 2017) (cleaned up). Likewise, the seventh—actual confusion—constitutes the “best evidence of a likelihood of confusion.” Viacom, 891 F.3d at 197 (cleaned up). “[A] finding of a likelihood of confusion need not be supported by a majority of the [digits].” Streamline Prod., 851 F.3d at 453 (cleaned up). And district courts may weigh those digits “differently from case to case, depending on the particular facts and circumstances involved.” Id. (cleaned up). The district court concluded that the third, fourth, and fifth digits favored granting the injunction. Neither party contests that decision, so we don’t consider those digits. Future Proof contends that the court erred in evaluating the first, second, sixth, seventh, and eighth digits. We examine each. Although the court made some errors, it correctly concluded that Future Proof failed to show a substantial likelihood of success on its trademark-infringement claim. A. Future Proof contends that the district court erred in determining that the first digit—”the type of mark allegedly infringed”—weighed against the injunction. Smack Apparel, 550 F.3d at 478 (cleaned up). We disagree. The type-of-mark digit refers to the strength of a mark: Strong marks receive “the widest ambit of protection,” and weak marks do not. Sun Banks of Fla., Inc. v. Sun Fed. Sav. & Loan Ass’n, 651 F.2d 311, 315 (5th Cir. July 1981). To determine the strength of a mark, we examine (1) “where the mark falls on a spectrum ” of categories and (2) “ the standing of the mark in the marketplace.” Am. Rice, 518 F.3d at 330. 1. The spectrum employs five categories: “(1) generic, (2) descriptive, (3) suggestive, (4) arbitrary, and (5) fanciful.” Streamline Prod., 851 F.3d at 451 (cleaned up). “[T]he strength of a mark, and of its protection, increases as one moves away from generic and descriptive marks toward arbitrary marks.” Am. Rice, 518 F.3d at 330 (cleaned up). We must determine (1) which categories on the spectrum are strong and (2) where BRIZZY falls on the spectrum. a. The relevant spectrum measures two separate aspects of trademarks—distinctiveness and strength. Future Proof confuses the two. First, distinctiveness is a “condition[] for registration” of a mark on the principal register: “The more distinctive the mark, the more readily it qualifies for the principal register.” U.S. Pat. & Trademark Off. v. Booking.com B. V., 140 S. Ct. 2298, 2302 (2020). Generic marks are never distinctive. Id. at 2303. Descriptive marks can become distinctive, but only by acquiring “secondary meaning.” Id. (cleaned up). Suggestive, arbitrary, and fanciful marks are “inherently distinctive . . . .” Id. at 2302 (cleaned up). Second, that spectrum helps us determine a mark’s strength. See Sun Banks, 651 F.2d at 315. But the strength inquiry is different from the distinctiveness inquiry in that suggestive marks don’t always make the cut. “A strong mark is usually fictitious, arbitrary or fanciful . . . .” Id. Suggestive marks, on the other hand, are “comparatively weak . . . .”[4] Moreover, classification of the mark on the spectrum is “not conclusive of ‘strength,’ . . . [because] a descriptive mark through vigorous promotion can become a strong mark, and an arbitrary mark that is not well known in the market can be a weak mark.”[5] In a word, suggestive marks are not necessarily strong marks that favor granting an injunction. For instance, we have found that suggestive marks support granting an injunction,[6] support denying an injunction,[7] or are indeterminate,[8] depending on other factors that help establish strength. At issue is whether BRIZZY is strong—not whether it’s distinctive.[9] Thus, Future Proof cannot prevail on that digit solely by showing that BRIZZY is suggestive. b. The district court concluded that BRIZZY is descriptive. Coors agrees. Future Proof, on the other hand, contends that BRIZZY is arbitrary or suggestive. Neither party suggests that BRIZZY is generic or fanciful, so we don’t analyze those categories. We conclude that BRIZZY is suggestive. Courts define descriptive, suggestive, and arbitrary marks by their relationship to the products they represent. Descriptive marks “convey[] an immediate idea of the qualities, characteristics, effect, purpose, or ingredients of a product or service.”[10] Suggestive marks “require[] the consumer to exercise the imagination in order to draw a conclusion as to the nature of the goods . . . .” Amazing Spaces, 608 F.3d at 241 (emphasis added) (cleaned up). Arbitrary marks “bear no relationship to the products or services to which they are applied.” Id. (emphasis added) (cleaned up). We must, therefore, determine whether BRIZZY (1) conveys an immediate idea of the qualities of seltzers, (2) requires consumers to exercise their imaginations to draw conclusions about seltzers, or (3) bears no relationship to seltzers. First, BRIZZY is not descriptive, because it does not “convey[] an immediate idea” about the characteristics of hard seltzers. Zatarains, 698 F.2d at 792. To conclude anything about hard seltzers from “brizzy,” a consumer must make an inference. For instance, a consumer might surmise that “brizzy”—because it rhymes with “fizzy”—denotes carbonation. But that consumer must first infer that “brizzy” is a play on “fizzy.” Using the same logic, a consumer might surmise that “brizzy”—because it rhymes with “dizzy”—denotes the sensation a consumer gets after a few rounds (because of a high alcohol content). But that consumer must first infer that “brizzy” is a play on “dizzy.” Either result constitutes conjecture. Given the necessary inference, BRIZZY does not “convey[] an immediate idea” about the characteristics of hard seltzers. Id. (emphasis added).[11] Second, BRIZZY is not arbitrary, because it “bear[s] [a] relationship to the products” to which it refers. Amazing Spaces, 608 F.3d at 241 (cleaned up). It rhymes with “fizzy,” which denotes the carbonated quality of hard seltzers. Moreover, arbitrary marks are usually “ordinary words which do not suggest or describe the services involved.”[12] And Future Proof concedes that “brizzy” is not an ordinary word in the English language, because Future Proof coined it.[13] BRIZZY is thus not arbitrary. Third, BRIZZY is suggestive. “Brizzy” relates to carbonation only because it rhymes with “fizzy.”[14] To make that connection—and make a conclusion about the carbonation of Brizzy—consumers must “exercise the imagination . . . .” Id. at 241 (cleaned up). A consumer must infer that “brizzy” is a play on “fizzy” and not on other words. That exercise of the imagination renders BRIZZY suggestive. Although we disagree with the district court’s classification of BRIZZY as a descriptive mark, suggestive marks—like descriptive marks— are “comparatively weak.” Sun Banks, 651 F.2d at 315. Thus, even if the district court erred in making that initial classification, its ultimate conclusion that BRIZZY is weak was not necessarily erroneous.[15] c. Future Proof contends that the district court should have applied a presumption that BRIZZY was valid. We disagree. Registration of a mark on the principal register provides “prima facie evidence of the validity of the registered mark . . . .” 15 U.S.C. § 1057(b). That provision provides a presumption that a registered mark is inherently distinctive. All. for Good Gov’t v. Coal. for Better Gov’t, 901 F.3d 498, 507, 508 (5th Cir. 2018). The rationale is that, if the USPTO didn’t require evidence of secondary meaning—which it would have needed if the mark were merely descriptive—the USPTO must have registered the mark based on its conclusion that the mark is inherently distinctive. Nola Spice Designs, L.L.C. v. Haydel Enters., Inc., 783 F.3d 527, 537 n.1 (5th Cir. 2015). But “the presumption of validity that attaches to a service mark is not relevant to the issue of infringement.” Sun Banks, 651 F.2d at 315 (cleaned up). And that makes sense. For an infringement claim, we examine the strength of a mark, not its distinctiveness or validity.[16] Thus, Future Proof’s claim that the district court should have given “weight to the presumption of distinctiveness” is inconsistent with our precedent.[17] 2. Although Future Proof largely closes its argument by contending that BRIZZY is distinctive, that is not the end of our analysis. Besides the spectrum, to determine strength, “the key inquiry before us, is determined by . . . standing in the marketplace.” Sun Banks, 651 F.2d at 315. Specifically, evidence of “third-party single and multi-word uses” of a mark tends to show weakness. Id. at 316. We do not require third-party usage involving the plaintiff’s entire mark, but instead only the portion of the plaintiff’s mark that the defendant also uses.[18] Brizzy and Vizzy share the common “IZZY” root. Several third parties also brand their products with the “IZZY” root. Those include Malibu’s “FIZZY PINK LEMONADE,” Malibu’s “FIZZY MANGO” drink, “IZZE” sparkling juice, “BIZZY” coffee, “FIZZY FOX” sparkling shrub, IGA’s “FIZZY” sparkling water, and Hubble’s “FIZZY JUICE” sparkling drink. That third-party usage of the “IZZY” root tends to show that BRIZZY is weak. In fact, we recently concluded that a suggestive mark did not “enjoy strong standing in the market,” where the record showed six examples of third-party usage, which constituted “widespread third-party use.” Springboards to Educ., 912 F.3d at 815. Future Proof raises two objections to that analysis. First, it objects that “not one beverage product identified by Coors is a hard seltzer.” But we do not confine our analysis of third-party usage to products of the exact type that the plaintiff sells.[19] Second, Future Proof implies that we may not parse letters of a mark for the purposes of identifying third-party usage, contending that Coors “failed to identify a single competitor using the term ‘BRIZZY’ to market any product.” But we don’t require litigants to show third-party usage of the plaintiff’s entire mark. Instead, third-party usage involves the portion of the plaintiff’s mark that the defendant also uses. See Sun Banks, 651 F.2d at 314, 316; Holiday Inns, 481 F.2d at 448. At bottom, it’s Future Proof’s contention about how the two marks might confuse consumers that warrants our parsing of BRIZZY.[20] Future Proof contends that consumers will confuse Brizzy and Vizzy not because the “Br” and “V” letters are similar, but because the marks share the common “IZZY” root.[21] Because Future Proof has made the shared “IZZY” root the basis of its claim, it is likewise appropriate to analyze third-party usage of the “IZZY” root. Because BRIZZY is suggestive and there is significant evidence of third-party usage, the district court’s conclusion that BRIZZY is weak is not “clear error” as Future Proof suggests. B. Future Proof contends that the district court erred in determining that the second digit—”the similarity between the two marks,” Smack Apparel, 550 F.3d at 478 (cleaned up)—weighed “only marginally in favor of granting the injunction . . . .” That digit, the argument goes, weighs “heavily in favor of injunctive relief” for two reasons: the court erroneously (1) “focused on certain visual differences in product packaging” and (2) failed to consider the aural similarities of “ brizzy” and “vizzy.” We disagree on both points. First, Future Proof claims that the court shouldn’t have considered differences in product packaging. But to determine the similarity of two marks, we examine “whether, under the circumstances of use, the marks are similar enough that a reasonable person could believe the two products have a common origin or association.” Xtreme Lashes, 576 F.3d at 228. And it is well established that “[c]ourts consider marks in the context that a customer perceives them in the marketplace,” which can include “labels, packages, or . . . advertising material directed to the goods “[22] The district court correctly noted several differences in product packaging. Brizzy and Vizzy cans have different shapes. Moreover, Brizzy cans have “solid and dark backgrounds” with a “‘digital’ appearance with bubbles and the vague shapes of drinking glasses.” Vizzy cans, on the other hand, feature white backgrounds and pictures of fruit. Finally, the text surrounding each mark is different. Brizzy cans have the flavor and the words “seltzer cocktail” in small text. Conversely, Vizzy cans feature the words “hard seltzer,” the flavor, and the “bolded text ‘With Antioxidant Vitamin C.’” Second, Future Proof claims that the court failed to consider aural similarities of the “B” and “V” consonants. That claim is baseless, because the court did consider the aural similarities of “B” and “V.” It is true that the aural similarities of two marks are relevant. Capece, 141 F.3d at 201. Here, the only difference between the two marks is the “Br” and “V” consonants. And the Court of Customs and Patent Appeals did conclude that “‘VEEP’ and ‘BEEP’” are similar marks, because “[t]he consonant sounds ‘B’ and ‘V’ are likely to be misunderstood by the listener, depending as they do on such variables as the diction of the speaker and the hearing acuity of the listener.” Krim-Ko Corp. (Krim-Ko Div., Nat. Sugar Ref. Co.) v. Coca-Cola Bottling Co. of N.Y., 390 F.2d 728, 731–32 (C.C.P.A. 1968). The court didn’t provide any citation to support that claim, and the court didn’t analyze BRIZZY’s “Br” sound. Future Proof thus overemphasizes Krim-Ko‘s persuasive value, and, contrary to what Future Proof contends, Krim-Ko in no way “compels the conclusion” that BRIZZY and VIZZY are “confusingly similar . . . .” In any event, the district court concluded that the similarity of the “B” and “V” consonants “is relevant here.” Future Proof also contends that those aural similarities are “particularly important,” because consumers often purchase alcoholic drinks “by verbal request” in bars and restaurants. But Future Proof provides no evidence that its seltzers sell primarily—or at all—in bars and restaurants.[23] In fact, Future Proof tells us that consumers can find Brizzy at “retail locations” like “H-E-B” or “Circle K.” Even supposing that Brizzys pervade bars and restaurants, our precedent says only that “[s]imilarity of sound . . . may be taken into account,” not that we should place aural similarity on a pedestal. Marathon Mfg. Co. v. Enerlite Prod. Corp., 767 F.2d 214, 219 (5th Cir. 1985) (emphasis added) (cleaned up). The court did not err in concluding that that digit weighs “only marginally in favor of granting the injunction “ C. Future Proof claims that the court “did not correctly weigh” the sixth digit—”the defendant’s intent,” Smack Apparel, 550 F.3d at 478 (cleaned up)—because it found that that digit “weighs in favor of denying the injunction.” Specifically, it claims that the sixth digit is “no worse than neutral.” Even assuming that the district court erred, that would not be substantial enough to render the court’s ultimate decision about likelihood of confusion an abuse of discretion. First, Future Proof claims that Coors and its executives were “keenly aware” and had “constructive notice” of Brizzy at the time they began marketing Vizzy. “But mere awareness of the senior user’s mark does not establish . . . bad intent.” Streamline Prod., 851 F.3d at 456 (cleaned up). Instead, “[o]ur intent inquiry focuses on whether the defendant intended to derive benefits from the reputation of the plaintiff,” and we usually rely on evidence that a defendant “imitat[ed] packaging material” or “adopt[ed] similar distribution methods.” Id. at 455, 456 (cleaned up). Future Proof provides no such evidence. Given its failure to provide evidence, Future Proof’s claims about Coors’s knowledge do not establish bad intent. Second, Future Proof says that the district court erroneously concluded that “[t]his digit weighs in favor of denying the injunction” even though that digit can only be neutral. Our precedent has used inconsistent language on that issue. We usually indicate that an absence of intent renders that digit neutral.[24] Conversely, we have also found that the digit can “weigh[] against finding a likelihood of confusion.” Streamline Prod., 851 F.3d at 457. Coors is silent on that point. We do not resolve that inconsistency here. Even assuming, arguendo, that the district court got it wrong, that error would not tip the scales in favor of finding an abuse of discretion. Future Proof bears the burden of establishing a likelihood of success, and it failed to provide evidence on that digit. Thus, even if the court employed the wrong language in drawing its conclusion about that digit, it correctly concluded that the sixth digit does not support the injunction. D. Future Proof contends that the district court erred in determining that the seventh digit—”evidence of actual confusion” among consumers, Smack Apparel, 550 F.3d at 478 (cleaned up)—”weighs against granting the injunction.” The court did not err in weighing that digit. The district court considered “one instance of actual confusion,” namely that a wholesaler confused Brizzy and Vizzy. The court concluded that (1) that was not evidence of actual confusion, because it involved a wholesaler—not a consumer—and (2) regardless, the instance involved a fleeting mix-up of names, which is insufficient to establish actual confusion. Future Proof objects to that first conclusion, contending that the district court was wrong to conclude that “a wholesaler is not a consumer” for purposes of actual confusion. We agree on that point. A “ plaintiff need not . . . prove confusion in actual consumers,” where there is evidence of actual confusion “on [the] part of distributors.”[25] The district court thus incurrectly concluded that wholesalers do not count as consumers. But the district court expressly set that erroneous conclusion aside and determined that, regardless, the alleged isolated incident was not “actual confusion,” but instead a fleeting “mix-up.” And it is telling that Future Proof does not object to that second conclusion. Actual confusion must be “more than a fleeting mix-up of names . . . .” Streamline Prod., 851 F.3d at 457 (cleaned up). A plaintiff “must show that the confusion . . . swayed consumer purchases.” Id. (cleaned up). Future Proof offered evidence that a wholesaler once asked a Future Proof employee about Vizzy, instead of Brizzy.[26] But Future Proof provides no evidence that that confusion “swayed consumer purchases.” Id. (cleaned up). At best, it has shown a “fleeting mix-up of names,” and that isn’t sufficient to establish actual confusion. Id. (cleaned up). Consequently, the district court did not err in concluding that Future Proof failed to show actual confusion. E. Future Proof contends that the district court “did not correctly weigh” the eighth digit—”the degree of care exercised by potential purchasers,” Smack Apparel, 550 F.3d at 478 (cleaned up)—because it found that that digit “provides little or no relevance . . . .” The court did not err. Future Proof gave the district court one piece of evidence to prove the degree of care: A 12-pack of Brizzy sells for the “low cost” of $14.99. Although we have concluded that, “[w]here items are relatively inexpensive, a buyer may take less care in selecting the item,” id. at 458 (emphasis added), we have never concluded that a low price is sufficient to establish a dearth of care. In fact, we often rely on affidavits or testimony to show a lack of consumer care. See, e.g., Xtreme Lashes, 576 F.3d at 231; Smack Apparel, 550 F.3d at 483. Future Proof provides no such evidence. Future Proof moreover urges that consumers make “quick decisions” about hard seltzers “in a crowded array, often in a crowded bar or restaurant.” But Future Proof provides no affidavits, testimony, or other evidence supporting that proposition. In fact, Future Proof provides the pricing information for its 12-packs in “retail locations” like “H-E-B” or “Circle K”— not for individual cans a consumer might order at a bar or restaurant. Without evidence that consumers typically make snap decisions to purchase Brizzy in teeming bars and restaurants, the district court correctly concluded the eighth digit does not favor granting the injunction. IV. Because a finding of a likelihood of confusion “need not be supported by a majority” of the digits and each digit “may weigh differently from case to case,” Streamline Prod., 851 F.3d at 453 (cleaned up), we review the court’s ultimate conclusion about likelihood of success for clear error. See Am. Rice, 518 F.3d at 328. In sum, the district court concluded that three digits supported the injunction and one weighed “marginally in favor of granting the injunction . . . .” But the court correctly concluded that the other four factors did not support the injunction. And, notably, the court correctly concluded that the two digits that have special importance, namely the sixth—which “may alone be sufficient to justify an inference that there is a likelihood of confusion,” Streamline Prod., 851 F.3d at 455—and the seventh—which constitutes the “best evidence of a likelihood of confusion,” Viacom Int’l, 891 F.3d at 197 (cleaned up)—did not support the injunction. The court did not commit clear error in concluding that Future Proof failed to carry its burden of showing a likelihood of success on the merits. Therefore, the decision isn’t one of the “extraordinary circumstances,” in which a district court so clearly erred that we will “reverse the denial of a preliminary injunction.” Anderson, 556 F.3d at 355–56. The court did not abuse its discretion. AFFIRMED.
