Two years ago, Multi Time Machine brought a lawsuit against Amazon for trademark infringement, alleging that web pages on Amazon.com for “MTM special ops watches” keyword searches could be too confusing to consumers, since the MTM watches are not sold on the site. Now, a similar complaint was brought by Bodum versus Williams-Sonoma for French press coffeemakers. These cases illustrate significant risks for e-commerce sites.
Multi Time Machine’s complaint was based on a few different search results pages at Amazon that involved keywords associated with Multi Time Machine’s trademarks. When one searched for “mtm special ops watches” (and similar keyword searches that could be related to their marks), Amazon displayed what are essentially related search results. As mentioned before, MTM watches are not sold on Amazon — but the site associated those keyword searches with other watches that might be considered similar.
Initial Interest Confusion
Multi Time Machine claimed that this caused “Initial Interest Confusion” (IIC), which is a controversial theory of trademark law. IIC is a legal premise with a hazy definition that allows a court to find infringement has occurred when there is an initial confusion on the part of a consumer about the provenance of a product before it is ultimately bought — even if that confusion no longer exists at the point of sale. Wikipedia provides the following hypothetical example:
West Coast Video’s competitor, Blockbuster Video, puts a billboard on a stretch of highway advertising a West Coast Video at an upcoming exit. In reality, there is no West Coast Video at this exit; it is a Blockbuster Video instead. The consumer, expecting to find a West Coast Video store, sees the Blockbuster Video and decides to patronize the suitable replacement. Even though the confusion has been dispelled, Blockbuster is still misappropriating the acquired goodwill of West Coast Video’s trademark.
So, you can see that IIC is intended to address a potentially unfair attempt to profit from a competitor’s mark.
Critics of IIC believe that it may be too hazy a concept, and that the definition is too imprecise. From my layman’s understanding, the idea of a bait-and-switch scenario would seem to have some element of uncontroversial and unfair fraud attached to it, and perhaps the degree to which consumers will comprehend the mistake in brand identification by the time of making the purchase could be the deciding factor as to whether it’s legally actionable or not.
Attorneys have expressed that there may be insufficient case law precedent to make it clear how solid the IIC concept really is. There is uncertainty about how it may be interpreted, and that leaves open the door to lawsuits based upon an uncomfortably loose concept. Some of the more persuasive critics argue that the concept should be abandoned in favor of the more specific tests established for whether or not infringement has occurred, period.
A district judge initially dismissed the MTM/Amazon suit in a summary judgment, finding that Multi Time Machine failed to show that the search results actually confused consumers. Upon appeal, the Ninth Circuit court ruled that the search results could be confusing to consumers, and the suit could go forward. Then, a few months later, the appellate court reversed its decision, finding that since Amazon had clear labeling of the search results, “… no reasonably prudent consumer accustomed to shopping online would likely be confused as to the source of the products.”
Confused about Initial Interest Confusion yet?
The back-and-forth reversals garnered a lot of attention in legal circles. In the ultimate ruling, the court noted that the alleged confusion was caused not by the design of a competitor’s mark, but by the design of a web page that displayed MTM’s mark and offered the competing product for sale. The case could be decided simply by evaluating whether the web page was likely to cause a “reasonably prudent consumer” to be confused about the origin of the goods.
The court focused upon the labeling and appearance of the product listings and the surrounding context on the screen displaying the results page, and they cited that they made their determination based on two questions:
- Who is the relevant, reasonable consumer?
- What would he or she reasonably believe based on what he or she saw on the screen?
Bodum vs. Williams-Sonoma
Fast-forward to last year through this spring, and a similar case was brought against Williams-Sonoma by Bodum, a seller of French press coffeemakers.
The Bodum vs. Williams-Sonoma case varies somewhat from the MTM/Amazon one. Bodum French press coffeemakers were carried by Williams-Sonoma in the past, but the company eventually dropped Bodum as a vendor. Subsequently, Williams-Sonoma began selling their own French press coffeemakers.
For anyone who has worked in corporate America, one can imagine the sequence: after selling a brand-name product for a period of time, a retailer begins to consider that they could source the manufacture of the product themselves, cutting out their vendor in order to increase the profits. This dynamic happens all the time and, in of itself, would not necessarily be legally actionable.
However, Bodum claimed that Williams-Sonoma went on to purposefully create an infringing situation by continuing to use the Bodum name on their website, while displaying non-Bodum products. I was able to obtain a screen-grab of a “Bodum” results page from the Williams-Sonoma.com website prior to them suspending the page:
The case was settled in April, but I think the situation involved still raises a lot of questions.
Assuming the claims had been found to have merit, this case could have been an example to demonstrate Initial Interest Confusion. The difference between the MTM/Amazon case and the Bodum/Williams-Sonoma case is perhaps that one would imagine that frequent customers of Williams-Sonoma, already familiar with the Bodum products, could easily have been very confused about the origination of the more
recent Williams-Sonoma French press products that replaced the Bodum ones that were previously carried.
Further, when those consumers searched in Google or searched on the Williams-Sonoma website for “Bodum,” they were arguably wanting the specific brand’s products that they were familiar with, and if the online website serves up products that are quite similar to the Bodum ones they were previously selling, the customers could arguably be thinking that they are indeed purchasing actual Bodum brand products at the point of sale.
I would conjecture that an additional argument could have been that the Bodum and Williams-Sonoma brands might have been so closely associated with one another in the minds of Williams-Sonoma customers, and the products displayed were similar enough, that a consumer might have reasonably thought that he or she was being presented with Bodum products when shown the search results page that was in contention.
I did just one comparison, using a Bodum French press coffeemaker page from the Williams-Sonoma website I found in the Internet Archive’s Wayback Machine from when they still carried Bodum products (left), and you can see the non-Bodum French press currently found on their website (right) for comparison. The products and their styling in the catalog photos are strikingly similar:
Making a case: Factors for consideration
In the MTM/Amazon case, the court essentially found that there was not Initial Interest Confusion because of a few factors:
- The Amazon search results page states that it is showing “related” results.
- The pages of the products listed on that page do not represent that they were Multi Time Machine watches.
- Consumers should reasonably recognize that the products presented were too inexpensive to be Multi Time Machine watches.
From my layman’s perspective, Williams-Sonoma may not have the same mitigating factors that Amazon had. The williams-sonoma.com site customers had come to expect to see Bodum products on the site. Likewise, consumers searching in Google for “bodum french press” were presented with a Google search results page where the Williams-Sonoma “related search results” page for “bodum” was listed prominently in the number-one position. The products presented closely resemble the Bodum brand products, so those familiar with Bodum items would not be likely to recognize that they are different from the bona fide branded ones. The prices are quite similar.
Damages and corrective advertising
As I presented recently in a session at the SMX West conference, “What You Don’t Know In Trademark Law Can Hurt You,” the fact that the Williams-Sonoma website had optimized a page for Bodum brand keywords could have sharply increased their damages exposure in this infringement case if the court had found merit in Bodum’s claims. Each of the impressions that happened in Google when people searched for “Bodum french press” and related phrases could have eventually been determined to be an instance of infringement when the searcher saw the Bodum name associated with the Williams-Sonoma website materials. (These organic listings impressions are called “misimpressions” within the context of trademark infringement.)
Consumers do not even have to have bought the products on Williams-Sonoma for infringement to have occurred, so the search results impressions and on-site impressions numbers could easily add up to increase the theoretical damages.
Bodum had specifically cited “corrective advertising” in their complaint as a potential avenue for damages. Historically, trademark infringement cases may have incurred relatively low damages price tags when courts found in favor of plaintiffs, but more recent cases have seen monetary damage awards skyrocketing, due in part to corrective advertising as a means of reparation. The theory is straightforward: if infringement has occurred, then one might attempt to repair the confusion in the mind of consumers by performing equivalent brand exposure advertising campaigns in the same (or closely similar) medium where the infringement originally occurred.
So, when infringement occurs in search engines, such as via page listings optimized for a branded keyword, then one may factor in that equivalent-or-greater numbers of impressions need to be obtained to counter the original confusion. Thus, when one begins adding up the misimpressions that occurred on search engine results pages as well as on a website’s page itself, the numbers can add up really fast. Then, imagine multiplying those numbers by the dollar amounts necessary to obtain the same numbers of impressions at today’s market value, add on penalties, and the damages awards can become very significant.
No doubt, when Williams-Sonoma carried the Bodum products, having the “bodum” keyword results page on the website, optimized for search engines, made great sense. I don’t know if the keyword merely remained cached in the e-commerce website’s system after Bodum was dropped as a vendor, but I think it likely that it could have been retained accidentally, resulting in the keyword results page continuing to rank prominently in Google search results after they no longer carried Bodum products.
Concerningly, I’d also theorize that Williams-Sonoma’s PPC ad campaigns could have also caused a risk of increased damages exposure in this case, if they are found to be infringing due to the “Bodum” search results page. I discovered that Williams-Sonoma had had ads continuing to run for “Bodum” searches in Google, even after the lawsuit was initiated, although they apparently halted displaying their “related searches” page for organic search as the court case started heating up.
While it is not generally considered infringing to run ads online that are targeted to one’s competitors’ brand names, if they were determined to have infringed, the ad targeting might also be considered yet more evidence of an overall effort to unfairly divert sales from Bodum and might also have been considered to be part of a confusing sequence whereby consumers could be fooled into thinking the Williams-Sonoma French presses were actually Bodum’s.
I would theorize that if a company is found to be infringing, even the normally acceptable use of a competitor’s trademark could be considered in the context of the illegal mark use when determining intentionality of infringement, as well as in assessing total misimpressions.
I would theorize that such advertising could constitute yet another form of initial interest confusion, or an exacerbation of it, since a consumer might search for “Bodum,” see the Williams-Sonoma ads, click through to the website where they might also encounter the “Bodum” branded search results pages.
So, the impressions and click-throughs for Williams-Sonoma’s PPC ads targeted to Bodum keywords could have also theoretically driven up a potential damages award even higher in the case.
(Williams-Sonoma’s PPC ads targeting “Bodum” branded keywords were not mentioned in the complaint, so I doubt Bodum may have even been aware of that.)
Case settlement and results
Bodum and Williams-Sonoma agreed to settle the trademark infringement case in April, with prejudice against Williams-Sonoma.
One may reasonably speculate that the settlement terms were favorable to Bodum. I see now that when one searches for “Bodum” on williams-sonoma.com, the results page contains strong, clear language:
We do not sell Bodum branded products.
Entering more general keywords may yield results.
The page now also has text reading, “You May Also Like,” which appears above the related product search results. The page’s meta description is now blank. So, Bodum likely got what they were after.
(As an aside, Bodum probably should have hired a search engine optimization professional to help craft the settlement instructions or to check them afterwards, because there continues to be a “Bodum Thermal French Press” results page and other Bodum-branded pages on the site, and these pages are still appearing in Google search results — with no notice that Bodum-branded products are no longer carried on those pages. Or, if the settlement agreement did cover these other instances, Williams-Sonoma should have hired a competent SEO pro to help audit that the terms of the agreement were satisfied in order to ensure such pages were eliminated from search engine results.)
Interestingly, Bodum did not mention “initial interest confusion” in their lawsuit. Perhaps they intended to play their cards whichever way might be signaled by the court as the avenue most likely to succeed. But, this suit was definitely one that involved IIC in a manner parallel to the MTM/Amazon one.
It is also interesting that Williams-Sonoma decided that the risks of contesting the suit were maybe far too great to allow it to move forward, and they opted to settle. On one hand, Amazon prevailed in resisting this in the MTM case. Perhaps the court’s somewhat wishy-washy, back-and-forth decisions in MTM/Amazon made it hard to estimate the likely outcome. Or perhaps when counsel for Williams-Sonoma evaluated their case, they feared that the criteria that had cleared Amazon would not go as far in their favor. I think it’s reasonable to imagine that their customers would have been confused about the Bodum search results page.
But if Williams-Sonoma had contested the case and prevailed, they likely would ultimately have profited more. The “Bodum” keyword has considerable amounts of traffic, and having that high-ranking Bodum French press page continuing to appear in search results almost certainly translates into profitable sales.
In a great many lawsuits involving companies, the most reasonable course of action is to settle, reducing chances of undesirable outcomes for both sides.
In this case, though, we might have wished that the suit had not settled, so that there could have been more clarification of the law around initial interest confusion.
In the wake of these cases, my advice to retailers is to be extremely vigilant and timely in removing optimized pages and keywords tags from their databases when products or vendors are discontinued. You do not want to have a situation where a reasonable claim may be made that you are attempting to improperly profit from a competitor’s mark.
The other takeaway is this: label your product search results very carefully to make it clear when you do not carry a branded product — and when presenting alternative products.
Following these practices could save your company from having to defend against costly and time-consuming litigation.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.