OPINION
SMITH, Circuit Judge.“A waived claim or defense is one that a party has knowingly and intelligently relinquished.” Wood v. Milyard, — U.S. -, 132 S.Ct. 1826, 1832 n.4, 182 L.Ed.2d 733 (2012). How, then, can a party waive a right “in a situation in which no right existed[?]” Ackerberg v. Johnson, 892 F.2d 1328, 1333 (8th Cir. 1989). The answer is: it cannot. Every circuit to have answered this question has held that “a litigant [need not] engage in futile gestures merely to avoid a 'claim of waiver.” Miller v. Drexel Burnham Lambert, Inc., 791 F.2d 850, 854 (11th Cir. 1986), abrogated on other grounds by Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U.S. 271, 108 S.Ct. 1133, 99 L.Ed.2d 296 (1988). We too adopt this position and therefore hold that futility can excuse the delayed invocation of the right to compel arbitration.Because we also conclude that any attempt to compel bipolar — that is, individual — arbitration in this case prior to the Supreme Court’s decision in AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 131 S.Ct. 1740, 179 L.Ed.2d 742 (2011), would have been futile, we will remand with - instructions that the District Court compel bipolar arbitration of all Plaintiffs’ arbitrable claims in accordance with its May 14, 2015, order and this opinion.
I.
Plaintiffs represent a putative class of New Jersey real estate purchasers and refinancers who were overcharged between $70 and $350 in fees stemming from the recording of their deeds and mortgage instruments. Plaintiffs allege that the settlement agents — title agents and attorneys who were, in turn, agents of the Defendants — intentionally charged Plaintiffs more than the county clerk charged for recording these documents and pocketed the difference. Plaintiffs further allege that the class claims add up to over $50 million, exclusive of treble damages and interest.
On January -22, 2009, Plaintiffs filed a complaint in the U.S. District Court for the District of New Jersey alleging both breach of contract and violation of New Jersey law. In response, Defendants sought to dismiss a number of these claims and raised several affirmative defenses. They did not, however, seek to compel arbitration based on the arbitration clauses present in their contracts with Plaintiffs.1 While no explanation for this inaction was given, we conclude infra that an attempt to compel bipolar arbitration *294would have been futile under prevailing New Jersey law. Because arbitration was not sought, the case was litigated for two and a half years with the focus primarily on class certification. In that time, both sides conducted broad discovery and contested several substantive motions on their merits. Plaintiffs have also extensively documented their efforts in this case and note that they served over 130 non-party subpoenas and spent over $50,000 on experts before Defendants sought bipolar arbitration.
However, on April 27, 2011, the Supreme Court in Concepcion held that the Federal Arbitration Act (FAA) preempted state laws that had previously prohibited a party from compelling bipolar arbitration in certain situations even when it was specifically agreed to by contract. 563 U.S. at 352, 131 S.Ct. 1740.2 Sensing an opportunity to ward off this potential class action, Defendants sent a letter to Plaintiffs on June 8, 2011, demanding enforcement of the arbitration agreements in light of this change in the law.3 Plaintiffs rejected this arbitration demand and noted that the question of arbitration “might have been an interesting issue had you thought of it two and a half years ago when the case was filed.... ” JA 816. Defendants then filed a motion in the District Court to compel bipolar arbitration on August 1, 2011. The District Court concluded that any attempt to compel bipolar arbitration prior to Concepcion would have been futile. Accordingly, the District Court granted the motion, stayed the case, and ordered bipolar arbitration.
This decision, however, led to a barrage of motions for reconsideration that, as the District Court observed, “has come to resemble a ping pong match between the parties.” JA 42. Ultimately, the District Court conducted an evidentiary hearing and found that all but two Plaintiffs had agreed to bipolar arbitration. Accordingly, over Plaintiffs’ strong protest, it again compelled bipolar arbitration of the remaining claims.
Three issues relating to this ruling were certified for interlocutory appeal pursuant to 28 U.S.C. § 1292(b):
(a) Whether the [District] Court’s decision to allow Defendants to assert the affirmative defense of arbitration due to a change in law even though a substantial period of time elapsed between [D]e-fendant[s’] answer and filing of the motion to assert the defense was in error; (b) Whether Plaintiffs’ claims against Defendants alleging a violation of the New Jersey Consumer Fraud Act, N.J.S.A. §§ 56:8-1 et seq., are barred; and (c) any other issues arising out of the District Court’s decision concerning the arbitration of the claims.
*295JA 195-96 (internal citations omitted). Because we conclude that the District Court did not err in compelling bipolar arbitration, we need not address any other issues on appeal.4
II.
We first consider whether, as an abstract matter, the futility of raising bipolar arbitration as a defense should excuse the delay in doing so. Generally, “[a] written provision in any ... contract ... to settle by arbitration a controversy thereafter arising out of such contract ... shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.” Federal Arbitration Act, 9 U.S.C. § 2. The Supreme Court has stated that this provision reflects a “ ‘liberal federal policy favoring arbitration.’ ” Concepcion, 563 U.S. at 339, 131 S.Ct. 1740 (quoting Moses H. Cone Mem’l Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 24, 103 S.Ct. 927, 74 L.Ed.2d 765 (1983)). Thus “any doubts concerning the scope of arbi-trable issues should be resolved in favor of arbitration, whether the .problem at hand is the construction of the contract language itself or any allegation of waiver, delay, or a like defense to arbitrability.” Moses H. Cone, 460 U.S. at 24-25, 103 S.Ct. 927; see also Volt Info. Scis., Inc. v. Bd. of Trs. of Leland Stanford Junior Univ., 489 U.S. 468, 476, 109 S.Ct. 1248, 103 L.Ed.2d 488 (1989) (“There is no federal policy favoring arbitration under a certain set of procedural rules; the federal policy is simply to ensure the enforceability, according to their terms, of private agreements to arbitrate.”). Given this strong preference to enforce arbitration agreements, we have concluded that a party waives the right to arbitrate “only where the demand for arbitration came long after the suit commenced and when both parties had engaged in extensive discovery.” PaineWebber Inc. v. Faragalli, 61 F.3d 1063, 1068-69 (3d Cir. 1995) (internal quotation marks and citation omitted).
For example, in Ehleiter v. Grapetree Shores, Inc., we held that a defendant seeking to compel arbitration may waive the right to do so when the plaintiff demonstrates unfair prejudice arising from the defendant’s delay in raising arbitration as an affirmative defense. 482 F.3d 207, 223 (3d Cir. 2007). To guide this prejudice inquiry, in Hoxworth v. Blinder, Robinson & Co. we identified six nonexclusive factors that courts should consider when determining if the defense of arbitration has been waived: (1) the timeliness or lack thereof of the motion to arbitrate; (2) the extent to which the party seeking arbitration has contested the merits of the opposing party’s claims; (3) whether the party seeking arbitration informed its adversary of its intent to pursue arbitration prior to seeking to enjoin the court proceedings; (4) the extent to which the party seeking arbitration engaged in non-merits motion practice; (5) the party’s acquiescence to the court’s pretrial orders; and (6) the extent to which the parties have engaged in discovery. 980 F.2d 912, 926-27 (3d Cir. 1992).
None of these cases, however, dealt with a situation in which the prejudicial delay occurred solely because an attempt to raise the defense of arbitration earlier would have been futile. Accordingly, we believe that our analysis in a futility case cannot rely solely on Hoxworth because there, waiver was premised on the fact that the right in question indisputably ex*296isted and could have been raised earlier. Wood, 132 S.Ct. at 1832 (defining waiver as the known relinquishment of a claim or defense). It makes no sense to consider, for example, whether a party “assent[ed] to the district court’s pretrial orders,” Hoxworth, 980 F.2d at 927, when, at the time the orders were issued, the party had no grounds on which it could plausibly contest them. In other words, Hoxworth was simply not a futility case. Thus, its list of factors for determining prejudice is ill suited for our analysis here.
Accordingly, we must look beyond Hox-worth to determine whether and when futility can excuse the delayed invocation of an arbitration clause as an affirmative defense. In so doing, we identify three primary reasons why we believe that adoption of the futility exception in this context is appropriate.
First, we believe that in the arbitration context, the logic of our waiver analysis— in addition to its specific factors — is undercut to the extent that the delay we are considering was the result of futility. This suggests to us that the standard waiver analysis should not apply in the futility context. Cf. Moses H. Cone, 460 U.S. at 24-25, 103 S.Ct. 927 (“[A]ny doubts concerning the scope of arbitrable issues should be resolved in favor of arbitration, whether the problem at hand is the construction of the contract language itself or any allegation of waiver, delay, or a like defense to arbitrability.”). As we noted in Hoxworth, “ ‘where a party fails to demand arbitration during pretrial proceedings, and, in the meantime, engages in pretrial activity inconsistent with an intent to arbitrate, the party later opposing a motion to compel arbitration may more easily show that its position has been compromised, i.e., prejudiced.’ ” 980 F.2d at 926 (quoting Price v. Drexel Burnham Lambert, Inc., 791 F.2d 1156, 1161 (5th Cir. 1986)). In other words, one of the primary justifications for waiver is that the party attempting to raise it as a belated defense acted inconsistently with his earlier known right to do so. However, if an earlier attempt to assert the defense of arbitration would have been futile, this failure to take a futile action is not inconsistent with that defense. See Fisher, 791 F.2d at 697 (“[T]he Fishers have failed to demonstrate that Becker acted inconsistently with a known existing right to compel arbitration.”). Thus, applicability of the standard waiver analysis seems to miss the key fact that while there may be some prejudice resulting from the delay in raising a particular defense, that prejudice is attributable to a change in the applicable law, not to any negligent action on the part of either party.
Second, we have recognized futility as an exception to both ripeness and administrative exhaustion. See Sammon v. N. J. Bd. of Med. Examiners, 66 F.3d 639, 641 (3d Cir. 1995) (engaging in a “futile gesture to establish ripeness” would be unnecessary); Berger v. Edgewater Steel Co., 911 F.2d 911, 916 (3d Cir. 1990) (“Although the exhaustion requirement is strictly enforced, courts have recognized an exception when resort to the administrative process would be futile.”); Harrow v. Prudential Ins. Co. of Am., 279 F.3d 244, 249 (3d Cir. 2002) (“A plaintiff is excused from exhausting administrative procedures under ERISA if it would be futile to do so.”). We see no reason why this logic would not extend to the case at hand. Why would we require a party to make a “futile gesture” to prevent waiver when we do not require such gestures in other scenarios?
Finally, as Plaintiffs’ admitted at oral argument, all four circuits to have considered this issue have undertaken similar analyses and come to the same conclusion. Fisher v. A.G. Becker Paribas Inc., *297791 F.2d 691, 697 (9th Cir. 1986) (“Until the Supreme Court’s decision in Byrd, the arbitration agreement in this case was unenforceable. Therefore, the Fishers have failed to demonstrate that Becker acted inconsistently with a known existing right to compel arbitration.”); Miller, 791 F.2d at 854 (“This circuit does not require a litigant to engage in futile gestures merely to avoid a claim of waiver. Thus, appellees’ failure to request arbitration prior to the Byrd decision is irrelevant to the issue of waiver.”); Peterson v. Shearson/Am. Exp., Inc., 849 F.2d 464, 466 (10th Cir. 1988) (“There was no requirement that Shearson make a futile attempt to obtain arbitration on the federal claim given the state of the law; indeed, it would be difficult to argue that such an attempt had a basis in existing law.”); Ackerberg, 892 F.2d at 1338 (“[W]e cannot find waiver, the voluntary relinquishment of a known right, in a situation in which no right existed. ... To find that the [ ] defendants waived a right they did not have ... is not only illogical, but also would encourage litigants, in order to avoid a finding of waiver, to file motions they knew to be futile.”). We therefore hold that futility can excuse the delayed invocation of the defense of arbitration.
III.
We next look at the futility exception in the context of this case. To do this, we first consider the existing bipolar and class arbitration case law and conclude that courts have typically treated each form of arbitration as substantively distinct. We therefore conclude that each can be independently waived, thereby requiring that each receive a separate futility analysis. We next ask what it means for the assertion of a right to have been futile, concluding that the appropriate test is whether assertion of that right was almost certain to fail. Finally, we apply this framework and hold that here, a motion to compel bipolar arbitration prior to Concepcion was almost certain to fail.
A.
Plaintiffs contend that it would not have been futile for Defendants to have moved to enforce the arbitration clauses prior to Concepcion. Defendants, however, assert that this is not the proper question. Instead, they suggest that the real question is whether they could have compelled bipolar arbitration prior to Concepcion.5 We agree with Defendants that the latter question is the proper inquiry, largely because profound differences distinguish class and bipolar arbitration. Cf. David S. Clancy, Matthew M.K. Stein, An Uninvited Guest: Class Arbitration and the Federal Arbitration Act’s Legislative History, 63 Bus. Law. 55, 62 (2007) (“Class arbitration is very different from the arbitration contemplated by Congress when Congress passed the FAA, and it is different in ways that plainly matter: its characteristics are the opposite of those that impressed Congress about arbitration.”). In Opalinski v. Robert Half International, Inc., we undertook a thorough review of Supreme Court and Third Circuit cases dealing with class and bipolar arbitration. 761 F.3d 326 (3d Cir. 2014). In doing so, we concluded that bipolar and class arbitration are distinct at a fundamental, substantive level. They are thus not merely different adjudicative procedures that are easily interchanged:
[W]e read the Supreme Court as characterizing the permissibility of classwide arbitration not solely as a question of procedure or contract interpretation but *298as a substantive gateway dispute qualitatively separate from deciding an individual quarrel. Traditional individual arbitration and class arbitration are so distinct that a choice between the two goes, we believe, to the very type of controversy to be resolved.
Id. at 334. Indeed, we further noted that “[t]he [Supreme] Court’s line of post-Baz-zle opinions ... indicates that, because of the fundamental differences between class-wide and bilateral arbitration, and the consequences of proceeding with one rather than the other, the availability of classwide arbitrability is a substantive gateway question ....” Id. at 335 (discussing Green Tree Fin. Corp. v. Bazzle, 539 U.S. 444, 459, 123 5.Ct. 2402, 156 L.Ed.2d 414 (2003)). Accordingly, we held that — absent clear and unmistakable evidence of agreement to the contrary — the court, not the arbitrator, must decide whether a contract permits either bipolar or class arbitration. Id. at 330, 335.
Indeed, the Supreme Court in Concepcion also highlighted three reasons why class arbitration is fundamentally different from bipolar arbitration. “First, the switch from bilateral to class arbitration sacrifices the principal advantage of arbitration — its informality — and makes the process slower, more costly, and more likely to generate procedural morass than final judgment.” 563 U.S. at 348, 131 S.Ct. 1740. “Second, class arbitration requires procedural formality. The AAA’s rules governing class arbitrations mimic the Federal Rules of Civil Procedure for class litigation.” Id. at 349, 131 S.Ct. 1740. “Third, class arbitration greatly increases risks to defendants.” Id. at 350, 131 S.Ct. 1740. Indeed, as Concepcion concluded, “[a]rbi-tration is poorly suited to the higher stakes of class litigation.” Id. These differences make it clear why parties, to put it bluntly, care so much about whether the agreement to arbitrate permits class arbitration. They also make clear why having the right to compel class arbitration is not the same as having the right to compel bipolar arbitration.6 We thus part ways with our dissenting colleague who believes that the “dispositive question here” is whether “it would have been futile in 2009 for [Defendants] to move to enforce the arbitration clauses.” By so broadening the dispositive inquiry in this case, it is no wonder she concludes that the futility exception does not apply. Had Defendants sought to enforce the arbitration clauses in 2009, they would have been, as we explain below, forced into class arbitration with near certainty.7 Even the dissent begrudgingly acknowledges that they “might not have succeeded in getting what they wanted: individual arbitration.” How, then, can we say that Defendants waived the right to individual arbitration by failing to seek to enforce an arbitration clause which almost certainly would have resulted in class arbi*299tration? We believe that the right to individual-arbitration is a distinct right separate from the right to class arbitration. We therefore hold thak — whether or not class arbitration was permissible — a court must also determine whether it would have been futile to move for bipolar arbitration under the prevailing law. We turn, then, to that question.
B.
To determine whether Defendants’ motion to compel bipolar arbitration would have been futile, we must first define futility as understood in this context. Other courts to address this issue have concluded that futility does not mean something is absolutely impossible; nor does it mean something is merely improbable. See generally Garcia-v. Wachovia Corp., 699 F.3d 1273, 1279 (11th Cir. 2012) (“The more lenient unlikely-to-succeed standard ... would only encourage litigants to delay moving to compel arbitration until they could' ascertain how the case was going in federal district court.” (internal citations and quotation marks omitted)). Instead, when asking whether arbitration would have been futile, courts ask whether “it was almost certain ... that a motion to compel arbitration would have been denied.” Id. (emphasis added); see also Miller, 791 F.2d at 854 (“[A]ny motion to compel arbitration would almost certainly have been futile.”); Peterson, 849 F.2d at 466 (“Because Shearson almost certainly could not have obtained an order for arbitration of the Rule 10b-5 claim prior to McMahon, it did not waive its right to arbitrate the claim.”).8
We agree, which requires us to ask whether a motion to compel bipolar arbitration filed by Defendants prior to Concepcion was almost certain to fail. We hold that it was. Then-existing New Jersey law prohibited courts from compelling bipolar arbitration in a certain subset of adhesive consumer contracts. Muhammad v. Cty. Bank of Rehoboth Beach, DE, 189 N.J. 1, 912 A.2d 88 (2006). Because we conclude that the contracts here fall within the subset of contracts for which compelled bipolar arbitration would have been unconscionable under Muhammad, we hold that any attempt to seek bipolar arbitration in this case prior to Concepcion was almost certain to fail.
In Muhammad, the’' New Jersey Supreme Court held that “it was unconscionable for defendants to deprive Muhammad of the mechanism of a class-wide action, whether in arbitration or in court litigation.” 189 N.J. at 22, 912 A.2d 88. This was because “[t]he public interest at stake in [the plaintiffs] ability and the ability of her fellow consumers effectively to pursue their statutory rights under this State’s consumer protection laws overrides the defendants’ right to seek enforcement of the class-arbitration bar in their agreement.” Id. Specifically, the court considered “the small amount of damages being pursued in this action involving complicated financial arrangements and multiple out-of-state entities.” Id. It then went on to note that the small value of each claim — at most $600— would make it difficult to pursue individually, even if recovery of attorney’s fees were permitted. Id. at 21, 912 A.2d 88 (“One may be hard-pressed to find an attorney willing to work on a consumer-*300fraud complaint involving complex arrangements ... when the recovery is so small.”)- Accordingly, the court concluded- — -after performing a fact intensive inquiry — that in a certain subset of adhesive contracts, a consumer cannot be compelled to submit to bipolar arbitration. This was primarily because the court concluded that individual plaintiffs were frequently discouraged from asserting meritorious claims by the relatively high costs inherent in doing so. Thus, as a policy matter, the court concluded that consumers in such situations could not be forbidden from pursuing their claims in a class setting. Id.
Three years later in Homa v. American Express Co., we also confirmed that this New Jersey policy was not preempted by the FAA. 558 F.3d 225 (3d Cir. 2009). In Homa, a putative class of consumers alleged that American Express misrepresented the terms of the rewards program for one of its credit cards. 558 F.3d at 227. Like Muhammad, the claims at issue in Homa were small claims that were not generally worth litigating on an individual basis. Id. The District Court in Homa, however, held that despite Muhammad, American Express could compel bipolar arbitration by enforcing the class arbitration waiver provision in its contracts. Id. at 230-231. We read Muhammad differently and reversed. First, we noted that the District Court placed too much emphasis on language in Muhammad explaining that class-arbitration waivers are not per se unenforceable. Id. at 230. We concluded that this focus on the lack of a per se rule inappropriately minimized the broader applicability of Muhammad and its conclusion that certain contracts barring class arbitration violated New Jersey’s public policy against exculpatory clauses. Id. Second, we compared the facts of Homa to those in Muhammad and concluded that
the contract at issue [in Homa] bears the hallmarks of a contract of adhesion — it was presented on a take-it-or-leave-it basis, ... in a standardized printed form, without opportunity for the adhering party to negotiate except perhaps on a few particulars and, as Appellant’s underlying claim implicates less than five percent of a cardholder’s overall credit card balance, predictably involves a small amount of damages.... [Thus,] the District Court should have denied the 12(b)(6) motion and concluded that, in light of Muhammad, at this stage the class-arbitration waiver at issue violates New Jersey’s fundamental public policy.
Id. at 231. Thus, “we-h[e]ld that, if the claims at issue are of such a low value as effectively to preclude relief if decided individually, then, under Muhammad, ... the class-arbitration waiver is unconscionable.”9 Here, however, because we know the value of the individual claims, we need not require the District Court to conduct further inquiry into this issue on remand.
Applying Homa and Muhammad, we must undertake a fact specific inquiry into whether the contracts in this case have the *301same characteristics as those discussed above. See Rudbart v. N. Jersey Dist. Water Supply Comm’n, 127 N.J. 344, 605 A.2d 681, 687 (1992) (“[I]n determining whether to enforce the terms of a contract of adhesion, courts have looked not only to the take-it-or-leave-it nature or the standardized form of the document but also to the subject matter of the contract, the parties’ relative bargaining positions, the degree of economic compulsion motivating the ‘adhering’ party, and the public interests affected by the contract.”). If so, we can confidently conclude that any attempt by the Defendants to compel bipolar arbitration' — and “deprive [Plaintiffs] of the mechanism of a class-wide action, whether in arbitration or in court litigation,” Muhammad, 189 N.J. at 22, 912 A.2d 88—was almost certain to fail.
In doing so, we conclude that the contracts here contain the same characteristics as those in Muhammad and Homa. Therefore, a New Jersey court confronted with this case in 2009 would almost certainly have found Muhammad controlling and would have denied Defendants’ motion to compel bipolar arbitration. Specifically, we highlight three key similarities. First, the value of the individual claims in this case are small, between $70 and $350; the claims in Muhammad were all for less than $600. This suggests that it would be, in most cases, impractical to bring an individual claim for relief. See Delta Funding Corp. v. Harris, 189 N.J. 28, 912 A.2d 104, 115 (2006) (“Harris has adequate incentive to bring her claim as an individual action. Not only are her damages substantial, but the fact that her home is at stake in the foreclosure proceeding makes it likely that she would contact an attorney. The same cannot be said of low-value claims where individuals have little, if any, incentive to seek out an attorney.”). Second, there is little doubt that these contracts are adhesive consumer contracts; they were presented on a “take-it-or-leave-it basis, commonly in a standardized printed form, without opportunity for the ‘adhering’ party to negotiate except perhaps on a few particulars.” Muhammad, 189 N.J. at 15, 912 A.2d 88.10 Finally, it is likely that here, “without the availability of a class-action mechanism, many consumer-fraud victims may never realize that they may have been wronged.” Id. at 20, 912 A.2d 88. Most consumers would lack the necessary motivation to spend time looking into the costs associated with filing records in their county. Because these similarities all evidence the same unconscionability concerns highlighted in Muhammad, we hold that any attempt by Defendants to compel bipolar arbitration prior to Concepcion would almost certainly have failed.11
*302Indeed, even the Plaintiffs at one point argued that under Muhammad, the contracts here would be unconscionable if read to prohibit class arbitration and compel bipolar arbitration. In a section of Plaintiffs’ response to Defendants’ motion to compel arbitration that is aptly titled “Under New Jersey Law Chosen by Defendants, a Requirement of Individual Arbitration of Plaintiffs’ Claims Would be Unconscionable,” the Plaintiffs explained that “[t]he factors identified by the Court in Muhammad as rendering the class action ban unconscionable are present here to the extent Defendants invoke their arbitration clause solely in order to prohibit any class action.” Pl.’s Br. in Opp. to Joint Mot. to Stay, ECF No. 221 at *39.12 While not giving this inconsistency any real weight, we simply note its existence.
IY.
We next address two minor issues. First, having concluded that the futility exception applies, we go on to excuse the pr e-Concepcion delay and analyze the post-Concepcion delay under Hoxworth. Second, we consider the scope of the arbitration clauses here and conclude that their breadth makes it clear they encompass the New Jersey Consumer Fraud Act (NJCFA) claims that are contested on appeal.
A.
Because any attempt to compel bipolar arbitration in this case prior to Concepcion was almost certain to fail, we will disregard Defendants’ pr e-Concepcion delay13 when determining whether waiver is appropriate. See Miller, 791 F.2d at 854 *303(“Subsequent to [the relevant change in the law], appellees delayed only two and a half months in making their request. Much of that delay is attributable to consolidation of the cases and appellant’s filing of an amended complaint. In light of these facts, we find no waiver of the right to arbitrate.”).14 Plaintiffs also do not claim that they suffered any additional unfair prejudice apart from the costs associated with maintaining and prosecuting this case pri- or to Concepcion. See generally Fisher, 791 F.2d at 698 (“The Arbitration Act requires district courts to compel arbitration even where the result would be the possibly inefficient maintenance of separate proceedings in different forums.”).15
Because we fully excused the pre-Coneepcion delay, we consider only whether Defendants’ approximately three-month delay between April 27, 2011 (when Concepcion was decided) and August 1, 2011 (when Defendants filed their motion to compel arbitration) was prejudicial to Plaintiffs. In doing so, we look to the Hox-worth factors outlined supra for guidance and conclude that waiver would be inappropriate for three reasons. First, Defendants notified Plaintiffs on June 8, 2011' — ■ just over a month after Concepcion was decided' — that they were demanding bipolar arbitration. This promptly put Plaintiffs on notice that a motion for arbitration was coming if the demand was rejected. Second, only three months passed between the accretion of the right and Defendant’s motion to compel arbitration. This is not an unreasonable amount of time. Third, virtually no substantive or procedural litigation occurred during this delay. Hoxworth, 980 F.2d at 926-27.16
B.
Finally, we note that the broad scope of the arbitration clauses here make *304it clear that we do not need to reach the viability of Plaintiffs’ NJCFA claims.17 The 1987 Owner’s Policy states that “arbitration shall decide any matter in dispute between you and the Company,” while the 1992 Loan Policy states that “[ajrbitrable matters include, but are not limited, to any controversy or claim between the company and the insured arising out of or relating to the policy.... ” JA 33. As we held in CardioNet, when the plain language of an arbitration clause is clear, it controls. 751 F.3d at 173. We went further, however, and also concluded that if the language of the contract is ambiguous, “the presumption of arbitrability applies” because “[w]e must resolve ‘any doubts concerning the scope of arbitrable issues ... in favor of arbitration.’ ” Id. (quoting Moses H. Cone, 460 U.S. at 24-25, 103 S.Ct. 927). Thus, we have no doubt that the NJCFA claims raised here are subject to arbitration.
V.
For the reasons previously articulated, we will remand this case to the District Court with instructions to compel bipolar arbitration of Plaintiffs’ arbitrable claims in accordance with the District Court’s May 14, 2015, order and this opinion.
. There are two arbitration clauses at issue here; some putative class members have claims under the 1987 Owner’s Policy while others have claims under the 1992 Loan Policy. For purposes of this analysis, the minor differences between the two are immaterial.
. It cannot be disputed that Concepcion marked a change in our jurisprudence. See Litman v. Cellco P’ship, 655 F.3d 225, 230 (3d Cir. 2011) ("The specific question before us remains whether the FAA preempts the New Jersey Supreme Court’s ruling in Muhammad. As noted above, we had previously held that, pursuant to Homa, it did not. We now examine that decision anew and hold that Homa has been abrogated by Concepcion and that Muhammad is preempted by the FAA.”).
. The dissent characterizes Defendants' "about-face” as a "litigation tactic [intended] to derail the court proceedings.” Such a characterization would seem to apply regardless of when Defendants moved for arbitration, as arbitration by definition derails court proceedings. In addition, deciding whether to invoke an arbitration clause is certainly a litigation ‘-'tactic,” but tactical decision making alone does not counsel against enforcement of a valid arbitration agreement. Indeed, this general disapproval of arbitration clauses has been roundly rejected by the Supreme Court. Concepcion, 563 U.S. at 339, 131 S.Ct. 1740 (explaining that the FAA reflects a "liberal federal policy favoring arbitration.”).
. The District Court had jurisdiction pursuant to 28 U.S.C. § 1332(d)(2) and this Court has jurisdiction pursuant to 28 U.S.C. § 1292(b).
. Indeed, Plaintiffs acknowledge that “Defendants have not sought arbitration other than on an individual basis” and at one point in-this litigation argued th^it the motion to compel such arbitration should be denied because it would not allow for class arbitration. ECF No. 221 at * 40.
. We recognize that both Concepcion and Opalinski were decided after this case was brought and thus were not available to Defendants when they had to decide whether to move for bipolar arbitration. That said, we believe that these cases fairly summarize the rather uncontroversial position that class arbitration differs greatly from bipolar arbitration. See Bazzle, 539 U.S. at 459, 123, S.Ct. 2402 (recognizing that parties who have agreed to bipolar arbitration cannot be forced into class arbitration because the two procedures are substantively distinct); EEOC v. Waffle House, Inc., 534 U.S. 279, 289, 122 S.Ct. 754, 151 L.Ed.2d 755 (2002) (''[N]othing in the [FAA] authorizes a court to compel arbitration of any issues, or by any parties, that are not already covered in the agreement.”). This was well understood in 2009 when the complaint was filed.
. As discussed infra, the New Jersey Supreme Court made it clear that “it [is] unconscionable for defendants to deprive [plaintiff] of the mechanism of a class-wide action, whether in arbitration or in court litigation.” Muhammad, 189 N.J. at 22, 912 A.2d 88.
. We see no conflict between the "almost certain to fail” standard and the Eighth Circuit's position that waiver is appropriate if "it should have been clear ... that the arbitration agreement was at least arguably enforceable.” Se. Stud & Components, Inc. v. Am. Eagle Design Build Studios, LLC, 588 F.3d 963, 967 (8th Cir. 2009). If an arbitration clause is not at least arguably enforcéable, it is almost certain to fail. These are, if not essentially the same standard, logically parallel.
. In a concurring opinion, Judge Weis also discussed some of the specific factors that New Jersey courts seemed to focus on when determining if a contract is unconscionable. This list, too, makes it clear that one of the key factors is the relatively small value of the individual claims: "[mjatters bearing on the Court’s appraisal included the lawsuit’s complexity, the amount of damages involved, and the availability of attorneys’ fees and statutory multipliers. The size of potential damages was considered to be an important consideration and was used to limit the holding to 'low-value' cases.” Homa, 558 F.3d at 233. He did, however, express some doubt about our ability to determine with certainty how the New Jersey Supreme Court would rule if confronted with this case, and thus explained that the District Court should perform a more in depth application of Muhammad on remand, /d.-at 233-34.
. We recognize that Plaintiffs had the option to augment a few specific terms in the contract by, for example, asking for a pre-printed Arbitration Endorsement which would have required both parties to consent to arbitration. Based on the evidence presented to the District Court, this option was rarely, if ever, requested. Nonetheless, the ability to exercise this option in a real estate closing hardly provides support for the position that this is anything but a contract of adhesion. Indeed, Muhammad recognized that the ability to negotiate "on a few particulars” does not prevent a contract from being adhesive. 189 N.J. at 15, 912 A.2d 88.
. While we readily acknowledge the concerns which the New Jersey Supreme Court has suggested necessitate a finding of procedural unconscionability, we need not opine on the public policy considerations inherent in deciding whether to enforce similar arbitration clauses because Congress has already made this difficult decision. As the Supreme Court made clear in Concepcion, "[t]he principle purpose of the FAA is to ensure that private arbitration agreements are enforced according to their terms.” 563 U.S. at 344, 131 S.Ct. 1740. And because the Supreme Court has decided that the FAA preempts contrary state law, we need not address this issue further.
. This further highlights the key error the dissent makes when suggesting that the lack of a class arbitration waiver meaningfully distinguishes this case from Muhammad. The dissent argues that in our case, it is unlikely that a New Jersey court would have held the arbitration clauses unconscionable because the clauses could have been read to permit class arbitration (due to their lack of a class-arbitration waiver). While this is true, the right question to ask when assessing futility here is not whether the arbitration clauses, as a whole, would have been unconscionable. The question is whether a motion to compel bipolar arbitration would have been futile pri- or to Concepcion. Indeed, practically speaking, the dissent's approach to futility would lead to an absurd outcome in this case. In order to prevent waiver of Defendants’ now-existing right to compel bipolar arbitration, Defendants’ lawyers would have had to seek to enforce an arbitration agreement that they knew — or should have known — would almost certainly be read to allow Plaintiffs to arbitrate as a class. Accordingly, had they pursued arbitration in 2009, they would have been exposing their clients to the very real possibility that a multi-million dollar class action case would be decided by an arbitral panel; a procedural outcome far less preferable for Defendants than a class action in federal court. See generally Concepcion, 563 U.S. at 350, 131 S.Ct. 1740 (''Arbitration is poorly suited to the higher stakes of class litigation.”); Jonathan R. Bunch, To Be Announced: Silence from the United States Supreme Court and Disagreement Among Lower Courts Suggest an Uncertain Future for Class-Wide Arbitration, 2004 J. Disp. Resol. 259, 272 (2004) (explaining that class arbitration often seems to "bring[ ] the burdens of litigation into the arbitral forum” without bringing any of the benefits along with it). It would be a cruel joke to tell Defendants that they waived their right to bipolar arbitration by not seeking to enforce an arbitration agreement that would have almost certainly led to class arbitration.
. The length of the delay itself — if excused under the futility exception — is not relevant to our inquiry. Yet even were we to consider it, we note that delays of a similar length have been excused by the Eighth and Eleventh Circuits. See Benoay v. Pindential-Bache Sec., Inc., 805 F.2d 1437, 1440 (11th Cir. 1986) (“Despite the fact that Bache and Stark's motion to compel arbitration was made two and one-half years after initiation of the civil action, we conclude that the motion was timely in light of a change in law affecting the parties’ rights.”); Nesslage v. York Sec., Inc., 823 F.2d 231, 234 (8th Cir. 1987) (excusing two year delay).
. We also believe that our decision to excuse the delay and its resulting expense to Plain- ■ tiffs here is supported by the " 'liberal federal policy favoring arbitration.' " Concepcion, 563 U.S. at 339, 131 S.Ct. 1740 (quoting Moses H. Cone, 460 U.S. at 24, 103 S.Ct. 927). We therefore leave it to courts in future cases to decide whether and how to apply futility outside of the arbitration context.
. In In re California Title Insurance Antitrust Litigation, No. CIV. 08-1341, 2011 WL 2566449 (N.D. Cal. June 27, 2011), a case practically identical to ours, Judge White concluded that the delay resulting from the belated invocation of arbitration as a defense was not unfairly prejudicial despite the fact that the case had been going on for over three years. Specifically, he noted, "[tjhere is nothing in the record to support Plaintiffs' conclu-sory contention that granting the motion to compel arbitration would unfairly prejudice Plaintiffs.” Id. at *3 (internal citations and quotation marks omitted). He then noted that the trial date was still over a year away and that only limited substantive discovery had taken place. We too are not convinced by Plaintiffs’ allegations of prejudice here and, in light of the FAA’s strong policy favoring arbitration, reject the argument that the litigation costs associated with a delay in this case can alone qualify as sufficient unfair prejudice to prevent application of the futility exception. Cf. Martin v. Yasuda, No. 15-55696, 829 F.3d 1118, 1124, 2016 WL 3924381 (9th Cir. July 21, 2016) ("A determination of whether 'the right to compel arbitration has been waived must be conducted in light of the strong federal policy favoring enforcement of arbitration agreements.’ ” (quoting Fisher, 791 F.2d at 694)).
.We also summarily reject Plaintiffs' argument that the Arbitration Endorsement was a mandatory addendum to every title insurance contract at issue here. Not only are the District Court’s findings on this issue compelling, but we are also reviewing the court’s factual findings for clear error, despite Plaintiffs attempt to dress up their argument as a matter of law by suggesting that the District Court ignored probative evidence. See N. River Ins. Co. v. CIGNA Reinsurance Co., 52 F.3d 1194, 1218 (3d Cir. 1995). The District Court here properly considered and weighed the competing evidence and found the Defendants’ position more compelling. This was not clearly erroneous.
. The scope of an arbitration clause is decided by the court absent clear and unmistakable evidence that the parties agreed to submit this issue to the arbitrator. First Options of Chicago, Inc. v. Kaplan, 514 U.S. 938, 944, 115 S.Ct. 1920, 131 L.Ed.2d 985 (1995); CardioNet, Inc. v. Cigna Health Corp., 751 F.3d 165, 171 (3d Cir. 2014). We find no language in the contracts suggesting an intent that this issue be submitted to the arbitrator.