More Evidence That the Initial Interest Confusion Doctrine is Dying–Dwyer v. Sensocon
By Eric Goldman
Dwyer Instruments, Inc. v. Sensocon, Inc., 2012 WL 2049921 (N.D. Ind. June 5, 2012)
Earlier this year, I blogged about some research I had done suggesting the declining fortunes of the initial interest confusion doctrine. I anticipated turning the research into an article declaring the death of the initial interest confusion doctrine, and I had some pseudo-empirical evidence supporting that…and then 2011 happened, and a few too many cases embraced the initial interest confusion doctrine (such as the crappy Facebook v. Teachbook case), screwed up my empirics and effectively scuttled the article. I’ve put the article on hiatus while I let more evidence accrue that the initial interest confusion doctrine is waning and washes away 2011′s blip.
While I wait, I continue to read every case discussing the initial interest confusion doctrine. I thought this overlong opinion, resolving a mildly interesting trademark dispute in the “differential pressure gauge” industry regarding comparative advertising, provided a nice snapshot of the modern era’s judicial hostility to the plaintiffs’ overclaiming of initial interest confusion. The court says:
The Plaintiff submits, without any legal analysis, that the Defendants’ rampant use of its Mark is for the purposes of increasing the Defendants’ webpage search results for Magnehelic. This reference to Internet advertising implicates the discussion in Promatek Industries, Ltd. v. Equitrac Corp., 300 F.3d 808 (7th Cir.2002), regarding initial interest confusion. In Promatek, the defendant placed the plaintiff’s trademark in its metatag, thereby diverting web consumers to the defendant’s website. Id. at 812. The court held that this was a misappropriation of the plaintiff’s goodwill even if the consumers were no longer misled once they reached the defendant’s website. Id. (“[T]hat confusion as to the source of a product or service is eventually dispelled does not eliminate the trademark infringement which has already occurred.”) (quoting Forum Corp. of N. Am. v. Forum, Ltd., 903 F.2d 434, 442 n. 2 (7th Cir.1990)); see also Dorr–Oliver, Inc. v. Fluid–Quip, Inc., 94 F.3d 376, 382 (7th Cir.1996) (explaining that initial interest confusion is actionable under the Lanham Act and occurs when a competitor gets its foot in the door by confusing consumers with the similarity of the mark, even if the customer realizes the true source of the goods before the sale is consummated). The Promatek court explained that using another’s trademark in metatags was much like posting a sign with another’s trademark in front of a store. Id. at 813 (“Customers believing they are entering the first store rather than the second are still likely to mill around before they leave.”) (citing Brookfield Comm’ns, Inc. v. W. Coast Entm’t Corp., 174 F.3d 1036, 1064 (9th Cir.1999)); cf. Eli Lilly & Co., 233 F.3d at 465 (noting that the defendant’s references to the plaintiff’s trademark (PROZAC®) in its website source codes was evidence of wrongful intent to divert Internet users searching for information on PROZAC® to the defendant’s website). The court found that the danger of initial interest confusion that applied to terms used in metatags also applied to terms used on websites. 300 F.3d at 813.FN5
FN5. Additionally, as one district court noted, “modern search engines make little if any use of metatags.” Standard Process, Inc. v. Banks, 554 F.Supp.2d 866, 871 (E.D.Wis.2008) (quoting 4 McCarthy on Trademarks & Unfair Competition § 25:69 (4th ed.2003)). Instead of relying on keywords manipulated by webmasters, which became an increasingly poor indicator of relevancy, search engines now primarily use algorithms that rank a website by the number of other sites that link or point to it. 554 F.Supp.2d at 871 (citing 4 McCarthy on Trademarks § 25:69).
Initial customer confusion could thus exist if the Defendants’ repeated use of the Magnehelic® Mark on websites selling Sensocon products caused consumers who were looking for the Plaintiff’s product to visit the Defendants’ website under the mistaken impression that it was a site where Magnehelic® gauges were offered for sale—even if these potential buyers then saw Sensocon’s attempt to define its product as an alternative to the Plaintiff’s Magnehelic® and realized that the Plaintiff did not sponsor or endorse the Defendants’ product. However, in an infringement action, the Plaintiff bears the burden of proving likelihood of confusion, and the Plaintiff has not presented any evidence to suggest that, even if the Defendants intended to increase Sensocon’s Internet traffic through search results for Magnehelic, any such purpose was successful achieved. For example, the Plaintiff has not designated any evidence that individuals who conduct website searches containing the word Magnehelic have been or will be directed to websites offering the Defendants’ products instead of to the Plaintiff’s website or authorized dealer websites. “A genuine issue of material fact arises only if sufficient evidence favoring the nonmoving party exists to permit a jury to return a verdict for that party.” Faas v. Sears, Roebuck & Co., 532 F.3d 633, 640–41 (7th Cir.2008) (quotation marks omitted). Here, there is no evidence from which a reasonably jury could find that the Defendants’ use of the Magnehlic® Mark on the Sensocon website created customer confusion, initial or otherwise.
Notice what the court did here (overlook the stupid suggestion that comparative advertising could still create initial interest confusion). It required the plaintiff to provide some hard evidence of initial interest confusion, not just vague hypothetical assertions. Specifically, the court asks to see how any searchers actually saw any search results or other information that might have impacted their decision. The plaintiff had no such evidence, nor is any trademark owner likely to be able to generate such evidence because it’s so hard to rank in organic search for a competitor’s trademark.
This reinforces that initial interest confusion is a relic doctrine designed to redress really weak consumer search processes like guessing domain names or search engines rankings influenced by inaccurate metatags. I don’t think we needed the initial interest confusion doctrine even back in the day, but I’m confident we don’t need it in the modern search engine era. I discuss both the importance of judges requiring plaintiffs to demonstrate actual harm to support an initial interest confusion claim, and the likelihood that technological evolution would moot the problems that the initial interest confusion doctrine was designed to address, in my 2005 Deregulating Relevancy article.
Even though this opinion lays out a sensible evidentiary hurdle for trademark owners to overcome if they are going to whine about initial interest confusion, it would have been even better if the judge had taken the logical next step and just declared the initial interest confusion doctrine dead. The fact that judges can’t bring themselves to do this, and instead engage in doctrinal twists and contortions to effectuate a de facto evisceration of the doctrine, provides some interesting insights into the so-called flexibility of the common law and the dangers of common law experiments. When I write my paper about the death of the initial interest confusion doctrine a decade from now, I’ll amplify that point too.