MEMORANDUM OPINION SETTING OUT FINDINGS OF FACT AND CONCLUSIONS OF LAW1

Hon. Keith P. Ellison, United States District Judge.

In this trademark infringement action, Plaintiffs ("UH") move for a preliminary injunction to prevent Defendant Houston College of Law (formerly known as South Texas College of Law, and referred to herein as "Defendant") from using the mark "HOUSTON COLLEGE OF LAW" as a name, mark, or source identifier for its legal education services.2

Based on the pleadings, the numerous briefs and submissions,3 the arguments and evidence presented at a hearing on the motion, and the applicable law, the Court enters the findings of fact and conclusions of law set forth below. Based on the findings and conclusions, the Court GRANTS UH's motion for a preliminary injunction.

I. BACKGROUND

The University of Houston is a public university founded in 1927 that provides undergraduate- and graduate-level courses in a variety of academic disciplines. It is ranked as a Tier 1 research university in the Carnegie Research University Rankings, serves more than 42,000 students annually, and has achieved national recognition for both its football and basketball programs. Its campus is located just outside downtown Houston.

UH's law school was founded in 1947 and was known as the University of Houston College of Law until 1967, when it became the Bates College of Law. This name was relatively short-lived, however, as another name change occurred in 1982 when UH became the University of Houston Law Center ("UHLC"). It has been known and marketed as such ever since. UH is ranked as the 50th best law school in the country in the U.S. News and World Report rankings.

Defendant, located in downtown Houston, is a private law school that was founded in 1923 as South Texas College of Law and Commerce. It exists on its own, which is to say that it is not owned by or materially affiliated with a university or other colleges. For most of its existence, the law school was known as South Texas College of Law, and more recently as South Texas College of Law/Houston. It is unranked in the U.S. News rankings.

On June 22, 2016, South Texas College of Law announced that it was changing its name to "Houston College of Law." UH publicly voiced its opposition to the name change almost immediately thereafter, expressing its concern about the potential for "significant confusion" between UH and Defendant's newly-branded law school. Five days after Defendant announced its name change, UH filed a complaint alleging trademark infringement, among other claims.

II. LEGAL STANDARDS

A. The Preliminary Injunction Standard

A court may grant an application for a preliminary injunction "only if the movant establishes (1) a substantial likelihood of success on the merits, (2) a substantial threat of irreparable injury if the injunction is not issued, (3) that the threatened injury if the injunction is denied outweighs any harm that will result if the injunction is granted, and (4) that the grant of an injunction will not disserve the public interest."4 "Although the grant or denial of a preliminary injunction rests in the discretion of the trial court, `[the Fifth Circuit has] cautioned repeatedly that a preliminary injunction is an extraordinary remedy which should not be granted unless the party seeking it has clearly carried the burden of persuasion on all four requirements.'"5

B. The Lanham Act Standard

The Lanham Act makes liable "[a]ny person who ... uses in commerce any word, term, name, symbol, or device,... which ... is likely to cause confusion, or to cause mistake ... as to the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person."6 To obtain an injunction for trademark infringement, a party must show that (1) the claimed mark is eligible for protection, (2) the party seeking protection is the mark's senior user, (3) there is a likelihood of confusion between the plaintiff's mark and the defendant's mark, and (4) this likelihood of confusion will cause the plaintiff irreparable injury for which there is no adequate legal remedy.7 The parties do not dispute that, to the extent UH's claimed marks are eligible for protection, UH is the senior user.

III. LIKELIHOOD OF SUCCESS ON THE MERITS

Analysis for trademark infringement under the Lanham Act proceeds in two steps. "[A] party must first show that it has a protectable right in the mark and, second, show that there is a likelihood of confusion between the marks.8

A. Protectable Right in the Claimed Marks

UH seeks protection of both federally registered marks and common law marks. The Court will address each type of mark in turn.

(1) UH's Registered Trademarks

UH has presented proof that it owns Reg. No. 0,747,078, Reg. No. 3,025,231, and Reg. No. 4,116,569, each of which registers the word mark "UNIVERSITY OF HOUSTON."9 "Proof of registration of a service mark or trademark is ... prima facie evidence of the registrant's exclusive right to use the mark in commerce for the services specified in the registration."10 Here, Defendant does not challenge UH's exclusive right to use this mark, but argues that the services specified in the registration are narrower than UH alleges.11

The UNIVERSITY OF HOUSTON mark is the subject of numerous federal registrations, and the services specified in each registration vary. Although Defendant correctly notes that Reg. Nos. 0,747,078 and 4,650,772 are not directed at "educational services themselves," Defendant fails to raise a similar objection — or even mention — Reg. No. 4,116,569,12 which expressly provides for use of the mark in the context of "educational services, namely, providing college and graduate level courses of instruction, continuing education courses, and seminars in the ... legal... field[]."13 By Defendant's own implicit admission,14 this is precisely the context in which UH is bringing the instant litigation. The Court therefore holds that UH possesses a protectable right in the "UNIVERSITY OF HOUSTON" mark.

(2) UH's Claimed Common Law Trademarks

UH also alleges that it holds several common law (i.e., federally unregistered) word marks and "color marks" that are eligible for protection, including "UNIVERSITY OF HOUSTON LAW CENTER," "HOUSTON LAW," and the colors red and white.15

"To determine whether a word or phrase is protectable, it must first be determined into which category, (1) generic, (2) descriptive, (3) suggestive, or (4) arbitrary or fanciful, the word or phrase belongs."16 Generic marks are never eligible for protection, but here, Defendant concedes that the claimed marks are descriptive.17 "Descriptive terms may only be protected after proof that they have acquired secondary meaning."18 "Such secondary meaning is achieved when, in the minds of the public, the primary significance of a product feature or term is to identify the source of the product rather than the product itself."19

UH has provided evidence that the UNIVERSITY OF HOUSTON LAW CENTER mark has acquired secondary meaning,20 and Defendant seems to concede as much in the briefing.21 The Court therefore finds that the mark has acquired secondary meaning and concludes that the mark is protectable. The protectability of UH's other alleged common law marks need not be addressed because, as discussed below, the Court finds that Defendant's mark creates a likelihood of confusion on the basis of the UNIVERSITY OF HOUSTON and UNIVERSITY OF HOUSTON LAW CENTER marks alone.

(B) Likelihood of Confusion

To demonstrate a likelihood of success on the merits of its trademark infringement claim, UH must show that Defendant's new mark, HOUSTON COLLEGE OF LAW, is likely to cause confusion with one of UH's protectable marks.22 "Likelihood of confusion is synonymous with a probability of confusion, which is more than a mere possibility of confusion."23 A finding of a likelihood of confusion is a finding of fact.24 In making this finding, the Court considers a non-exhaustive list of so-called "digits of confusion," including:

"The digits may weigh differently from case to case, depending on the particular facts and circumstances involved."26 "No one factor is dispositive, and a finding of a likelihood of confusion does not even require a positive finding on a majority of these digits of confusion."27 "In addition to the listed factors, a court is free to consider other relevant factors in determining whether a likelihood of confusion exists."28

Important to note at the outset is the precise nature of UH's argument. UH alleges not only that prospective purchasers (i.e., students considering law school) will confuse Defendant with UHLC, they argue more broadly that prospective purchasers are likely to believe that Defendant and the University of Houston are somehow associated.29 Likelihood of confusion analysis must be conducted with this broader argument in mind.

(1) The Strength of UH's Marks

The first digit of confusion raises the question of whether UH's marks are strong or weak. Strength of a trademark is determined by two factors.30 Courts first examine the mark's conceptual strength by evaluating whether the mark should be classified as "generic, descriptive, suggestive, or arbitrary and fanciful."31 The strength of a mark — and, by extension, the protection that it is afforded — increases as one moves away from generic and descriptive marks toward arbitrary marks.32 The parties seem to agree that the marks at issue here are descriptive,33 and the Court agrees with their assessment.34 The marks are therefore relatively lacking in conceptual strength.

But "[t]he ultimate strength of the mark, the key inquiry before us, is determined by a number of factors which establish its standing of the mark in the marketplace."35 In other words, the second factor that courts consider — and the more important of the two — is a mark's commercial strength.36 "Marks may be strengthened in the marketplace by extensive advertising, length of time in business, public recognition, and uniqueness."37 UH has introduced evidence indicating that:

This evidence suggests that the commercial strength of UH's marks is relatively strong,44 and that this is especially true within the legal industry and geographic markets most relevant to this litigation — both UH and Defendant overwhelmingly target prospective students in just two states, Texas and Florida.45

The Fifth Circuit has recognized, however, that extensive third-party use of a similar mark can be "impressive evidence that there would be no likelihood of confusion."46 Moreover, district courts must consider all third-party use of a mark — even if the use occurs outside of the plaintiff's particular industry — in determining whether the plaintiff's mark is strong or weak.47 "The logic behind this rule is that customers [may] have become so conditioned by a plethora of such similar marks that customers have become educated to distinguish between different marks on the basis of minute distinctions."48 Based on this logic, argues Defendant, UH's marks should be deemed relatively weak; more than 25,000 registered businesses use the word "Houston" in their names.49

But multiple courts in this district have acknowledged that, although the Fifth Circuit requires a broad examination of all third-party usage, it does not necessarily require fact-finders to assign equal weight to all third-party usage.50 And common sense dictates that the weight will vary significantly based on the characteristics of the third-party user and the extent of its use.51 For example, here, although Defendant identifies a number of Houston-based institutions of higher learning that use either "University" or "Houston" in their name,52 none has law schools and there is no evidence that any are well known in the marketplace. To the extent consumers are unaware of third-party use, the logic behind the third-party use rule is inapplicable; the consumers have not been conditioned to distinguish among the marks. Thus, although Defendant's evidence of third-party use weakens the strength of UH's marks, it is only to a limited degree.

The Court finds that the UNIVERSITY OF HOUSTON and the UNIVERSITY OF HOUSTON LAW CENTER marks are somewhat strong — particularly in the markets relevant to this litigation — but not overwhelmingly so. This digit of confusion therefore weighs in favor of finding a likelihood of confusion.

(2) The Similarity of the Marks

"The Fifth Circuit uses a `subjective eyeball test' to determine whether two marks are similar."53 In other words, "The similarity of design is determined by considering the overall impression created by the mark as a whole rather than simply comparing individual features of the marks."54 In making this determination, courts consider "the marks' appearance, sound, and meaning."55

The appearance of the UNIVERSITY OF HOUSTON LAW CENTER and HOUSTON COLLEGE OF LAW marks are strikingly similar. As an initial matter, two of the three words in Defendant's mark appear in UH's mark ("Houston" and "Law"), which is a noteworthy fact in and of itself.56 Far more troubling, however, is the way in which Defendant deploys its mark in the marketplace.57 Consider Defendant's logo, for example, which is prominently featured in most (if not all) of the promotional material in evidence. As is true of the UHLC logo, Defendant's logo uses block letters,58 emphasizes the word "HOUSTON,"59 and utilizes a red and white color scheme.60 Indeed, these features are ubiquitous throughout Defendant's marketing materials.

Defendant identifies several relatively small differences in the two logos, but none of the distinctions is sufficient to alter the "overall impression" that the logos elicit. For example, Defendant's logo includes the phrase "EST. 1923" in relatively small font.61 Smaller still — and not always present62 — is the font that states "FORMERLY SOUTH TEXAS COLLEGE OF LAW."63 And the generic image of the scales of justice in Defendant's logo does little to dispel confusion between the marks of two law schools,64 particularly given that the image is displayed in red and white. Indeed, rare is the case where a defendant has made an exact copy of the plaintiff's mark. As one district court has noted, "the most successful form of copying is to employ enough points of similarity to confuse the public, with enough points of difference to confuse the courts."65

But even if the Court assumed that "prospective purchasers [would] recognize that the two designations are distinct" in their appearance, "confusion may [still] result if purchasers are likely to assume that the similarities in the designations indicate a connection between the two users."66 And here, the meaning of the parties' marks makes potential purchasers likely to assume such an affiliation between Defendant, on the one hand, and UH and/or UHLC on the other.

The meanings of the marks HOUSTON COLLEGE OF LAW and UNIVERSITY OF HOUSTON LAW CENTER are practically identical, and this alone presents a source of potential confusion.67 Compounding this confusion — and making it far more difficult to dispel — are the interrelated meanings of Defendant's mark and UNIVERSITY OF HOUSTON. Universities often serve as umbrella organizations to multiple colleges that are each responsible for educating students within certain academic disciplines. The University of Houston is no exception. In addition to housing the Law Center, UH is home to, for example, the University of Houston College of Arts, the University of Houston College of Education, and the University of Houston College of Pharmacy. "Houston College of Law" fits almost perfectly within this framework, creating a substantial risk that potential purchasers will "think [Defendant's] services [have] some connection with [UH]."68 This is particularly true when the HOUSTON COLLEGE OF LAW mark is prominently displayed in the red and white colors commonly associated with UH.

Based on the strong similarities in both appearance and meaning, the Court finds that the second digit of confusion weighs very heavily in favor of finding a likelihood of confusion.69

(3) The Similarity of the Products or Services

"The greater the similarity between [the parties'] services, the greater the likelihood of confusion,"70 and here, UHLC and Defendant offer practically identical services.71 Defendant attempts to distance itself from the services offered by UH by noting that the University has a broad array of athletic programs and non-legal educational opportunities while Defendant has none. But this distinction is of no moment. UH alleges trademark infringement based in part on an affiliation-confusion theory. In other words, UH is not alleging that consumers will mistake Defendant for the University of Houston, but rather that consumers will mistakenly associate Defendant with the University of Houston (e.g., by assuming that Houston College of Law is one of the many colleges under UH's umbrella). The fact that, at a university level, UH participates in intercollegiate athletics and offers non-legal educational opportunities does not dispel association-based confusion. To the contrary, it is perfectly consistent with it.

The Court finds that the third digit of confusion weighs heavily in favor of finding a likelihood of confusion.

(4) The Identity of the Retail Outlets and Purchasers

"Differences in the parties' customer bases can lessen the likelihood of confusion."72 There is no dispute that the parties target the same segments of the market and provide their services in the same geographic area — the campuses are separated by only three miles.73 The Court finds that this digit of confusion weighs heavily in favor of finding a likelihood of confusion.

(5) The Similarity of the Advertising Media Used

"Courts also consider the similarity between parties' marketing efforts: the greater the degree of overlap in the marketing approaches of the two entities, the greater the likelihood of confusion."74 Defendant is forced to concede that it will use channels of advertising similar to those employed by Plaintiffs.75 The Court finds that this digit of confusion weighs heavily in favor of finding a likelihood of confusion.

(6) Defendant's Intent

"Proof of an intent to confuse the public is not necessary to a finding of a likelihood of confusion," but "[i]f a mark was adopted with the intent to confuse the public, that alone may be sufficient to justify an inference of a likelihood of confusion."76 "[A] junior user's knowledge or awareness of the senior user's trademark" is insufficient to create an inference of intent.77 Instead, the proper inquiry "is whether [the junior user] had the intent of deriving benefit from the reputation or goodwill of [the senior user]."78 "Direct proof of bad faith is rarely present," so courts must often base their decision on circumstantial evidence.79

Defendant contends that the decision to rebrand itself as Houston College of Law was driven by a desire to align its name with its location — Houston — in the minds of consumers. As evidence, the law school cites to the results of an extensive market survey that it commissioned in 2013 (the "Simpson Report") to explore the market's perception of "the South Texas College of Law name."80 In particular, Defendant emphasizes the Simpson Report's finding that many respondents were in favor of a change because the name "South Texas College of Law" can lead people to mistakenly believe that the school is located in the Rio Grande Valley.81 When respondents were asked to provide suggestions for a new name, the most frequently mentioned suggestion (32%) was to "include [a] reference to the location in Houston."82

UH, on the other hand, contends that the Simpson Report actually cuts against Defendant's position. While true that some respondents were concerned by the law school's association with the Valley, that concern was far from the most frequently mentioned response in the survey. Instead, "The most frequently mentioned reason" for changing the law school's name was "that the name South Texas College of Law is often confused with other schools, particularly Texas Southern University."83 The Simpson Report further notes that the confusion with Texas Southern "detracts from STCL's prestige and national reputation."84 Based on this evidence, UH suggests that Defendant's name change was intended to distance itself from Texas Southern while manufacturing a perceived affiliation with UH — one that would enhance Defendant's prestige rather than detract from it.

To bolster this point, UH presents strong evidence that affiliating with a major university — or generating the mistaken perception of an affiliation — would be beneficial to Defendant. And again, UH turns to the Simpson Report on which Defendant so heavily relies: "The second most frequently mentioned response in favor of [a] name change [is] if [South Texas] affiliates with a university such as Rice or Texas A & M."85 Indeed, Defendant has actually discussed this prospect with several universities throughout the past two decades,86 and Dean Guter acknowledged at his deposition that "using the University of Houston or Texas A & M University or Rice ... name or affiliating with one of those institutions" could provide benefits such as "broader exposure"87 and "attracting [more] students."88

The benefits of affiliation were thrown into sharp relief for Defendant when the entity formerly known as Texas Wesleyan University Law School leapfrogged South Texas in the U.S. News rankings shortly after affiliating with Texas A & M.89 In fact, 2016 was the first time in the history of the school that Texas Wesleyan had been ranked, and the first time in three years that South Texas had fallen out of the rankings entirely.90 Dean Guter testified that he fielded several calls from dissatisfied South Texas board members and alumni following this development.91 According to UH, it was shortly after this sequence of events — and three years after the Simpson Report — that South Texas decided to change its name to Houston College of Law.

If this were the only evidence of intent before the Court, resolution of this issue would be relatively straightforward. Although it has conceptual merit, UH's theory is rooted in highly circumstantial evidence that would be insufficient to meet the substantial burden imposed at this procedural stage. But UH presents additional evidence that more directly calls Defendant's intent into question: in connection with the name change, it appears that Defendant adopted a new red and white color scheme that very closely resembles the University of Houston's.

Defendant notes that its official school colors are red and gold, and have been since 1973.92 The school's use of color in practice, however, tells a different story. To the extent that Defendant has used red in the past — even its use of red appears somewhat inconsistent throughout the years93 — the shade has been a dark crimson, and the crimson has typically been accompanied by the color gold in some form or fashion.94 Defendant's Houston College of Law logo now consists exclusively of red and white, and its promotional materials similarly feature red and white as the overwhelmingly dominant colors. According to Defendant, it began employing white (rather than gold) because it provides a better contrast for red lettering.95 Regardless, the similarity to UH's color scheme is striking.

Even more troubling is the shade of red that Defendant now employs. UH has presented numerous images of Defendant's Houston College of Law promotions next to its South Texas College of Law merchandise, and the current shade of red is unmistakably brighter than the classic South Texas crimson.96 Instead, it appears more reminiscent of the hue that has long been employed by UH.

At the hearing, Defendant argued that the varied shade of red was the result of using a variety of vendors and inconsistent paper quality.97 This argument is not entirely without precedent. A court in this district recently acknowledged that colors can vary with the material on which they are printed, but the court's conclusion was expressly based on "credibl[e] testi[mony]" from a witness as well as internal documents corroborating the testimony. Both are absent here.98

Based on the above-referenced evidence, the Court finds that Defendant at least knew of the likelihood that its new mark — particularly dressed in the new red and white color scheme — was likely to cause a mistaken association with UH. But awareness is not sufficient to establish intent, and, though a close call, the Court cannot find that Defendant acted with the "specific intent" of "deriv[ing] benefit from the reputation or goodwill of plaintiff."99 Defendant's rationale for emphasizing "Houston" in its name is entirely plausible, and the Court is wary of relying too heavily on select snapshots of promotions and merchandise produced by various vendors on various types of materials. The Court therefore finds that this factor does not weigh in favor of either party.100

(7) The Evidence of Actual Confusion

"Evidence of actual confusion is not necessary to a finding of a likelihood of confusion, but it is nevertheless the best evidence of a likelihood of confusion."101 "Moreover, reason tells us that while very little proof of actual confusion would be necessary to prove the likelihood of confusion, an almost overwhelming amount of proof would be necessary to refute such proof."102 In fact, the Fifth Circuit has affirmed a district court's bench-trial finding of actual confusion based on a single known instance.103

Although point-of-sale confusion is the "most common and widely recognized type of confusion that creates infringement," it does not "mark the outer boundaries of trademark infringement."104 Trademark infringement can also "be based upon confusion that creates initial consumer interest, even though no actual sale is finally completed as a result of the confusion."105 This type of confusion, known as "initial-interest confusion," gives "the junior user credibility during the early stages of a transaction and can possibly bar the senior user from consideration by the consumer once the confusion is dissipated."106

"To show actual confusion, a plaintiff may rely on anecdotal instances of consumer confusion or consumer surveys."107 UH relies on both.

(i) Consumer Surveys

UH's survey expert, Hal Poret, surveyed two hundred law students using the well-accepted Eveready methodology,108 and found a net confusion rate of 25%.109 Defendant takes issue with certain aspects of Mr. Poret's methodology, and instead touts the survey of its own expert, Gabriel Gelb. Mr. Gelb found a net confusion rate of only 6%. Thus, as is so often the case in high-stakes litigation, highly qualified experts have presented dueling reports that reach significantly different results. The Court must parse the methodology of each to determine which is deserving of greater evidentiary weight.

"In assessing the validity of a survey, courts look to two factors: first, the manner of conducting the survey, including especially the adequacy of the universe; and second, the way in which participants are questioned."110 Defendant makes no objection to the universe of respondents in the Poret Survey,111 but lodges a strong objection to the way in which the survey participants in the "Webpage Test Group" were questioned. These participants were shown an image that was identical to the Houston College of Law homepage, but with one exception: the image omits two banners that rotate prominently across the webpage ("South Texas College of Law Changes to Houston College of Law" and "Houston College of Law Stands Behind Name Change; Is Prepared to Defend Decision in Court").112 Defendant argues Mr. Poret's survey should therefore be given no weight because it failed to "test the alleged infringing use as it's actually seen in the real world today."113 The Court disagrees.

Defendant's use of the alleged infringing mark "in the real world today" comprises far more than its use of the mark on one page of one website. To the contrary, UH has presented evidence that Defendant has already begun aggressively marketing its new name by advertising on large billboards on major Houston highways,114 sending out mailers to prospective law students and members of the legal community,115 and selling merchandise bearing its new name and logo.116 None of those uses contains the purported disclaimers on which Defendant's argument so heavily relies. Further, Defendant's argument even overstates the prevalence of the two banners on the Houston College of Law website. The banners are apparently limited to the website's home page, meaning that anyone who clicks a direct link from Google to the "Admissions" page, for example, would never see them. And even these two banners are not always displayed on the home page; a third, unrelated banner rotates along with them.117

Mr. Gelb's survey fails to take this consideration into account, and the weight of his findings is diminished as a result. In fact, the only image of the webpage that respondents saw in Mr. Gelb's survey included the prominent "South Texas ... Changes to Houston College of Law" banner. His omission of the other two rotating banners is somewhat surprising given his criticism of Mr. Poret's decision to alter the presentation of the same subject matter.118

Moreover, to the extent that the two banners on Defendant's homepage are the only readily available means of dispelling consumers' initial confusion, Defendant's position is all the more troubling. While those banners are admittedly a prominent feature of the webpage, the most prominent feature of its homepage is a series of eight rotating banners that tout Defendant's primary selling points.119 In other words, even if a consumer's initial-interest confusion only persists long enough to lead him to the homepage, then Defendant has "br[ought] the patrons in the door.... [T]he confusion has succeeded."120 Defendant is able to inform prospective purchasers of its "best value" status, acclaimed advocacy program, and "high bar-passage rate," for example, and is able to do so only because of the brand recognition and goodwill that UH has "expensively and carefully built" over decades.121

The Court concludes that, while neither survey is without flaw, Mr. Poret's is substantially stronger.122 Further, precedent in this district indicates that Mr. Poret's finding of a 25% net confusion rate among prospective purchasers is significant and provides support for finding that this digit of confusion weighs in UH's favor.123

(ii) Anecdotal Instances of Consumer Confusion

UH has documented numerous instances of actual confusion. These instances include the following:

Defendant argues that these instances of actual confusion are irrelevant to the extent that they do not involve prospective law students. But this position is at odds with case law in this district. While true that "`[t]rademark infringement occurs only when the use ... is likely to confuse purchasers,'"132 courts and commentators have also repeatedly recognized that "evidence of confusion in others permits the inference of confusion in purchasers."133 In other words, evidence of confusion in non-purchasers is relevant to the extent that it proves actual purchasers would be similarly confused by the junior user's mark.

The first two instances of confusion are deserving of relatively little weight — they involved individuals who are unfamiliar with the legal education industry.134 The next four instances are significantly more noteworthy, however, as they evince mistakes made by individuals who are active participants in the field. Most important, the last two anecdotes are instances of actual confusion by prospective law students, and even suggest that the confusion is not quickly dispelled. One involved a prospective law student who assumed that Defendant was affiliated with UH even after visiting Defendant's booth and speaking with its representative. Similarly, it seems reasonable to presume that the prospective student in the other anecdote was well into the process of researching law schools — and Defendant in particular — if she was calling about Defendant's $55 application fee. The fact that confusion could persist at the point of paying to apply for admission is particularly significant in the context of initial-interest confusion.135

The Court finds that this digit of confusion weighs heavily in favor of finding a likelihood of confusion. UH has provided relatively cogent survey evidence as well as strong anecdotal examples of actual confusion. The quantity of anecdotes is made all the more compelling given the short amount of time since Defendant's name change and the summer lull in the law school application cycle.136 This constitutes far more than the "very little proof of actual confusion" that the Fifth Circuit has required to tip this digit of confusion in plaintiffs' favor.137

(8) Degree of Care Exercised by Potential Purchasers

The eighth digit is the degree of care exercised by purchasers. "Where [services] are relatively inexpensive, a buyer may take less care in selecting [them], thereby increasing the risk of confusion."138 But if the items are expensive and the buyers are sophisticated, then confusion is less likely to occur.139 Nonetheless, "a high price tag alone does not negate other indicia of likelihood of confusion, especially if the goods or marks are similar,"140 and "even a sophisticated purchaser can be subject to initial interest confusion."141

There can be no dispute that the cost of a legal education is tremendous, nor can there be much debate regarding the relative sophistication of prospective law students at the point of sale — but only at the point of sale. Prospective law students are not endowed with an inbuilt knowledge of the legal education industry. It is only after their interest in legal education is first piqued that they begin the process of becoming sophisticated. In other words, there exists a period of time in every prospective law student's career where, not only is he unsophisticated, he knows practically nothing about the industry and is particularly susceptible to confusion.

The Court finds that this digit of confusion weighs heavily against finding a likelihood of point-of-sale confusion. At best, however, it weighs only slightly in Defendant's favor as it relates to initial-interest confusion.

(9) Weighing the Digits of Confusion

Before weighing the digits of confusion, the Court must calibrate the scale. In the context of point-of-sale confusion, this inquiry is relatively straightforward: the Court weighs the digits of confusion to determine whether a prospective purchaser will likely be confused as to a product's source at the time he purchases the product. The concept of initial-interest confusion is more challenging to delineate, and the Court requested supplemental briefing from the parties on this subject.

(i) The Initial-Interest Confusion Doctrine

The Fifth Circuit first held that a trademark infringement claim "can be based upon confusion that creates initial consumer interest" in Elvis Presley Enterprises, Inc. v. Capece.142 This holding was particularly noteworthy because, unlike in cases involving point-of-sale confusion, it provided that a claim of infringement can be valid "even though no actual sale is finally completed as a result of the confusion."143 Instead, the court expressed concern that "[i]nitial-interest confusion gives the junior user credibility during the early stages of a transaction," which "can possibly bar the senior user from consideration" down the road.144

Defendant argues that the Fifth Circuit's holding in Elvis was driven by the plaintiff's evidence that consumers were likely to purchase something from the defendant once they entered his bar.145 But, while this evidence may have been relevant to the court's decision, there is nothing to suggest that it was a necessary factor — the court mentioned it only in passing,146 and did so directly on the heels of holding that no sale need actually occur. Rather, Elvis focused on the concern that the misappropriated credibility of the senior user would "bring[] patrons in the door."147 "Once in the door, the confusion has succeeded."148 In other words, Elvis suggests that the initial-interest confusion doctrine is intended to address situations where a junior user trades on the goodwill of the senior user for its own financial gain.149

This interpretation is supported both by the authority to which Elvis cites, as well as by cases that have cited to it. Elvis cited to the Second Circuit's decision in Mobil Oil, for example, which held that the likelihood of initial interest confusion was based "not in the fact that a third party would do business with [the junior user], but rather in the likelihood that [the junior user] would gain crucial credibility during the initial phases of a deal."150 "[A]n oil trader might listen to a cold phone call from [the junior user] ... when otherwise he might not, because of the possibility that [the junior user] is related to [the senior user]."151 Courts in several other circuits have held similarly.152

More recently, a court in this district granted a preliminary injunction based, at least in part, on the potential for initial-interest confusion. Again, the court's focus was on the fact that the junior user would "benefit from the consumer's belief that [the junior user] is affiliated with the [senior user]," and that the consumer "will impute ... the goodwill that [the senior user] has expensively and carefully built since 2002."153

(ii) Weighing the Digits of Confusion

The first factor favors finding a likelihood of confusion, and the second, third, fourth, fifth, and seventh digits of confusion weigh heavily in favor of such a finding. Only the eighth factor cuts against UH's case, and the sixth factor weighs in no one's favor at all. This strongly suggests that prospective law students are likely to assume that Houston College of Law is affiliated with the University of Houston.154 Although this confusion will likely dissipate by the point of sale, this type of confusion implicates the very concerns that the initial-interest confusion doctrine is intended to address.

Defendant cautions against reading this doctrine too broadly, and their admonition is well taken. As the 6th Circuit noted in Groeneveld, "[W]hat appears to concern Groeneveld is not so much initial-interest confusion, but initial interest, period."155 "Simply invoking the term `initial-interest confusion' does not state a viable claim."156 This Court agrees; more is required.

But Defendant fails to acknowledge that, here, more is provided. Prospective students are likely to further investigate Houston College of Law not necessarily because of their initial interest in the law school, as Defendant suggests, but rather because the mark seemingly bears the imprimatur of UH's well-known brand — in other words, because of initial-interest confusion.157 And not only is Defendant benefitting from the goodwill of a stronger brand,158 the stronger brand is that of a direct competitor.159 UH and Defendant offer the same set of services through the same media to the same potential purchasers in practically the same location. In other words, not only does the confusion afford Defendant credibility during the early stages of a transaction, it does so at the direct expense of the University of Houston. Indeed, while Defendant notes that consumers' confusion will dissipate as soon as they visit Defendant's homepage, the most prominent portion of the webpage is essentially a list of the best reasons to choose Defendant's law school over UH's.160

Notwithstanding the opposing digits of confusion, Defendant contends prospective law students exercise such a high degree of care that they are highly unlikely to choose a law school while still confused as to the school's identity. That may be true161 — purchasing a legal education typically (and hopefully) involves months of thorough research — but this fact alone does not preclude finding a likelihood of confusion prior to the purchase (i.e., initial-interest confusion).162 If it did, sellers of goods or services that involve extended purchasing processes would be effectively outside the ambit of the Lanham Act's protection, leaving competitors free to appropriate the senior user's goodwill with impunity, and allowing them to gain "credibility during the early stages of a transaction."163 And it is in the early stages of the transaction when prospective law students are the least sophisticated and most susceptible to confusion, especially when such similar marks are involved.

Defendant additionally attempts to counterbalance the opposing digits of confusion by noting that it did not act with intent. According to Defendant, cases finding a likelihood of initial interest confusion commonly involve instances where the junior user intentionally adopted its mark to capitalize on the confusion.164 But, while intent may be a common thread among the cases to which Defendant cites, nothing in those cases indicates that intent is a necessary showing in the context of initial-interest confusion cases — especially where, as here, the other digits of confusion are so overwhelmingly one-sided. If anything, Defendant's Fifth Circuit precedent speaks directly to the contrary: the Elvis court expressly held that "[p]roof of an intent to confuse the public is not necessary to a finding of a likelihood of confusion."165 Moreover, intent is a notoriously challenging element to prove under even the best of circumstances.166 The Court would be reluctant to place undue weight on the "intent" digit of confusion, particularly in the context of a preliminary injunction proceeding where, as here, the plaintiff is handcuffed by an expedited timeline and limited discovery.

It also bears repeating that UH's inability to conclusively prevail on this digit of confusion does not mean that it weighs in Defendant's favor. As the Fifth Circuit noted in Xtreme Lashes, even "with no evidence of [defendant's] intent, this factor is neutral."167 And here, UH has presented considerable evidence that Defendant intended to derive a benefit from UH's goodwill.168 Although the evidence was insufficient for the Court to find intent as a matter of fact, the Court reached this finding by a very narrow margin. In sum, Defendant attempts to counterbalance the six digits of confusion that weigh heavily against them with a digit of confusion that, at best, weighs in no one's favor at all.

The Court finds that the digits of confusion weigh heavily in favor of finding a likelihood of confusion, and that UH has therefore clearly carried its burden of establishing a substantial likelihood of success that it will prevail on the merits of its trademark infringement claim.

IV. IRREPARABLE HARM

UH argues that, in the Fifth Circuit, showing a likelihood of confusion presumptively shows a substantial threat of irreparable harm. Defendant disagrees, and, as it has done throughout this litigation, provides several cogent and well-researched arguments to the contrary. But the disagreement need not be addressed here. Irrespective of the presumption, the Court concludes that monetary damages will not adequately compensate UH, and that UH will be irreparably harmed by Defendant's continued use of the "Houston College of Law" mark to identify and market its brand.

A court in this district recently held that a likelihood of initial-interest confusion would cause irreparable injury to the senior user in two ways, both of which apply here.169 First, UH's lack of control over the quality of Defendant's conduct — conduct that prospective law students will likely attribute to UH by mistake — constitutes an irreparable injury.170 In fact, UH's evidence of actual confusion exemplifies this consideration. For example, Defendant's law professors may speak to audiences that include prospective law students, and Defendant's recruiting department has in fact attended school fairs at which its representatives directly interacted with prospective law students. Second, UH's "time, effort, and expense exerted to create and define its brand has been unfairly exploited," which monetary damages cannot compensate.171

Finally, the Court cannot help but note the peculiar context of Defendant's contention that no irreparable harm will occur from their infringement. Defendant insists that the findings of the Simpson Report were the impetus for its recent name change. The Report found that "the most frequently mentioned reason" that respondents were in favor of a name change was that "`South Texas College of Law' is often confused with other schools, particularly Texas Southern University."172 It must be with a great sense of irony that Defendant now attempts to downplay the effects of the same type of affiliation confusion that prompted Defendant to spend hundreds of thousands of dollars to rebrand itself.

V. BALANCE OF HARDSHIPS

The third requirement for obtaining a preliminary injunction requires the plaintiff to establish that his irreparable harm is greater than the hardship that the preliminary injunction would cause the defendant.173 Defendant argues that it incurred substantial costs in conjunction with its name change, and that it would similarly incur substantial costs if compelled to revert to South Texas College of Law. Dean Guter was quoted on June 23, 2016, as stating, "I feel safe in saying we haven't spent $35,000 to $40,000 extra over anything we would have spent anyway. And so the biggest cost that we see going forward is changing the external signage."174 Those subsequent changes proved costly. Diane Summers, Defendant's Director of Marketing and Communications, submitted a sworn declaration indicating that UH incurred approximately $458,000 in additional costs to publicize the name change.175 Ms. Summers also notes that her staff had spent considerable time preparing for the name change in the months leading up to the rebranding.

But Defendant's costs, while substantial, engender little sympathy when viewed in the broader context of Defendant's conduct. A party's claim of hardship is given less weight if he chooses "to incur costs in the face of a possible infringement claim,"176 and here, a law suit was practically inevitable. Dean Guter essentially admitted that Defendant changed its name knowing that UH would sue to stop it.177 Indeed, as the Court has already found, Defendant was at least aware that its new mark would likely cause prospective purchasers to mistakenly associate Defendant with UH. It consequently must have come as no surprise when UH promptly released a statement on June 23rd — one day after Defendant's announcement — expressing its concern about the "significant confusion" between UHLC and Defendant's newly-branded law school.178 Defendant soon boldly proclaimed that it "Stands Behind its Name Change" and "Is Prepared to Defend Decision in Court" on the homepage of its website. Yet it was only after news of UH's objection that Defendant evidently incurred the overwhelming majority of these costs, and destroyed "[m]uch of the older stationary and signage bearing the name `South Texas College of Law.'"179 In other words, Defendant opted to double down, yet cites to the high stakes of the game as a reason to call off the bet.

As explained above, UH has much to lose if the injunction is not granted. Defendant has spent hundreds of thousands of dollars marketing its new name, and its market presence has been (and will continue to be) enhanced on the strength of UH's well-recognized brand. If an injunction does not issue, this harm will go unaddressed until a trial on the merits, which is unlikely to take place prior to the close of this application cycle. Moreover, as T-Mobile held on a similar set of facts, "The harm to [UH's] brand, which it has spent [millions] of dollars and [many decades] creating, substantially outweighs the ... harms that [Defendant] will suffer in stopping the use" of the HOUSTON COLLEGE OF LAW mark.180 And whatever the harm, Defendant's own actions suggest that the harm caused by affiliation-confusion is significant enough to scrap a 93-year-old brand to avoid it.

The Court finds that UH's irreparable harm is greater than the hardship that the preliminary injunction would cause Defendant.

VI. THE PUBLIC INTEREST

"The public interest is always served by requiring compliance with Congressional statutes such as the Lanham Act and by enjoining the use of infringing marks."181 The Court finds that protecting UH in this case from an infringing junior competitor does not disserve the public's interest.

VII. CONCLUSION

The Court set out its conclusions of law throughout its above discussion of the case, but it repeats them here for the sake of clarity:

The Court therefore GRANTS UH's Motion for Preliminary Injunction. A hearing is set for Wednesday, October 19, 2016 at which time the parties may present argument as to: (A) a feasible timeline for Defendant to comply with the injunction and (B) whether a bond is necessary and, if so, the proper amount. The injunction will issue under separate order consistent with this opinion following the hearing.