A Comprehensive Guide to the Law of Online Contracts

//

Date of Last Update: April 25, 2022

TL;DR

This is a long article (nearly 13,000 words!) about a complicated subject. It provides substantial amounts of historical context and legal analysis. You might not care about any of that. But you can read the bolded sections and the infographics, and you’ll get gist of it.

One thing that courts often say when they’re analyzing the enforceability of online contracts is that “the fundamentals of online contract formation should not be different from ordinary contract formation.” See, e.g., Sgouros v. TransUnion Corp., 817 F.3d 1029, 1034 (7th Cir. 2016). But intuitively and analytically, the reality is that the principles of online contract formation are very different from those of the offline world.

The problems with online contracts are well known. Among the standard list of complaints: Most people don’t read them; there’s no practical way to read them even if you wanted to; most people don’t understand them; they routinely purport to make benign, commonplace conduct illegal, and counterparties almost never have the power to negotiate their terms. And finally, and perhaps most importantly, online contracts often do not require “assent” or “agreement” in the same sense as offline contracts to be enforceable.

Because software is cannibalizing so many industries, this “contract-creep” with its associated reduction of consumer legal rights, is pervading more industries. The world of meat-space has relatively few requirements with respect to entering into contracts for normal day-to-day transactions. Buying a house or even renting one requires you to sign a contract. But if you go grocery shopping, buy a pair of shoes, or get a haircut in the offline world, you almost certainly do not enter into a contract during any of those transactions. But if you buy a pair of shoes online or purchase groceries online, there’s a good chance that you may be required to enter into a legal contract to do those things. And most modern software-focused services require contractual agreements to participate. You don’t sign a contract when you hop in a taxi. But when you sign up for an Uber, they try to bind you to their terms and services agreement. And as software “eats the world” in more industries, so too do these contractual agreements that limit legal rights creep into our lives with ever-greater frequency.

Clicking “I agree” to an Apple software update doesn’t feel like signing an ink-and-paper contract. But the reality is that these online contracts are often still enforceable, just as when you sign on the dotted line. Though the circumstances that lead to their enforceability are often not what you might expect.

The purpose of this article is not to dive deeply into the ethical or “normative” questions about whether online contracts should be enforceable, but rather to provide a descriptive summary of when under United States law online contracts are enforceable, and when they are not. I’ll make a few snarky comments here and there, but for the most part, I’m going to leave the “should” part of the conversation out of this analysis. We’ll leave that to the professors and the high courts.

This article is about the current state of the law of online contracts. And that current state of the law is a framework built around a bizarre nomenclature derived from the world of CD-ROMs, DVDs, and shrink-wrap software packaging from the late 1990s. As software and the online world have evolved, the law of online contracts has sort of evolved with it, but the principles of online contract formation still hark back to the 1990s, and in many ways, are still stuck in the legal thinking from that era.

Part I of this article will take you through the history of this legal framework. Part II will take you step by step through the different legal categories and how they are applied. In Part III, we’ll consider some of the most important legal fact patterns where legal disputes frequently occur. And in Part IV, we’ll consider some of the nuances related to different jurisdictions and explain the places where the laws may be enforced differently. In Part V, we’ll have a brief discussion of “best practices” with respect to creating enforceable online agreements. Part VI will explain “best practices” from the perspective of the web user.

Part I – The Birth of Online Contracts with No Agreement

  1. Offline Origins

To understand the law of online contracts, we need to start with the case of Carnival Cruise Lines, Inc. v. Shute, 499 U.S. 585, 111 S.Ct. 1522, 113 L.Ed.2d 622 (1991). This case pre-dates the modern internet. It a case about analog contracts, not digital agreements. But if you want to understand how and why you can be bound to long, dense, unfavorable contracts that you never read, that’s where we should begin.

Eulala Shute lived in the state of Washington. She bought a ticket on a carnival cruise line. While on the cruise, she hurt herself slipping on a deck mat. This happened, according to the legal documents, while sailing off the Mexican coast.

When she got back home, she sued Carnival in Washington state. Unfortunately for her, there was a choice-of-venue provision in her ticket that required all litigation to occur in Dade County, Florida. This was stated in fine print on the back of her ticket. She sued in Washington anyway, because she never read or agreed to those terms (or so she claimed), and it was extraordinarily inconvenient to file litigation in Florida. The district court granted summary judgment in favor of Carnival based on the choice-of-venue provision.

On appeal to the Ninth Circuit, the appellate court reversed. It found that requiring Shute to litigate in Miami was unfair and inconvenient, so the choice-of-forum provision was invalid.

The Supreme Court reversed the Ninth Circuit, saying:

[R]espondents’ passage contract was purely routine and doubtless nearly identical to every commercial passage contract issued by petitioner and most other cruise lines. See, e.g., Hodes v. S.N.C. Achille Lauro ed Altri-Gestione, 858 F.2d 905, 910 (CA3 1988), cert. dism’d, 490 U.S. 1001109 S.Ct. 1633, 104 L.Ed.2d 149 (1989). In this context, it would be entirely unreasonable for us to assume that respondents—or any other cruise passenger—would negotiate with petitioner the terms of a forum-selection clause in an ordinary commercial cruise ticket. Common sense dictates that a ticket of this kind will be a form contract the terms of which are not subject to negotiation, and that an individual purchasing the ticket will not have bargaining parity with the cruise line. But by ignoring the crucial differences in the business contexts in which the respective contracts were executed, the Court of Appeals’ analysis seems to us to have distorted somewhat this Court’s holding in The Bremen.

And with this paragraph, the Supreme Court glossed over what would become a very consequential issue in the world of online contracting: Namely, that companies can bind their customers to contracts that everyone knows they never read or agreed to.

In the early-to-mid-1990s, the law of online contracts as we know it today did not yet exist. The internet was still relatively new, and online commerce was in its infancy. But the internet was exploding around the world, and courts were soon forced to grapple with the legal issues associated with it.

  1. Online Integrations

To understand how the reasoning in Carnival Cruise got applied to the digital world, we need to start with the case of ProCD, Inc. v. Zeidenberg, 86 F.3d 1447, 1448−49 (7th Cir. 1996). The time was 1996. Bill Clinton was re-elected for his second term; Michael Jordan had just come out of retirement to wreak havoc again on the NBA; and no court had yet taken seriously the idea that posting a unilateral statement of legal preferences on the internet was sufficient to create a binding contract. Before ProCD v. Zeidenberg, every court that had assessed the validity of a “shrinkwrap” contractual agreement, namely, one that states the terms of a contract on the packaging of software, had held that such a “contract” was unenforceable. But that was about to change.

Here are the facts of ProCD, Inc. v. Zeidenberg:

Defendants Matthew Zeidenberg and Silken Mountain Web Services, Inc., a one-person corporation formed by Zeidenberg, purchased copies of plaintiff’s Select Phone™ CD-ROM software program, downloaded telephone listings stored on the CD-ROM discs to Zeidenberg’s computer and made the listings available to Internet users by placing the data onto an Internet host computer. Plaintiff contends that defendants’ actions constitute copyright infringement, breach of the express terms of the parties’ software licensing agreement, a violation of Wisconsin’s Computer Crimes Act, misappropriation and unfair competition. Defendants argue that the data they downloaded from plaintiff’s Select Phone™ program were not protected by copyright, that defendants did not use Select Phone™ in a manner inconsistent with plaintiff’s copyright, that they are not bound by the software licensing agreement and that plaintiff’s state law claims are preempted by federal copyright law.

At the district court, it was determined that the defendants were entitled to summary judgment in their favor. Although the software owned by the plaintiffs was protected by copyright, that did not extend to the telephone listings included on the CD-ROM discs. Also, it was determined that the defendants only distributed “unprotected data.” The district court also found it persuasive that the license was on the inside, rather than the outside of the packaging. So consumers were required to buy the license first, and only then did they learn about the terms of the license. As such, according to the district court, they could not be bound by a contract they had not agreed to.

On appeal to the 7th Circuit, the decision was reversed and remanded. The judge who issued the opinion, Frank Easterbrook, is one of the most famous judges in the country, well-known for his “law and economics approach” to legal decision-making.

Transactions in which the exchange of money precedes the communication of detailed terms are common…. Consumer goods work the same way. Someone who wants to buy a radio set visits a store, pays, and walks out with a box. Inside the box is a leaflet containing some terms, the most important of which usually is the warranty, read for the first time in the comfort of home. By Zeidenberg’s lights, the warranty in the box is irrelevant; every consumer gets the standard warranty implied by the UCC in the event the contract is silent; yet so far as we are aware no state disregards warranties furnished with consumer products. Drugs come with a list of ingredients on the outside and an elaborate package insert on the inside. The package insert describes drug interactions, contraindications, and other vital information — but, if Zeidenberg is right, the purchaser need not read the package insert, because it is not part of the contract…. Terms of use are no less a part of “the product” than are the size of the database and the speed with which the software compiles listings. Competition among vendors, not judicial revision of a package’s contents, is how consumers are protected in a market economy. Digital Equipment Corp. v. Uniq Digital Technologies, Inc., 73 F.3d 756 (7th Cir.1996).

Judge Easterbrook’s logic is that economic theory dictates that market forces will ensure that standard form online contracts will contain terms that are not only socially efficient but also beneficial to non-drafting parties as a class compared to other possible combinations of price and terms. That the efficiency in drafting contracts this way allows producers of goods and services to provide cheaper products than they would otherwise, and that this benefits consumers. For that reason, even though everyone knows that the defendant (and that other similarly situated defendants) will probably never read the contract or agree to its terms when buying the software, the person who did not read the terms should still be bound by those terms as an enforceable contract.[1]

This was the rationale that gave birth to the modern canon and understanding of online contracting.

Judge Easterbrook justified his decision on Section 2-204 of the Uniform Commercial Code[2], which says that “a contract can be formed in any way the parties agree.” But one of the basic principles of contracts is that all parties must agree to all material terms of a contract before a contract may be formed. According to Section 2-209 of the Uniform Commercial Code, new terms can become a part of the contract, but only if they do not make material changes to the contract. Judge Easterbrook glossed over this issue, and appellate courts have not really revisited the issue with seriousness ever since.

By this logic, courts over the last 25 years have generally come to the conclusion (with some notable exceptions) that parties who agree to use any service agree to be bound by whatever legal terms that a website dictates through its terms of use, as long as they have received reasonable notice of the terms.

  1. A New Emerging (Diverging?) Standard

Contracts are a state-law issue. And online contracts, even though they exist in the friction-less, boundary-less world of the internet, are also generally governed by state-law principles.

Which is why it is perhaps odd that the law of online contracts is such an echo chamber of federal court opinions interpreting other federal court opinions, and that so much of that law is basically identical. There are relatively few state-court cases with outsized influence in the law of online contracts. Federal court judges are required, under the Erie doctrine, to apply state substantive law when exercising supplemental or diversity jurisdiction over breach of contract cases with online agreements. But the most influential online contract cases appear in federal court, and federal courts are considered more prestigious than state courts. So the practical reality is that federal courts have dictated online breach of contracts law to the states, rather than the other way around.

A few significant cases have arisen in the past few years that might change that.

Kauders v. Uber Technologies, Inc., 486 Mass. 557 (Mass. Jan. 4, 2021), a case out of the Supreme Court of Massachusetts, was yet another case to address the enforceability of Uber’s online sign-in wrap agreement. In this case, a user of the service filed suit against the company for refusing to offer rides to a user who was blind and accompanied by a guide dog. For the umpteen-millionth time, Uber filed to compel arbitration. The Supreme Court of Massachusetts denied Uber’s motion to compel arbitration, on the grounds that its online agreement was unenforceable.

In rejecting Uber’s motion to compel, the Massachusetts Supreme Court adopted a slightly different articulation of the legal standard regarding the enforceability of online contracts.

The court wrote:

We conclude that this two-prong test, focusing on whether there is reasonable notice of the terms and a reasonable manifestation of assent to those terms, is the proper framework for analyzing issues of online contract formation. Setting out these general fundamental contract principles is not, however, the difficult part of analysis. “The trick here is to know how to apply these general principles to newer forms of contracting” over the Internet….

Reasonable notice. The first prong requires that the offeree receive reasonable notice of the terms of the online agreement. Where the offeree has actual notice of the terms, this prong is satisfied without further inquiry. Miller, 448 Mass. at 680, 863 N.E.2d 537 (party bound by terms of contract regardless of whether party actually read terms). Actual notice will exist where the user has reviewed the terms. It will also generally be found where the user must somehow interact with the terms before agreeing to them.

Reasonable manifestation of assent. When considering whether the user assented to the terms of the online agreement, we consider the specific actions required to manifest assent. A user may be required to expressly and affirmatively manifest assent to an online agreement by clicking or checking a box that states that the user agrees to the terms and condition. (internal citations omitted)

As one court has observed, “[w]hile clickwrap agreements … are not necessarily required …, they are certainly the easiest method of ensuring that terms are agreed to.” Nicosia, 834 F.3d at 237-238. These are the clearest manifestations of assent.

Where no such express agreement is required by the offeror, we must turn to other less obvious manifestations of assent to the terms. This makes the task of the court more difficult. See Cullinane II, 893 F.3d at 62 (“We note at the outset that Uber chose not to use a common method of conspicuously informing users of the existence and location of terms and conditions: requiring users to click a box stating that they agree to a set of terms, often provided by hyperlink, before continuing to the next screen”). In these cases, courts must again carefully consider the totality of the circumstances, and assent may be inferred from other actions the users have taken. Where the connection between the action taken and the terms is unclear, or where the action taken does not clearly signify assent, it will be difficult for the offeror to carry its burden to show that the user assented to the terms.

Kauders at 572-574. (emphasis added)

For the last four years or so, the 1st Circuit has shown itself to be the most progressive in rejecting the enforceability of dubious online agreements. Kauders, an opinion from Massachusetts, the most populous state in the 1st Circuit, further cements that status.

While the two-prong approach is not radically different from the federal legal consensus in terms of the enforceability of online contracts, it is sufficiently different that we might expect different results in cases there compared to other parts of the country.

And the influence of that opinion is growing. In early 2022, the Maine Supreme Court adopted an identical standard to that of Kauders. Sarchi v. Uber Technologies, Inc., 2022 ME 8 (Me. Supreme Judicial Ct. Jan. 27, 2022). The question then is whether the rest of the country will follow suit.

Part II – Wrapper’s Delight

Over the years, Judge Easterbrook’s opinion in the ProCD case became incredibly influential. So much so, that courts began to think of online contracts in similar terms, even when the process related to those types of agreements bore little resemblance to the “shrinkwrap” contract that was the subject of that dispute. In the intervening years, courts have devised a nomenclature based on four general terms to categorize the large variety of online legal contracts: 1) Clickwrap 2) Browsewrap 3) Sign-in Wrap and 4) Scrollwrap. Many legal scholars hate these terms, since they are contrived and do little to resolve the underlying legal issues that face courts. But with a quarter century of history behind them, I think it’s safe to say that these terms are now entrenched in our understanding of the enforceability of online contracts.  

  1. Clickwrap

The names we use to describe online contracts began to take shape in the late 1990s. There were a few cases in the late 1990s that said that if you click “I agree” to an online agreement, that it was an enforceable contract. But the first case that I found that used the phraseology “clickwrap” to describe such a contract was Stomp, Inc. v. Neato, LLC, 61 F. Supp. 2d 1074 (C.D. Cal. 1999).

In a footnote, that case defined a clickwrap agreement as follows:

A “clickwrap agreement” allows the consumer to manifest its assent to the terms of a contract by “clicking” on an acceptance button on the website. If the consumer does not agree to the contract terms, the website will not accept the consumer’s order. Such agreements are common on websites that sell or distribute software programs that the consumer downloads from the website. The term “clickwrap agreement” is borrowed from the idea of “shrinkwrap agreements,” which are generally license agreements placed inside the cellophane “shrinkwrap” of computer software boxes that, by their terms, become effective once the “shrinkwrap” is opened.

Id. at fn. 11.

And so it began. With the definition of clickwrap agreements in that case, the court in Stomp solidified the fact that the clickwrap agreements were presumptively enforceable contracts.

By the middle of the next decade, courts across the country had adopted similar terminology and begun to solidify the reasoning that a manifestation of agreement by clicking “I agree” to a terms of use agreement or other online contract was an enforceable contract. See Davidson & Assocs. v. Jung, 422 F.3d 630, 638−39 (8th Cir. 2005); Recursion Software, Inc. v. Interactive Intelligence, Inc., 425 F. Supp. 2d 756, 781−83 (N.D. Tex. 2006); Salco Distribs., L.L.C. v. iCode, Inc., No. 8:05-CV- 642-T-27TGW, 2006 WL 449156, at *2 (M.D. Fla. Feb. 22, 2006); Mortgage Plus, Inc. v. DocMagic, Inc., No. 03-2592, 2004 WL 2331918, at *4−5 (D. Kan. Aug. 23, 2004); i-Sys., Inc. v. Softwares, Inc., No. Civ. 02-1951(JRT/FLN), 2004 WL 742082, at *6 (D. Minn. Mar. 29, 2004); Novak v. Overture Servs., Inc., 309 F. Supp. 2d 446, 451−52 (E.D.N.Y. 2004); i.LAN Sys., Inc. v. Netscout Serv. Level Corp., 183 F. Supp. 2d 328, 330−31 (D. Mass. 2002); Caspi v. Microsoft Network, L.L.C., 732 A.2d 528, 532−33 (N.J. Super. Ct. App. Div. 1999); Hotmail Corp. v. Van$ Money Pie, Inc., No. C-98-20064JW, 1998 WL 388389, at *3−9 (N.D. Cal. Apr. 16, 1998).

In 2022, the law related to clickwrap agreements is iron-clad. Absent extraordinary circumstances such as unconscionability, extreme unfairness, misdirection, or public policy issues, clickwrap agreements are enforceable contracts.

  1. Browsewrap

The clickwrap agreement discussed in the prior section is the least controversial of the types of online agreements. While there are problems inherent with negotiating contracts online with a mass audience, with clickwrap agreements, at least we are dealing with a situation where the person agreeing to the contract takes the affirmative step of clicking a button that acknowledges their agreement to those terms.

But what about the types of contracts where a person never clicks “I agree”? What about situations where a website posts a terms-of-use to a website, and the user of the website never sees it or agrees to it?

This type of contract is called a “browsewrap agreement.” It is the most common type of online agreement, and unlike the clickwrap agreement, the browsewrap agreement is presumptively unenforceable, with some noteworthy exceptions.

The first court to use the language “browsewrap” agreement in a published case was Pollstar v. Gigmania, Ltd., another California case, this time from 2000.  The court said:

Gigmania contends that the breach of contract claim fails as a matter of law because Pollstar cannot allege the required contract element of mutual consent. Viewing the web site, the court agrees with the defendant that many visitors to the site may not be aware of the license agreement. Notice of the license agreement is provided by small gray text on a gray background.

Moreover, unlike the shrinkwrap license held enforceable in ProCD v. Zeidenberg, 86 F.3d 1447 (7th Cir.1996), the license agreement at issue is a browse wrap license. A shrinkwrap license appears on the screen when the CD or diskette is inserted and does not let the consumer proceed without indicating acceptance. By contrast, a browse wrap license is part of the web site and the user assents to the contract when the user visits the web site. No reported cases have ruled on the enforceability of a browse wrap license. However, the Seventh Circuit’s opinion in ProCD provides some policy considerations that are helpful to the court.

As might be inferred from the quote above, the Pollstar court picked up right where the ProCD court left off. The plaintiffs were allowed to proceed with their breach-of-contract claim against the defendant, even though there was no evidence in the record that the defendant had ever read or consented to the terms.

The Pollstar case did not go into too much detail to give future defendants guidance on what makes a browsewrap agreement enforceable or not, but two later cases out of the 2nd Circuit over the next five years would provide that clarity.

The first of these two cases was Specht v. Netscape Commc’ns. Corp., 306 F. 3d 17 (2d Cir. 2002). In that case, the opinion was written by now-Supreme Court Justice Sonia Sotomayor. She refused to enforce an online browsewrap agreement to compel arbitration on the grounds that a “reasonably prudent Internet user in circumstances such as these would not have known or learned of the existence of the license terms before responding to defendants’ invitation to download the free software, and that defendants therefore did not provide reasonable notice of the license terms.” Id. at 20.

This was an important case, in that it was one of the first Circuit Court cases to push back on the contours of enforceability of online agreements after ProCD in a way that was more pro-consumer. It created a boundary that imposed minimum notice requirements on the enforceability of online contracts.

The next important case to address this issue was Register.com, Inc. v. Verio, Inc, 356 F.3d 393 (2d Cir. 2004). While this case wasn’t the first case to address the enforceability of an online browsewrap agreements, it was the first circuit-court case to enforce such an agreement, and it has since set the standard for such agreements ever since.

The case involved the appeal of a preliminary injunction to the 2nd Circuit. The plaintiff, Register.com, was a domain-name registrar. The defendant, Verio, Inc., accessed Register.com’s databases by web scraping, collecting the names, addresses, and telephone numbers of the domain owners, and then spammed them to sell website development services, among other things. [3]

Register.com, Inc. sued Verio, Inc. for breach of contract, trespass to chattels, and violation of the Lanham Act, and in so doing, requested injunctive relief in the form of a preliminary injunction.

The district court granted the injunction, which Verio, Inc. appealed.

On appeal, Verio argued “that it was not bound by Register’s terms because it rejected them. Even assuming Register is entitled to demand compliance with its terms in exchange for Verio’s entry into its systems to take WHOIS data, and even acknowledging that Verio was fully aware of Register’s terms, Verio contends that it still is not bound by Register’s terms because it did not agree to be bound.” Id. at 402.

The Second Circuit was not persuaded. And in explaining its reasoning, it provided the following analogy:

The situation might be compared to one in which plaintiff P maintains a roadside fruit stand displaying bins of apples. A visitor, defendant D, takes an apple and bites into it. As D turns to leave, D sees a sign, visible only as one turns to exit, which says “Apples—50 cents apiece.” D does not pay for the apple. D believes he has no obligation to pay because he had no notice when he bit into the apple that 50 cents was expected in return. D’s view is that he never agreed to pay for the apple. Thereafter, each day, several times a day, D revisits the stand, takes an apple, and eats it. D never leaves money.

P sues D in contract for the price of the apples taken. D defends on the ground that on no occasion did he see P’s price notice until after he had bitten into the apples. D may well prevail as to the first apple taken. D had no reason to understand upon taking it that P was demanding the payment. In our view, however, D cannot continue on a daily basis to take apples for free, knowing full well that P is offering them only in exchange for 50 cents in compensation, merely because the sign demanding payment is so placed that on each occasion D does not see it until he has bitten into the apple.

Verio’s circumstance is effectively the same. Each day Verio repeatedly enters Register’s computers and takes that day’s new WHOIS data…. Verio acknowledges that it continued drawing the data from Register’s computers with full knowledge that Register offered access subject to these restrictions. Verio is no more free to take Register’s data without being bound by the terms on which Register offers it, than D was free, in the example, once he became aware of the terms of P’s offer, to take P’s apples without obligation to pay the 50 cent price at which P offered them.

Returning to the apple stand, the visitor, who sees apples offered for 50 cents apiece and takes an apple, owes 50 cents, regardless whether he did or did not say, “I agree.” The choice offered in such circumstances is to take the apple on the known terms of the offer or not to take the apple.

The logic behind this case is that if an internet user has notice of a website’s terms, and the user nonetheless continues to access the website after having received notice of those terms, and the user continues to receive the benefit of the website, then the user will be bound by those terms as an enforceable contract.

But the thing about this type of case is that it is a property claim masquerading as a contract claim. The person who goes into the apple stand and eats apples without paying is stealing. A person who goes into a convenience store and takes a pack of gum without paying is shoplifting.

What the law should have done with this type of case, is develop a kind of quesi-intellectual property regime related to data. Instead, the law has developed a regime where someone can form a contract by sending them a cease-and-desist letter.

As explained by Stanford Professor Mark Lemley in his seminal article Terms of Use:

The problem is that the shift from property law to contract law takes the job of defining the Web site owner’s rights out the hands of the law and into the hands of the site owner. Property law may or may not prohibit a particular “intrusion” on a Web site, but it is the law that determines the answer to that question. The reason my “no-trespassing” sign is effective in the real world is not because there is any sort of agreement to abide by it, but because the law already protects my land against intrusion by another. If the sign read “no walking on the road outside my property,” no one would think of it as an enforceable agreement. If we make the conceptual leap to assuming that refusing to act in the way the site owner wants is also a breach of contract, it becomes the site owner rather than the law that determines what actions are forbidden. The law then enforces that private decision. One might like or dislike the vesting of such control in a site owner as a matter of policy, but doing so is an abandonment of the notion of assent.

Now, in 2022, this “conceptual leap” is likely governing law in every jurisdiction in the country.

The current regime of browsewrap contracts was laid out in the case, Nguyen v. Barnes & Noble, Inc., 763 F. 3d 1171, a 9th Circuit case from 2014.

“Unlike a clickwrap agreement, a browsewrap agreement does not require the user to manifest assent to the terms and conditions expressly … [a] party instead gives his assent simply by using the website.” Hines v. Overstock.com, Inc., 668 F.Supp.2d 362, 366-67 (E.D.N.Y.2009) (Indeed, “in a pure— form browsewrap agreement, ‘the website will contain a notice that—by merely using the services of, obtaining information from, or initiating applications within the website—the user is agreeing to and is bound by the site’s terms of service.’” Fteja v. Facebook, Inc., 841 F.Supp.2d 829, 837 (S.D.N.Y.2012) (quoting United States v. Drew, 259 F.R.D. 449, 462 n. 22 (C.D.Cal. 2009)). Thus, “by visiting the website—something that the user has already done—the user agrees to the Terms of Use not listed on the site itself but available only by clicking a hyperlink.” Id. “The defining feature of browsewrap agreements is that the user can continue to use the website or its services without visiting the page hosting the browsewrap agreement or even knowing that such a webpage exists.” Be In, Inc. v. Google Inc., No. 12-CV-03373-LHK, 2013 WL 5568706, at *6 (N.D.Cal. Oct. 9, 2013). “Because no affirmative action is required by the website user to agree to the terms of a contract other than his or her use of the website, the determination of the validity of the browsewrap contract depends on whether the user has actual or constructive knowledge of a website’s terms and conditions.” Van Tassell v. United Mktg. Grp., LLC, 795 F.Supp.2d 770, 790 (N.D.Ill.2011) (citing Sw. Airlines Co. v. BoardFirst, LLC, No. 06-CV-0891-B, 2007 WL 4823761, at *4 (N.D.Tex. Sept. 12, 2007)); see also Mark A. Lemley, Terms of Use, 91 Minn. L.Rev. 459, 477 (2006) (“Courts may be willing to overlook the utter absence of assent only when there are reasons to believe that the [website user] is aware of the [website owner’s] terms.”).

I’ve had the good fortune to discuss this issue with Professor Lemley. He has since said (in an amicus brief on a case where I was co-counsel) that:

In Boardfirst, the court accurately noted my observation that ‘[c]ourts may be willing to overlook the utter absence of assent only when there are reasons to believe that the defendant is aware of the plaintiff’s terms,’ but it apparently mistook a criticism of what had happened in a small number of cases as being advocacy for what should happen.

Since Boardfirst was decided, my words have been taken to support the opposite of what I had intended – in no small part because of Boardfirst’s misinterpretation of my point. This has proved, in my opinion and those of many other observers, to be a terrible development, taking contract law in exactly the wrong direction. Offer, acceptance, and consideration are at the heart of what we teach our students to make up a contract. But they are all absent in cases like this. Courts asked to enforce browsewrap or other contracts of adhesion should be even more zealous about ensuring that offerees’ conduct truly manifests assent to the contract terms, not less so. Adequate notice of the terms is an absolute minimum prerequisite, but it is neither a substitute for, nor sufficient proof of assent, as Boardfirst has regularly been read to suggest.

Mark Lemley, Amicus Brief, Southwest Airlines Co. v. Kiwi.com et al, 3:21-cv-00098-E.

In sum, while the clickwrap agreements are the least controversial type of online agreement, browsewrap agreements remain more controversial. But what is important to remember is that, at least for now, validity of the browsewrap contract depends on whether the user has actual or constructive knowledge of a website’s terms and conditions. If a website owner can show that a user has actual or constructive notice of a website’s terms, and that the user continues to access the site after receiving notice, the user may be held liable for any subsequent breach of contract.

  1. Scrollwrap

The first case to address the enforceability of a scrollwrap agreement and to name the term was Berkson v. Gogo LLC, 97 F. Supp. 3d 359 at 398-99 (E.D.N.Y. 2015).

Basically, a scrollwrap agreement is a subset of clickwrap agreements where a user is required scroll through a page before they are allowed to agree to an online agreement. This type of contract is seen as an extra degree of protection with respect to the clickwrap agreement. In every instance that I have seen where the enforceability of a scrollwrap agreement was addressed by a federal court, the court enforced the scrollwrap agreement. As such, the scrollwrap agreement may be viewed as the most ironclad of all online agreements, even when compared to the generic clickwrap agreement.

  1. Sign-in Wrap

The final category of wrap contracts (at least for now, until some other judge concocts another category of “wrap” agreements) is the “sign-in wrap agreement.”  This was also named by the Berkson v. Gogo LLC court, which called it “[a] questionable form of internet contracting.” Id. at 399.

If clickwraps are the least controversial and browsewraps are the most controversial, the law related to sign-in wraps is probably the most confusing for designers of websites.

These types of contracts are becoming increasingly common in recent years. The idea is that a website owner makes signing up for a service or website contingent upon agreeing to a terms-of-use. According to the Berkson court, “the website is designed so that a user is notified of the existence and applicability of the site’s ‘terms of use’ when proceeding through the website’s sign-in or login process.” Id.

One company that has had its “sign-in wrap” terms of use scrutinized by multiple federal courts is Uber. To sign up for Uber, you have to click “I agree” to use the service, but you do not have to separately agree to the terms of service. The Uber App did not require users to “click” any of these buttons or access the Terms of Service before they could complete the registration process.

The only Circuit Court to address this issue, the 1st Circuit, held that this was not an enforceable contract: “Because the Plaintiffs were not reasonably notified of the terms of the Agreement, they did not provide their unambiguous assent to those terms. We therefore find that Uber has failed to carry its burden on its motion to compel arbitration. For these reasons we reverse the district court’s grant of Uber’s motion to compel arbitration, and remand the case for further proceedings consistent with this opinion.” Cullinane v. Uber Technologies, Inc., 893 F. 3d 53 (1st Cir. 2018).

The recent Berman v. Freedom Financial case out of the Ninth Circuit is also illustrative. In this case, the defendant tried to enforce an online contract that included an arbitration agreement (we’ll discuss that more in Part III). The online agreement was a sign-in wrap agreement. The wrinkle in this case is that rather than have language that explicitly said, “By clicking CONTINUE, I agree to the Terms & Conditions which includes mandatory arbitration and Privacy Policy,” or something along those lines, the defendant simply had alarge box that said “CONTINUE” and then small print adjacent to the large “CONTINUE” button that said “I understand and agree to the Terms & Conditions which includes mandatory arbitration and [sic] Privacy Policy.” This small-print language was right next to the large CONTINUE button, but according to the Ninth Circuit, it was not clear enough to show that users manifested assent to the online contract. The motion to compel arbitration was denied.

More and more, the online sign-in wrap agreements that are being enforced are ones that look like clickwrap agreements. The most important thing to know is that if you have sign-in wrap agreements, courts might be eager to find reasons not to enforce them.

Part III: Common Fact Patterns in Online Contract Litigation

In this Part, I’ll emphasize the two most common fact patterns where conflicts with online contracts arise.

The most common fact pattern for litigation related to online contracts is when consumers try to pursue litigation or a class action against a company, and the company seeks to compel arbitration, and claims that the parties agreed to arbitrate all disputes because of an online agreement.

The second common fact pattern is when a host website tries to stop another company from accessing their online data. The host website sends a cease-and-desist letter and then pursues litigation to enforce an online contract.

  1. The Motion to Compel Arbitration

If you peruse the most important and influential legal cases in the history of online contracting (see, e.g., Specht v. Netscape Commc’ns. Corp., 306 F. 3d 17 (2d Cir. 2002); Nguyen v. Barnes & Noble, Inc., 763 F. 3d 1171 (9th Cir. 2016); Berkson v. Gogo LLC, 97 F. Supp. 3d 359 at 398-99 (E.D.N.Y. 2015); Cullinane v. Uber Technologies, Inc., 893 F. 3d 53 (1st Cir. 2018)), the majority of them involve motions to compel arbitration.

The reason is simple: Corporate defendants hate class actions. Plaintiffs’ lawyers aren’t going to file a legal complaint for someone who was swindled out of a few hundred dollars. But if a corporate defendant purportedly swindled thousands or tens of thousands or millions of people out of a few dollars, then that’s something worth litigating, if the plaintiffs can certify a class action. But if the people who were supposedly swindled agreed to arbitrate their dispute, they cannot pursue class-action litigation. This makes the enforceability of online agreements to arbitrate a billion-dollar question, in many different scenarios.

  • Policy and motivation to arbitrate

This is how one court described the current public policy in favor of arbitration:

“In enacting [Section] 2 of the [FAA], Congress declared a national policy favoring arbitration and withdrew the power of the states to require a judicial forum for the resolution of claims which the contracting parties agreed to resolve by arbitration.”); Moses H. Cone Mem’l Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 22, 24, 103 S.Ct. 927, 74 L.Ed.2d 765 (1983) (“[A]ny doubts concerning the scope of arbitrable issues should be resolved in favor of arbitration.”); see also Katherine Van Wezel Stone, Rustic Justice: Community and Coercion Under the Federal Arbitration Act, 77 N.C.L. Rev. 931, 943-56 (1999) (describing the “extraordinarily expansive” reach of the FAA).

Fashioned by corporate America into one of its most potent weapons against its customers and its own employees, this single statute possesses the power largely to eviscerate the entire panoply of civil rights and consumer protection statutes so arduously constructed by the Congress over the years. See, e.g., American Exp. Co. v. Italian Colors Rest., 570 U.S. 228, 234, 236, 133 S.Ct. 2304, 186 L.Ed.2d 417 (2013) (declining to recognize a class’s “entitlement to class proceedings for vindication of statutory rights” and concluding that “the fact that [a claim] is not worth the expense involved in proving a statutory remedy does not constitute the elimination of the right to pursue that remedy.” (emphasis in original)). At the same time, its power warps our concepts of adhesion and unconscionability, see, e.g., AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 338, 351-52, 131 S.Ct. 1740, 179 L.Ed.2d 742 (2011),2 sets at naught the court rules allowing the little guys to band together, id. at 344, 131 S.Ct. 1740 (concluding that “[r]equiring the availability of classwide arbitration interferes with fundamental attributes of arbitration and thus creates a scheme inconsistent with the FAA”), and even marginalizes our constitutional right to trial by jury, see, e.g., Gilmer v. Interstate/Johnson Lane Corp., 500 U.S. 20, 35, 111 S.Ct. 1647, 114 L.Ed.2d 26 (1991) (holding an ADEA claimant can be subjected to compulsory arbitration and rejecting the argument that the ADEA was meant to protect claimants from a waiver of a judicial forum).

Fraga v. Premium Retail Services, Inc., — F.Supp.3d —- (2022); 2022 WL 279847 (D. Mass., Jan. 31, 2022).

With all that background, it’s very much in corporate defendants’ best interests to compel arbitration, particularly with class action claims, and it’s very much in plaintiffs’ best interests to avoid it.

Corporate defendants have a significant interest in pursuing arbitration. But all websites also have a significant interest in making their sign-up processes and simple and straightforward as possible. If you make people click on too many buttons and agree to sign too many forms, you’re liable to lose customers. So websites have a long history of riding the fine line between having streamlined sign-up processes and having enforceable online contracts. And it is in that way that much of the law related to online contracts was created.

  • Early limits to enforceability

We’ve already discussed the ProCD, Inc. v. Zeidenberg case, where a contract was enforced even though the user was likely never aware of its terms. After, Zeidenberg, most courts enforced all online agreements and their arbitration clauses, regardless of their sign-in process.

And then, in 2002, Second Circuit limited the enforceability of such agreements. Specifically, the Specht cour created a new standard whereby a party looking to compel arbitration had to prove that “a reasonably prudent user of software” would have known about the agreement to arbitrate.  In Specht, the Second Circuit declined to enforce an arbitration clause to which a user purportedly agreed when he clicked on a button to download software. The terms and conditions were not visible anywhere on the screen containing the download button. According to the Specht court, “[I]n circumstances such as these, where consumers are urged to download free software at the immediate click of a button… a reference to the existence of license terms on a submerged screen is not sufficient to place consumers on inquiry or constructive notice of those terms.” Id. at 32.

  • Clickwrap nuance

Another case that illustrates some of the nuances in this area of law is Grosvenor v. Qwest Communications Intern., Inc. 2010 WL 3906253 (Sep. 30, 2010). If you’ve made it this far, you’ve probably started to notice some patterns. Clickwraps are enforceable in general. Browsewraps are enforceable only when the party looking to enforce the contract can provide that the user had notice. That simple, right?

Well, not quite. In Grosvenor, the defendant Qwest Communications was seeking to compel arbitration based on a clickwrap agreement. The court acknowledged that “Courts have found clickwrap agreements to be valid and binding.” Id. at 8. But that wasn’t the end of the analysis.

The Qwest Subscriber Agreement and the Arbitration Clause do not appear on the same scroll down box or page as the “I Accept” and the “I Do Not Accept” buttons…. As presented, the Clickwrap Agreement does not clearly incorporate the Subscriber Agreement by reference and to reach the arbitration clause requires the user to leave the installation program, log onto the Internet (if possible), navigate to the proper page, and read the Subscriber Agreement, then return to the installation program’s scroll down window to read the remaining ten pages of the High–Speed Internet Modem Installation Legal Agreement before choosing whether to agree to the terms. In addition, the arbitration issue is confused by the fact that the readily available agreements that provide a forum in the court system for resolution of conflicts springing from the scroll box contracts. This creates an ambiguity regarding recourse in the event of a dispute. These circumstances demonstrate a genuine issue of fact.

Id.

Because there was a genuine issue of material fact as to whether the plaintiffs had agreed to arbitrate their claims, the court denied the motion to compel arbitration.

  • Sign-in Wrap Confusion

If you really want to get your head spinning on this issue, start diving into the case law around the enforceability of sign-in wrap agreements. To make a long story short: determination of whether a sign-in wrap agreement (and the arbitration clause that is at the heart of that agreement) is enforceable comes down to fine gradations of text, color, location, obviousness, and other forms of context (and, if we’re being honest, the likeability of the plaintiff and the defendant, and the court and the judge deciding the case).

Take, for example, the sign-in process for popular ride-sharing service Uber. This process has evolved somewhat over time, but at least two courts have found their online agreements and arbitration clauses to be enforceable, See, e.g., Meyer v. Uber Technologies, Inc. (2d Cir. 2017) 868 F.3d 66, 76 (Meyer); Metter v. Uber Technologies, Inc. (N.D.Cal., Apr. 17, 2017, No. 16-cv-06652- RS) 2017 WL 1374579; and at least two courts have found them to be unenforceable. Cullinane v. Uber Technologies, Inc. (1st Cir. 2018) 893 F.3d 53, 63; Sarchi v. Uber Technologies, Inc., 2022 ME 8 (Me. Supreme Judicial Ct. Jan. 27, 2022).

Each of these legal disputes involved a motion to compel arbitration. Half of the time, Uber won. Half of the time, Uber lost.

If you want to spend a few hours or weeks trying to play the game of why some courts found what is basically an identical agreement to be enforceable and some found it unenforceable, be my guest. There are those who try to rationalize the differences. In my mind, it’s every bit as much about the Who (the judges and how sympathetic the parties are) and the Where (how progressive and pro-plaintiff the judges are, which is often predictable by Court location) as it is about the What (the actual drafting, display, and context of the online agreements themselves).

Suffice it to say that if you incorporate your arbitration clause into an agreement as a “sign-in wrap” contract, you run the risk of a nail-biting experience in the court room as a judge may decide whether or not to enforce it.

  1. Contract Formation by Cease-and-Desist Letter

Another common fact-pattern that generates legal disputes is what I call “contract formation by cease-and-desist letter.” This is an issue that most frequently happens in the context of web scraping. Web scraping is when someone uses automated means of collecting data from another website, often without the website’s permission. This is an interesting area of the law, because, as a rule, data is not subject to copyright or other intellectual property protection. But of course web sites often don’t like it when someone else collects data from their site without permission. There are a variety of legal arguments that host websites use to argue that this is illegal: trademark infringement, trespass to chattels, and violation of the Computer Fraud and Abuse Act.

Ultimately, the most effective argument that host websites have been able to employ in these instances is breach of contract. Many, particularly websites owned and operated by larger, public companies, have websites with terms of use that prohibit web scraping. But the vast majority of those websites have browsewrap terms of use. If you’ve made it this far, you know that “absent evidence of actual or constructive notice of the terms,” a browsewrap online agreement is not an enforceable contract.

Well, one way that companies have learned that they can demonstrate “actual notice” of terms is to send someone a cease-and-desist letter. If you send someone a cease-and-desist letter, and that cease-and-desist letter calls attention to a terms-of-use that prohibits, for example, web scraping, automatic data collection, and/or commercial use of your website, that’s probably sufficient to provide evidence of actual notice of terms. And if the person or company that receives a cease-and-desist continues to access the website in violation of the purported terms of use after receiving this cease-and-desist notice, then that conduct likely constitutes a breach of contract.

At a high level, this is the basic fact pattern for dozens of legal cases over the past 20 years, starting with Register.com v. Verio, Inc., and continuing to the present day. See, e.g. Register.com, Inc. v. Verio, Inc., 356 F. 3d 393 (2d Cir. 2004); Ticketmaster LLC v. RMG Technologies, Inc., 507 F. Supp. 2d 1096 (C.D. Cal. 2007);  BeIn, Inc. v. Google Inc., (N.D. Cal. 2013) (“Most courts upholding the enforceability of browsewrap agreements have done so in circumstances where notice to the defendant was firmly established in the factual record)”; CouponCabin LLC v. Savings.com Inc., (D. Ind. 2017); hiQ Labs, Inc. v. LinkedIn Corp. 2021 WL 1531172 (N.D. Cal., April 19, 2021).

Indeed, one company, Southwest Airlines, has litigated this issue at least ten times. See, e.g., Southwest Airlines Co. v. Farechase, Inc., 318 F. Supp. 2d 435 (N.D. Tex. 2004); Southwest Airlines Co. v. QL2 Software Inc., Case No. 3:05-CV-1558 (N.D. Tex.); Southwest Airlines Co. v. BoardFirst, L.L.C., Case No. 3:06-CV-0891, 2007 WL 4823761 (N.D. Tex. Sept. 2007); Southwest Airlines Co. v. Checkinsooner.com, LLC, Case No. 3:10-CV-01512 (N.D. Tex. 2010); Southwest Airlines Co. v. Infare Solutions A/S, Case No. 3:10-CV-01674 (N.D. Tex. 2010); Southwest Airlines Co. v. SW Software Development, LLC, Case No. 3:12-CV-00591 (N.D. Tex. 2012); and Southwest Airlines v. Roundpipe LLC, et al., 375 F. Supp. 3d 687 (N.D. Tex. 2019); , Southwest Airlines Co. v. Kiwi.com et al, 3:21-cv-00098-E (N.D. Tex. 2021). If you’ve ever wondered why you can’t find Southwest Airlines’ flights on Kayak, Google Flights, Expedia, or other online travel agents, that’s why!

At the beginning of this article, I said that the law of online contract formation is very different from the offline world. To me, this fact pattern is a perfect example of that.

But this exact fact pattern is established legal precedent in the online world. It’s totally baffling to me that courts allow contracts to be formed by cease-and-desist letters. It’s hard to imagine better evidence that parties do not agree with each other on all terms than the fact that one party is forming the contract by sending a cease-and-desist letter. But this bizarro universe is the one we’re living in, so consider yourself forewarned!

Part IV: Jurisdictional Considerations

While there are not radical differences in the law governing online contracts in different parts of the United States, there are subtle differences in how the law is applied that impact litigation in different parts of the country. The nuances and contours of the United States federal judicial system are not well understood by non-lawyers, and for good reason. It’s a complicated system, with lots of different technical components that don’t lend themselves to easy explanation.

But the key thing to keep in mind is that courts in different parts of the country take on the personalities and politics of the constituents where they reside. It is likely not surprising to many to learn that the federal courts of Florida, Alabama, Mississippi, and Texas tend to be more conservative than those in Massachusetts or California.

In the traditional world of offline contracts, if two people sign a contract in an office in Tupelo, Mississippi, it should only be logical that the law of Tupelo, Mississippi should apply to that contract, and that the dispute should be adjudicated by those courts.

But of course, the internet has changed all that. And the dawn of remote work from post-COVID has changed that even more. Now it’s not common to have a Delaware company headquartered in Silicon Valley with its workers spread out across the globe. And then consumers can be in Buenos Aires or Reykjavik or anywhere in between. The server architecture might be hosted at Amazon Web Services in Washington and the customer service personnel might be in Bangalore, and the products might be manufactured in Vietnam or El Salvador.

Against that backdrop, the question of what jurisdiction, choice of law, and venue might apply to any given online contractual dispute becomes very complicated. In many instances, the choice of law and venue often determines the outcome of the dispute.

These questions could easily become the subject of an entirely separate comprehensive guide (maybe someday!), but for now, I will just outline the major issues associated with these questions.

  1. Contract Formation

In determining the preliminary question of whether a contract was formed, the alleged contract’s choice of law provision is not controlling. See, Edminster, Hinshaw, Russ and Assoc., Inc. v. Downe Twp., 953 F.3d 348, 351 (5th Cir. 2020) (“[T]he choice-of-law provision has force only if the parties validly formed a contract.”). If the parties never entered into a contract, then there is no contractual choice of law provision to apply to the question of formation.

This creates a bit of a chicken-and-the-egg situation with respect to where legal disputes are handled. Typically, a plaintiff wants to be able to sue using its preferred forum, venue, and choice of law. But they first must establish that they had a valid contract. But the question of whether a valid contract is formed might be a fact-intensive question that can only be determined at the end of litigation. But the procedural question of where litigation will take place is something that typically occurs at the beginning of litigation.

So how do courts solve this riddle?

  1. The first-to-file “rule”

One way that this issue plays out is with the first to file rule. Often, with an online contractual dispute, either party could make a legitimate claim to having the right to argue their claims in their preferred venue of choice. For example, one company might be headquartered in Delaware, and the CEO might live in Silicon Valley, but it might have customers in every state. And it might have a contractual dispute with a Texas company, which also has business operations in Delaware, California, and Texas. In theory, all three of these states could theoretically be the proper venue for a dispute, depending on the exact facts and circumstances of the case. Now, say that the Texas company wants to litigate in Texas, because they have a history of successful litigation in Texas. And then say that the Delaware company wants nothing to do with litigation in Texas, and they know that litigation is coming, whether they like it or not, and they know that the key question of law in that dispute is going to be whether a valid contract was formed between the parties.

Well, then the Delaware company might file a declaratory judgment in the state of Delaware as a means of avoiding litigation in the great state of Texas. 

A declaratory judgment is a type of lawsuit where a person or a company files a pre-emptive lawsuit to have a court make a determination whether their conduct is legal or not. It takes advantage of what’s called the first-to-file rule, which says that if two parties both have a legitimate claim to litigate in two different venues, that the first to file will take precedence.

The “first-to-file” rule isn’t actually a rule. But courts in the federal system maintain a system of mutual accommodation where they attempt to respect each other’s boundaries. When two parties file substantially the same suit against each other in different federal jurisdictions, generally, the judge in the court that filed first gets first crack at resolving the dispute. The general idea is that, absent compelling circumstances, federal courts will defer to actions previously filed in other federal courts when the parties and issues in the two suits are basically the same. That way, federal courts don’t talk over each other in the same dispute.

  1. Personal Jurisdiction

To exercise jurisdiction over a defendant, a plaintiff must establish that the court in question has “personal jurisdiction” over the defendant in the case. In the early 1900s, this was usually a simple question, did the person or company go to the place where the lawsuit was filed or not? In 2022, this is a much more complicated question, as courts must determine which types of internet conduct are sufficient to establish personal jurisdiction.

Now courts look to active contacts with a forum, such as internet sales to the forum residents, conducting business in the forum state through numerous contacts, or entering into specific dealings with forum residents. The actual number of visitors to a defendant’s website from citizens of the forum state might also be considered in an analysis of minimum contacts.

In Zippo Manufacturing Co. v. Zippo Dot Com, Inc., 952 F. Supp. 1119 (W.D. Pa. 1997), a federal court held that “the likelihood that general jurisdiction can be constitutionally exercised is directly proportionate to the nature and quality of commercial activity that an entity conducts over the Internet. This is measured through an examination of the website’s features and intended uses. Websites designed to facilitate or conduct business transactions will often be characterized as interactive. In contrast, a passive website that simply makes the information available to the user will be less likely to have a basis for personal jurisdiction.”

  1. Subject-Matter Jurisdiction

As discussed earlier, contracts are a state-law issue. But online contract cases usually find their way into federal court and are typically decided there. But to get there, they must meet certain requirements.

One way that federal courts exercise jurisdiction is through supplemental jurisdiction. What that means is that there is a claim involved in the dispute that invokes federal law, and that an online breach of contract claim is also included in that legal complaint.

This is something that happens often in the context of web-scraping lawsuits. Specifically, those lawsuits often include federal computer fraud and abuse act (“CFAA”) claims, federal trademark claims, and federal digital millennium copyright act (“DMCA”) claims. But they also often include breach of contract claims. Since these lawsuits include federal-law claims and state-law claims, a federal court would almost always exercise supplemental jurisdiction over such claims.

Another way that federal courts exercise jurisdiction is through the complete diversity requirement. The complete diversity requirement just means that the plaintiffs and defendants are located in different states or countries. This requirement also requires that the “amount in controversy” must be larger than $75k. But since any lawsuit would require way more than that in legal fees to go all the way, that’s rarely a factor in deciding where lawsuits take place.

  1. Venue/Forum

Another wrinkle to throw into the mix is venue. Some companies have their online contracts enforced under one state’s laws with venue in another state. For example, some companies like to have their contracts governed by Delaware law (because that’s where the business is formally registered) but they like to have venue for all legal disputes where they are headquartered, sometimes in California, Texas, Colorado, or elsewhere. Because, obviously, it’s most convenient to litigate where they are physically located.

This is a tricky situation, because courts tend to be experts in their own state’s laws, but far less so with respect to other states’ laws. Typically, clerks would research other states laws and try to apply them, but this is not always as easy as it sounds.

To decide whether “venue” is proper, courts usually find that where a company has their “usual place of business” is appropriate for venue. Factors that go into this include where management or employees work, where documents and records are stored, and where the core business of the company takes place. And again, the world of remote work is making this more complicated, as well, as many more companies are fully distributed now than ever before.

Venue only really becomes an issue when someone alleges that a company does not have legitimate operations in a certain location but rather is “forum shopping,” or is trying to litigate in a certain location not because of legitimate business reasons, but because they have a belief that the outcome might be preferable in that location.

  1. Choice of Law

But the biggest question of all is what state’s law apply in any given dispute.

If there is a dispute over which state’s laws might apply in a given dispute, courts follow what is called the restatement of the law of contracts, which says that, “in the absence of a choice of law by the parties, their rights ‘are determined by the local law of the state which, with respect to that issue, has the most significant relationship to the transaction and the parties under the principles stated in [the Restatement].’”

Section 6 of the Restatement instructs courts to look to the following factors in deciding which state has the most significant relationship:

  1. the needs of the interstate and international systems,
  2. the relevant policies of the forum,
  3. the relevant policies of other interested states and the relative interests of those states in the determination of the particular issue,
  4. the protection of justified expectations,
  5. the basic policies underlying the particular field of law,
  6. certainty, predictability and uniformity of result, and
  7. ease in the determination and application of the law to be applied.

I like to joke that whenever you see a seven-part test to answer a legal question, you might as well be looking at a million-part test, because there are an infinite number of ways that the court can look at the factors and determine the result.

As a rule, a federal court exercising federal question jurisdiction applies the choice of law rules of the forum state. Again, generally speaking, if there is no conflict between the local court’s rules and the rules of other interested states, then local laws would likely be applied. If there is a conflict between two states’ laws, the choice of law analysis would continue.

Within the United States alone,it would be an uphill argument[4] to say that there is a conflict between the laws of one state and others in terms of the law of online contracts. I have seen this argument made as recently as last year that there was a conflict between Massachusetts and California laws, but a California court rejected this argument. Mitch Oberstein, et al. v. Live Nation Entertainment, Inc., et al., 2021 WL 4772885 at *5 (C.D. Cal. Sept. 20, 2021).

That said, there may be good arguments to be made that certain foreign countries’ laws are more favorable with respect to the laws of online contracts. But this is a complex jurisdictional question that goes beyond the scope of this article.

Part V: Best Practices (From the Perspective of Website Owners)

Ok, so now you know (or at least skimmed) the law of online contracts. Given all the different issues, you could be forgiven for being completely lost. But if you’re a website owner, you’re probably just looking to figure out what to do next. That’s what this section is about.

  1. Two Clicks for Near-Certain Enforceability

If you want users to be bound by your online agreements, and you want to be certain that a court will enforce those agreements, users must agree to the terms of those agreements. The clearest and most unambiguous way to do that is to have users expressly agree to the terms of your online agreement through some form of affirmative click or assent.

To quote the Kauders v. Uber Technologies Inc. court:

A user may be required to expressly and affirmatively manifest assent to an online agreement by clicking or checking a box that states that the user agrees to the terms and conditions. See, e.g., Emmannuel v. Handy Techs., Inc., 442 F. Supp. 3d 385, 389 (D. Mass. 2020) (user required to affirmatively indicate assent by clicking “Accept” button); Covino v. Spirit Airlines, Inc., 406 F. Supp. 3d 147, 152-153 (D. Mass. 2019) (enforcing agreement where user checked box acknowledging agreement with terms and conditions set forth in offeror’s contract of carriage); Wickberg v. Lyft, Inc., 356 F. Supp. 3d 179, 181 (D. Mass. 2018) (screen required user to click box indicating that he “agree[d] to Lyft’s terms of services” before he could continue with registration process). These are often referred to as “clickwrap” agreements, and they are regularly enforced. (“Clickwrap contracts require Internet users to affirmatively click ‘I agree’ when assenting to the terms and conditions on a website or making online purchases”). As one court has observed, “[w]hile clickwrap agreements … are not necessarily required …, they are certainly the easiest method of ensuring that terms are agreed to.” Nicosia, 834 F.3d at 237-238. These are the clearest manifestations of assent.

Scholar Eric Goldman calls this the “two clicks” approach. One click is used when users sign up for the service. An entirely separate click requires them to acknowledge and agree to the terms of use.

Nothing in the law is completely bulletproof. If your website swindles children and retirees with dementia for profit, its online agreements might not be enforceable, regardless of your UX. But if you want the closest thing to certainty in the law, go for the “two clicks” approach in your online agreements.

  1. Sign-in Plus Subjective Prominence

There’s a reason that Uber keeps fighting this fight in courts. The more streamlined your process is for users, the more users you get, and the more money you get. If you have too many steps involved in the process to sign up for your site, you lose some users, and losing users is an expensive proposition.

Clearly, Uber thinks that it’s worth fighting at least a half a dozen lawsuits over this issue rather than adjusting to a clickwrap agreement. I can only assume that Uber’s lawyers are well aware that the more conservative approach would be to adopt a clickwrap or “two click” approach to onboarding users.

But all these years later, it still uses a sign-in wrap agreement for its sign-up process, one that has been enforced by some courts and not enforced by others.

So if you, like Uber, think that the risk is worth it, then what you need to know is that the enforceability of your agreement is likely to hinge on a series of highly subjective factors such as text, color, location, obviousness, and other forms of context. And while courts probably won’t say this in their opinion, the general perceived wholesomeness of your business will likely be called into question as well. For example, in a recent California case, a company called JustAnswer, LLC had its arbitration agreement not enforced, and reading between the lines, the nature of the defendant’s business model was a big part of that decision.  

  1. Document Everything, in Every Browser and Means of Accessing Your Site

If, as a host website looking to enforce your online agreements, you ever do find yourself in front of a court needing to enforce your online agreements, it will be your obligation to prove to the court that your online agreements were enforceable.

Since a user approaching your site from a mobile device as opposed to desktop might have a different experience, the enforceability analysis might be different for that particular user, depending on what the sign-up process was for that online agreement. So, too, might the experience of accessing a website be different for someone accessing via Safari, Chrome, or Firefox.

My free advice: Document it all! Because your online sign-up process might change many times over the course of your business’s lifetime, and the person suing you might not have accessed your site using the same process that your business uses now, you’ll need evidence from every point in time, with every browser, and mobile and desktop application. If you’re looking to prove that your online agreement was fully enforceable in February 2020, and you can’t prove to a court what a user’s online sign-up process would have been using a Safari mobile browser in February 2020, you probably won’t be able to prove to a legal tribunal that your online agreement is enforceable.

Part VI: Best Practices (From the Perspective of Website Users)

  1. Hear No Evil, See No Evil

As a website user, you likely have no liability for breaching an online agreement unless: 1) you have affirmatively agreed to its terms or 2) you have actual or constructive notice of its terms.

With that in mind, don’t waste your time reading online agreements where there is no process that requires you to click “I agree” to the terms. It only increases your potential liability associated with violating those online agreements.

By contrast, you should definitely read any online agreement that does require you to click “I agree” to its terms. Because if you breach those types of online agreements, you will almost certainly be liable for that breach.

At the risk of being very self-serving (I mean, I am a lawyer!), you can hire an attorney to review browsewrap online agreements and assess your potential liability under those terms. Such an analysis would likely be protected by attorney-client privilege and not discoverable in a potential legal dispute. The one context where this is most valuable is in the context of web-scraping businesses, as those companies have a high risk of receiving cease-and-desist letters that generate liability for breach of browsewrap terms of use.

  1. Don’t Click if You Can’t Comply (This Mostly Applies to Businesses)

From the perspective of a user looking to avoid liability associated with breach of an online agreement, a click is the coup de grace putting an end to any hopes of trying to avoid liability (absent extraordinarily rare extenuating circumstances).

From the perspective of an online individual user, there’s not a lot you can do to negotiate an online agreement. It’s not like you can email Apple, Uber, or Spotify and politely ask them to remove the arbitration clauses that prevent you from filing a class-action lawsuit.

But from a business perspective, especially for businesses where you have substantial relationships and significant revenue tied up in recurring agreements, those contracts are worth reading in their entirety before you click. Because for those types of agreements, liability is real and there are usually alternative services available. Online agreements are often similar but not identical, and there are nuances in online terms that may make a big difference in terms of your liability, business flexibility, and legal rights.

A good rule of thumb is that if the liability or revenue associated with an online contract is more than 10x what it would cost for a lawyer to review an agreement, it’s probably time to have a lawyer review that agreement.

  1. Get Creative

Some people operate on the belief that the law of online contracts exists within some gray area of the law. That’s not really true. While there are some types of online agreements, namely sign-in wrap agreements, that require fact-intensive inquiries, the law of online contracts is well established, for the most part.

Which means that if you find yourself in a position where one of the common types of online breach might apply to you or your business, you cannot simply go with the status quo and hope things will work out for you.

There are different strategies, options, theories, and ideas that might turn a losing case into a winning one. And this might be more or less successful depending on the court of law, depending on the jurisdiction, and the open-mindedness of the judge.

This is not something to do on your own. You’ll need legal assistance to pull this off. But if you or your business is faced with liability associated with a potential online breach claim, you’ll need to draw on a full range of legal options to have any chance at succeeding.

Part VII: Conclusion

If I had to summarize the law of online contracts in three sentences, I would do it like this:

  1. Clickwrap (and scrollwrap) agreements are enforceable, absent extraordinary circumstances;
  2. Browsewrap agreements are unenforceable, absent proof of actual or constructive notice;
  3. With sign-in wrap agreements, their enforceability depends on whether the text, color, location, prominence, and other subjective criteria are sufficient to put a user on notice of their terms and to show that a user manifested agreement to its terms.

Those three sentences don’t tell the whole story. If they did, I wouldn’t have just written 12,000-plus words about this topic. But they serve as a starting point for understanding this topic.

As with everything on the internet, this issue is evolving rapidly. Everything from smart contracts, to international jurisdictional issues, to antitrust law, will impact how we view and understand these laws in the years and decades to come.

But, given the increasing volume of transactions that occur online, we can safely assume that these issues will only grow in importance in the future. And that the legal consequences associated with these issues will grow in importance as well.

This blog post is for informational purposes only. The views expressed in this post may not reflect the opinions of other attorneys at the firm or the firm’s clients.


[1] It is worth noting that the defendant in the ProCD case did have to affirmatively accept the terms of the license, but only after buying the product and downloading it onto their computer.

[2] The Uniform Commercial Code is a comprehensive set of laws governing commercial transactions within the United States.

[3] It’s worth noting here that the real harm in this case, “spam” phone calls and emails, was addressed through a statutory remedy in 2003, the CAN-SPAM Act. Since the lawsuit was initiated in 2000, however, that remedy was not available to Register.com at the time.

[4] It’s an uphill argument because most federal courts do not think there is a divergence in state laws related to online contracts. Though if you ask me, there should be.

2022 © McCarthy Law Group, LLC, all rights reserved.