Troubling Trademark Ruling Over Amazon’s Internal Search Results–MTM v. Amazon (Forbes Cross-Post)

Photo credit: little 3d human and a question // ShutterStock

Photo credit: little 3d human and a question // ShutterStock

When a consumer asks a retailer for a product the retailer doesn’t carry, how should the retailer respond? A recent federal appellate court opinion suggested that gave the wrong answers to consumers searching for a watch brand that it didn’t carry.

The Ruling

Multi-Time Machine makes high-end military-style watches under brand names including “MTM Special Ops.” MTM tightly controls its distribution channels. As a result, doesn’t carry MTM’s watches. When Amazon consumers searched for “MTM Special Ops Watches” in Amazon’s internal search engine, consumers were provided a list of “aesthetically similar, multi-function watches manufactured by MTM’s competitors” such as Luminox and Chase-Durer, but the search results page did not expressly say that Amazon doesn’t carry MTM watches:

amazon screen shot 3

Amazon’s disclosures on its internal search results page differ from competitors and, both of whom “clearly announce that no search results match the ‘MTM Special Ops’ query and those websites do not route the visitor to a page with both MTM’s trademark ‘MTM Specials Ops’ repeatedly at the top and competitors’ watches below. Their pages show the search query playback but then forthrightly state that no results for the ‘MTM Special Ops’ search query were found, and then list competitors’ products.”

MTM claimed that Amazon’s search results constituted trademark infringement. The district court ruled for Amazon, saying that the search results page didn’t create actionable consumer confusion. In a split vote, recently the Ninth Circuit Court of Appeals reversed, holding that Amazon’s search results presentation might constitute trademark infringement and sending the case to the jury.

The majority opinion focuses on a much-criticized trademark doctrine called initial interest confusion. The Ninth Circuit has had a dozen or so cases addressing initial interest confusion, and its handling of the doctrine has vacillated wildly. In 1999, the Ninth Circuit adopted an exceptionally (and, in my opinion, unreasonably) overbroad definition of the concept. This led to a series of tortured and inconsistent rulings until 2011, when the Ninth Circuit adopted a more constrained definition that virtually killed the doctrine.

In this case, the Ninth Circuit bypasses its 2011 definition and instead defines initial interest confusion from a 2004 ruling:

Initial interest confusion occurs not where a customer is confused about the source of a product at the time of purchase, but earlier in the shopping process, if “customer confusion . . . creates initial interest in a competitor’s product.”

What a terrible definition! It potentially implicates a wide variety of standard, well-accepted, and socially beneficial retailing practices, such as the sale of house-branded products that are similarly labeled to, and positioned immediately adjacent to, mass-marketed products. I’ve cataloged more examples of ways retailers legitimately use third-party brands for merchandising practices in this article.

Using the lousy 2004 definition, the majority says that:

A jury could infer that the labeling of the search results, and Amazon’s failure to notify customers that it does not have results that match MTM’s mark, give rise to initial interest confusion.

One judge dissented:

Because Amazon’s search results page clearly labels the name and manufacturer of each product offered for sale and even includes photographs of the items, no reasonably prudent consumer accustomed to shopping online would likely be confused as to the source of the products….The search results page makes clear to anyone who can read English that Amazon only carries the brands of watches that are clearly and explicitly listed on the web page. The search results page is unambiguous.

Given the split vote and the Ninth Circuit’s 15 year struggle trying to define initial interest confusion, this case is a good candidate for an en banc rehearing. We’ll have to see if Amazon requests that. Even without en banc relief, Amazon could still win the jury trial, especially because both opinions acknowledged MTM’s difficulty showing that any consumers had been actually confused.


* Limited to internal search results. The opinion sidesteps its broader implications for offline retailers and search engine keyword advertising, but I think the ruling only applies to internal search engines run by online retailers. The other circumstances likely involve sufficient disclosures to consumers or other signals making the relationship between the trademark owner and rivals clear.

* Easy to fix. All Amazon would have had to do to avoid liability is expressly say that it doesn’t carry MTM’s watches. Presumably Amazon ran A/B tests showing that it makes more money without such disclosure, but a complete solution to the majority’s opinion is technically and legally simple to implement.

* Initial interest confusion revitalized. I don’t believe any trademark owner has won on initial interest confusion grounds since 2011, and many trademark experts considered the doctrine dead. This opinion potentially resurrects the doctrine like a zombie. That’s an unfortunate development. The initial interest confusion doctrine is solely based on judicial intuition; no empirical research validates its existence. It’s also an overly plastic doctrine; its boundaries and definition often change from case to case. This makes it’s impossible for a defendant to rebut and hard for litigants to predict outcomes. Having a revitalized doctrine will increase defendants’ litigation costs with no commensurate social benefit.

* Trademark vs. false advertising. Although the opinions never use the term “bait-and-switch,” this opinion only makes sense as a bait-and-switch case. In other words, consumers purportedly ask Amazon if it carries MTM watches, and instead of saying no, Amazon allegedly tries to “switch” the consumers to rival offerings. Pursuant to this deconstruction, the purported consumer harm would be Amazon’s alleged misrepresentation that it carries MTM watches when it doesn’t. Although trademark law sometimes extends to bait-and-switch situations, false advertising law is more directly applicable to this kind of misrepresentation. Rather than trying to push this case into false advertising law, the majority accepted it as a trademark case and used the initial interest confusion doctrine to bridge between trademark law and false advertising law. By contorting trademark law to redress a false advertising issue, the majority’s opinion creates a doctrinal mess that probably will take years for future courts to fix.

Case citation: Multi Time Machine, Inc. v., Inc., 2015 WL 4068877 (9th Cir. July 5, 2015)