Court Disregards Check-the-Box Agreement and Doesn’t Enforce Venue Clause — Dunstan v. comScore
[Post by Venkat Balasubramani with additional comments from Eric]
Dunstan v. comScore, Inc., 11-cv-05807 (N.D. Ill. Oct. 7, 2011)
A comScore Vice President testified that “before a user can install comScore software,” a customer must “click the box acknowledging” that the customer read and agreed to the terms. Plaintiffs, on the other hand, alleged that the forum-selection clause was not “apparent” when they downloaded the software. They also alleged that the terms of service were “obscured” during the installation process. From the court’s order, it seems like plaintiffs did not deny that they checked the box. The court resolves the apparent factual dispute as follows:
the court declines to infer that clicking a box acknowledging that a user has read an agreement indicates that the agreement was reasonably available to the user, particularly when the plaintiffs have alleged that the hyperlink to the agreement was obscured.
Whoa. Let’s take another look at this sentence. The court is saying that just because a user checked a box acknowledging the user had read the agreement, this does not mean that the court can infer that the user was able to read the agreement. (???)
comScore cited to several cases where courts enforced “click-through” agreements, including Specht v. Netscape. The court says that none of the cases involved an allegation of an obscured hyperlink. According to the court, Specht acknowledged the possibility that “a click-through agreement is not enforceable if its terms are not reasonably apparent to the user.” The court goes on to note:
it is not reasonable to expect a user casually downloading free software to search for such an agreement if it is not immediately available and obvious where to obtain it. As the Second Circuit noted, ‘when products are ‘free’ and users are invited to download them in the absence of reasonably conspicuous notice that they are about to bind themselves to contract terms, the transactional circumstances cannot be fully analogized to those in the paper world of arm’s-length bargaining.’ [U]nder the circumstances alleged here, including that the location of the license agreement was not readily apparent, the court concludes that the forum-selection clause was not reasonably communicated to the plaintiffs . . . .
This is definitely a double-take-worthy decision. The court relies on Specht v. Netscape, but Specht is a browsewrap case, where the user did not have to indicate assent to the terms before downloading the software. Given the circumstances (free download) and the fact that the terms were not in an obvious location, the court in Specht declined to enforce the terms.
There’s an easy way to solve the problem presented by Specht: have a mechanism to require the user to unequivocally indicate assent to the terms before downloading the software. Courts have upheld this type of contract formation because there is no ambiguity as to the user’s assent to the terms, and this was the type of agreement comScore had in place here. The consumer cannot say that he or she did not read the terms because prior to downloading, the user has to indicate that they read the terms. (See for example Feldman v. Google, which Eric discusses in this blog post: “Google Adwords Contract Upheld (Again)“.)
It’s tough to understate the importance of certainty in online contracting and the predictability of online agreement enforceability. They’re among the cornerstones of online commerce. Courts struggled with the enforceability of browsewrap terms, but check the box terms are widely acknowledged to be enforceable; at least there should be no bar as to mutual assent and basic contract formation. I’m not sure whether the formation process or the court went astray here (see Eric’s comments below regarding the former–he makes good points regarding implementation). If there were no issues with the UI implementation or the browser, then the court’s decision is off base.
[Interestingly, comScore did not argue that the dispute is subject to arbitration, which tends to indicate that the agreement did not have an arbitration clause.]
I have a couple theories about what went wrong here. Theory #1 is that the judge was overly willing to accept a plaintiff’s bald factual assertion that comScore didn’t adequately present the contract. (The judge says, “At this stage, however, the court must take the plaintiffs’ word for it.”). As Venkat indicates, judges have to do a little more gatekeeping than this, because plaintiffs will assert this defect in every lawsuit. If all it takes to survive a motion to dismiss is the plaintiff’s bald assertion, the contracts are nearly worthless.
Theory #2 is that comScore didn’t do its formation process properly. I think there is truth to this theory even if comScore went “by the book” and used what seemed like a mandatory non-leaky clickthrough agreement. It’s the responsibility of software vendors/website vendors to present the contract in such an unambiguous/can’t-miss-it process that NO ONE–plaintiffs’ lawyers, judges, Grandma–could possibly fail to see it. The fact that the judge gave the plaintiffs the benefit of the doubt is prima facie evidence that comScore failed to do this well enough.
The case might remind us of two key lessons for lawyers advising companies implementing user agreements:
1) I don’t care how brilliantly you draft your user agreement. It’s also your job as a lawyer to advise your clients HOW to form the contract and to ensure they follow your advice. If your brilliant contract isn’t properly formed, who cares what it says?
2) You need to look at the UI implementation across multiple browsers with a variety of settings. Even if your browser renders the agreement formation process just fine, another browser may chunk the display. This is even more crucial in the mobile environment, where UIs are even more constrained.