LA STELLA v. AQUION, INC.

*NOT FOR PUBLICATION* UNITED STATES DISTRICT COURT DISTRICT OF NEW JERSEY _______________________________________ JENNIFER LA STELLA, Plaintiff, Civil Action No.:19-cv-10082(FLW) v. OPINION AQUION, INC. d/b/a RAINSOFT, and HOME DEPOT, U.S.A., INC., Defendants. WOLFSON, Chief Judge: This matter arises out of a dispute over an in-home water precipitation test. Plaintiff Jennifer La Stella (“Plaintiff” or “La Stella”) sues Defendants Aquion, Inc. d/b/a Rainsoft (“Rainsoft”) and Home Depot, U.S.A., Inc. (“Home Depot”) under state law on behalf of a putative class of consumers. La Stella alleges that Rainsoft violated the New Jersey Consumer Fraud Act (“NJCFA”) by conducting deceptive in-home water precipitation tests designed to “invariably fail” a household’s water supply and induce consumers to purchase expensive water treatment systems. Plaintiff originally filed her Complaint in state court, but Defendants removed to federal court, asserting jurisdiction under the Class Action Fairness Act (“CAFA”). Presently before the Court is Plaintiff’s renewed Motion to Remand in which she contends that Defendants do not meet CAFA’s $5,000,000 amount in controversy requirement. I. FACTUAL BACKGROUND AND PROCEDURAL HISTORY La Stella contracted with an authorized Aquion dealer on February 9, 2018, to purchase a Rainsoft water treatment system for $7,980, which she thought this was necessary after an Aquion representative performed an in-home “evaluation” using Rainsoft’s water precipitation test, which her water supply failed. See Pl. Br., at 2. La Stella alleges that the test is “calculated to induce fear and anxiety in consumers as to the safety of their drinking water” so that they feel compelled to buy high-priced water treatment systems. See Compl., ¶¶ 2, 42. She filed a Complaint on this basis in New Jersey Superior Court, Law Division, Hunterdon County, asserting three causes of action: a violation of the NJCFA, unjust enrichment, and civil conspiracy. Id. Her Complaint defines the class as any consumer who purchased a water treatment system after a precipitation test, and the class period as April 2013 to March 2019. On April 6, 2019, Defendants removed the matter to federal court under CAFA, see ECF No. 1, contending that the amount in controversy exceeds $5,000,000. In support, Defendants provided a declaration from Clifford Leegard, an Aquion employee. See ECF No. 6, at 4; ECF No. 1, ¶¶ 21-25. Leegard stated that Defendants sold at least one “water conditioning system . . . drinking water system or . . . water filtration system to at least 4,626 customers” in New Jersey since March 8, 2013. See ECF No. 6, at 4; ECF No. 1-1, at 12. Defendants then calculated purported damages per class member as $2,000, added punitive damages and attorneys’ fees, and arrived at an amount in controversy well over $5,000,000. See ECF No. 1, ¶¶ 21-25. On June 17, 2019, La Stella moved for remand, arguing that Defendants added all consumer sales in New Jersey to meet the amount in controversy without adducing any evidence to show that every sale followed an in-home water precipitation test. See ECF No. 1-1, at 12; Pl Br., at 2-3. I issued an order on January 15, 2020, granting La Stella’s motion. See ECF No. 34. Specifically, I found that: [A]s defined in Plaintiff’s Complaint, the putative class members only include those individuals who were subject to an in-home water precipitation test prior to their purchase of a Rainsoft product. Contrary to Defendants’ contentions, Plaintiff has not specifically alleged that all sales of Rainsoft product are accompanied by in- home water testing; rather, Plaintiff’s consumer fraud allegations stem from the in- home water testing and the class is limited to individuals who were subject to such testing, and the alleged misrepresentations. Accordingly, neither the total number of sales to New Jersey residents nor the total proceeds of all New Jersey sales is an appropriate metric for determining the amount of controversy in this action, because both figures include transactions which would fall outside of the proposed class definition. The information Defendants have provided at this juncture gives rise to an over-inclusive damages figure. Id. ¶ 4. I then ordered “limited jurisdictional discovery” to ascertain the appropriate amount in controversy. Id. ¶ 5. The parties were entitled to two depositions and seven production requests, which they used primarily to determine the size of the class. The class size is dispositive in this case because the parties agree on all other variables: damages are about $2,000 per consumer, each consumer can treble damages under N.J.S.A. § 56:8-19, and attorneys’ fees can be as much as thirty percent of the judgment, bringing the total alleged damages to approximately $7,800 per consumer. See Pl. Br., at 14; Opp. Br., at 12. To prove the class size, Aquion submits a declaration and related testimony from Curtis Charles Wunder, owner of Atlantic Water Products (“AWP”), a New Jersey Rainsoft dealer. The Wunder Declaration states that “at least half” of AWP’s “at least 1,200” Rainsoft sales from 2018- 2019 followed an in-home precipitation test. See ECF No. 47, Ex. 7, ¶ 8. Wunder based this conclusion, not on first-hand knowledge or data, but on his business experience, sales background, and training program for salespeople. Id. Importantly, however, Wunder testified that the decision to perform a precipitation test is “up to each salesman” and is not tracked by AWP in any manner, he was not present for virtually any of AWP’s in-home sales between 2013 and 2018, and he would not know whether a precipitation test was performed unless he was there. See ECF No. 58, Ex. 2, at 5:12-16, 22:18-25, 43:8-44:1, 46:17-20, 64:22-25, 65:1-4, 65:5-8, 65:18-21, 66:1-67:7. Without adducing further documentary evidence, Aquion uses Wunder’s “educated guess” as to the test- to-sales ratio on this motion. Aquion also submits warranty data to add to the total number of sales Wunder alleges in his Declaration, which covers only 2018-2019, whereas the class period is April 2013 to March 2019. Specifically, because every water treatment system comes with a manufacturer guarantee, Aquion extrapolates that AWP sold 2,901 Rainsoft systems during the class period.1 see ECF No. 58, Ex. 1, at 55:21-56:3 (testifying that Aquion routinely tracks warranty data as reported by dealers such as AWP); Opp. Br., at 9-10 n.5 (explaining that the warranty data contains 17,000 rows of information, so it is summarized here rather than provided in full, but that La Stella received the raw data during discovery). Taking half of these sales, Aquion concludes that the class size is 1,451 and the amount in controversy is $11,310,000. La Stella seeks to depose Wunder in her motion. Without such a deposition, she argues, Aquion could use the information Wunder proffers “as a ‘sword’ to preemptively defeat any renewed motion to remand while concurrently using it as a ‘shield’ to block . . . discovery as to [its] the reliability and competency.” Pl. Br., at 5. In any event, La Stella argues that the Wunder Declaration is unreliable because it “fail[s] to provide a sufficient factual basis” for its conclusions. Id. at 2. Finally, assuming the Wunder Declaration is reliable, La Stella argues that it cannot establish the amount in controversy on its own terms because 600 sales multiplied by $7,800 per sale equals $4,680,000. Id. Defendants respond that Wunder “substantiates the basis” for his conclusion and that the test-to-sales ratio in the Declaration, in combination with the warranty data, suffices to put more than $5,000,000 at issue. See Opp. Br., at 7-8. 1 Defendants point out that this number does not include potential class members from approximately 1,725 other sales in New Jersey during the class period. See Opp. Br., at 3. However, because they provide no support or substantiation for these sales, I decline to consider them in determining the size of the class. II. LEGAL STANDARD A defendant may remove any civil action over which a federal court would have original jurisdiction. 28 U.S.C. § 1441(a); City of Chi. v. Int’l Coll. of Surgeons, 522 U.S. 156, 163 (1997) (“The propriety of removal . . . depends on whether the case originally could have been filed in federal court.”); Penn v. Wal-Mart Stores, 116 F. Supp. 2d 557, 561 (D.N.J. 2000). Original jurisdiction exists where a plaintiff’s well-pleaded complaint presents a federal question under 28 U.S.C. § 1331 or there is diversity of citizenship under 28 U.S.C. § 1332. See Caterpillar Inc. v. Williams, 482 U.S. 386, 392 (1987). Diversity jurisdiction, in turn, exists where “the matter in controversy [for each plaintiff] exceeds the sum or value of $75,000” and all plaintiffs are “citizens of different States” than all defendants, i.e., there is complete diversity. 28 U.S.C. § 1332(a); see also Swiger v. Allegheny Energy, Inc., 540 F.3d 179, 181-82 (3d Cir. 2008). As a result of these requirements, however, “many” early class actions could not be litigated in federal courts, “even if those actions implicated matters of national importance affecting millions of parties from many different states.” Kaufman v. Allstate N.J. Ins. Co., 561 F.3d 144, 148 (3d Cir. 2009). Congress therefore enacted CAFA in 2005, authorizing “[f]ederal court consideration of interstate cases of national importance” under a less stringent jurisdictional threshold than § 1332(a). Standard Fire Ins. Co. v. Knowles, 568 U.S. 588, 595 (2013); see also Farina v. Nokia Inc., 625 F.3d 97, 110 (3d Cir. 2010). CAFA confers jurisdiction where (1) the amount in controversy exceeds $5,000,000, as aggregated across all individual claims, see 28 U.S.C. §§ 1332(d)(2)-(6); (2) the citizenship of at least one class member differs from that of any defendant, i.e., there is minimal diversity, see 28 U.S.C. § 1332(d)(2)(A); and (3) the class consists of at least 100 members. See 28 U.S.C. § 1332(d)(5)(B). Because CAFA strongly favors jurisdiction, see Anthony v. Small Tube Mfg. Corp., 535 F. Supp. 2d 506, 512 (E.D. Pa. 2007) (citing Evans v. Walter Indus., Inc., 449 F.3d 1159, 1163 (11th Cir. 2006)), “no antiremoval presumption attends [CAFA] cases.” Dart Cherokee Basin Operating Co., LLC v. Owens, 574 U.S. 81, 89 (2014); cf. Boyer v. Snap-on Tools Corp., 913 F.2d 108, 111 (3d Cir. 1990) (explaining that, in non-CAFA cases, courts must strictly construe the removal provisions and resolve all doubts in favor of remand); Steel Valley Auth. v. Union Switch & Signal Div., 809 F.2d 1006, 1010 (3d Cir. 1987) (same). Instead, CAFA “‘should be read broadly, with a strong preference that interstate class actions [ ] be heard in a federal court if properly removed by any defendant.’”2 Dart, 574 U.S. at 89 (quoting S. Rep. No. 109-14, p. 43 (2005)). Still, even in CAFA cases, the defendant bears the burden of establishing diversity. See Samuel-Bassett v. Kia Motors America, Inc., 357 F.3d 392, 396, 403 (3d Cir. 2003) (“The party asserting jurisdiction bears the burden of showing that at all stages of the litigation the case is properly before the federal court.”); Frederico v. Home Depot, 507 F.3d 188, 193 (3d Cir. 2007); Morgan v. Gay, 471 F.3d 469, 473 (3d Cir. 2006) (“Under CAFA, the party seeking to remove the case to federal court bears the burden to establish the amount in controversy is satisfied.”); Kaufman, 561 F.3d at 151; Ellis v. Bradbeck, No. 06-0750, 2006 WL 3518590, *1 (D.N.J. Dec.4, 2006) (“[T]he party asserting jurisdiction bears the heavy burden of showing that the case is properly before the federal court.”). III. DISCUSSION The parties’ dispute in this case boils down to whether Defendants have satisfied the amount in controversy requirement under CAFA. Plaintiff advances two arguments that 2 For this reason, 28 U.S.C. § 1453(c)(1) also permits a defendant to request review of an order “granting or denying a motion to remand a class action.” Compare 28 U.S.C. § 1453(c)(1), with 28 U.S.C. § 1447(d) (explaining that an order remanding a case “is not reviewable on appeal or otherwise”). Defendants have not established jurisdiction: the Wunder Declaration lacks a sufficient factual basis, and even if it is reliable evidence, Wunder’s test-to-sales ratio combined with his total sales figure falls short of $5,000,000. Defendants contend that the Wunder Declaration suffices as evidence and, based on a combination of his representations and warranty data, the amount in controversy exceeds the statutory threshold. A. What Evidence May Be Used to Establish Jurisdiction Under CAFA on Removal As a preliminary matter, Plaintiff suggests that it is improper to consider Defendants’ Notice of Removal at all because the Complaint, which typically determines jurisdiction, “does not contain any allegations establishing the [$5,000,000] amount in controversy.” Pl. Br., at 7. To determine the amount in controversy, a court must aggregate “the claims of individual class members.” 28 U.S.C. § 1332(d)(6). Specifically, the court “adds up the value of the claim of each person who falls within the definition of [the] proposed class and determine whether the resulting sum exceeds $5 million.” Knowles, 568 U.S. at 592. Usually, in doing so, the court looks to the face of the complaint. See Frederico, 507 U.S. at 197; Penn, 116 F. Supp. 2d at 561; Judon v. Travelers Prop. Cas. Co. of Am., 773 F.3d 495, 507 (3d Cir. 2014) (explaining that a court must look to the complaint and engage in a “reasonable reading of the value of the rights being litigated”) (quoting Angus v. Shiley Inc., 989 F.2d 142, 145 (3d Cir. 1993)); Dolin v. Asian Am. Accessories, Inc., 449 Fed. App’x. 216, 218 (3d Cir. 2011) (holding that a court “discern[s] the amount in controversy by consulting the face of the complaint and accepting the plaintiff's good faith allegations”); Portillo v. Nat’l Freight, Inc., 169 F. Supp. 3d 585, 595 (D.N.J. 2016); Imperial Spirits, USA, Inc. v. Trans Marine Intern. Corp., No. 98-5469, 1999 WL 172292, at *2 (D.N.J. Feb. 17, 1999) (“Under normal circumstances, a court can rely entirely upon the complaint to determine the amount in controversy.”). However, “in removal cases, determining the amount in controversy” merely “begins with a reading of the complaint.” Samuel-Bassett, 357 F.3d at 398 (emphasis added). Where, as here, the Complaint is “open-ended,” Penn, 116 F. Supp. 2d at 561, and “fails to plead a specific amount of damages,” Lamond v. Pepsico, Inc., No. 06-3043, 2007 WL 1695401, at *4 (D.N.J. June 8, 2007), a defendant’s removal notice “serves the same function as the complaint would if filed in the district court.” Frederico, 507 F.3d at 197; see also Morgan, 471 F.3d at 474. Accordingly, I may consider Defendants’ Notice of Removal to ascertain the amount in controversy. Frederico, 507 F.3d at 197-98; Imperial Spirits, 1999 WL 172292, at *2 (collecting cases); see also Erie Ins. Exch. v. Erie Indem. Co., 722 F.3d 154, 158 (3d Cir. 2013). B. What Standard Applies to Defendants’ Jurisdictional Assertion Next, there is some confusion—between the parties and in the case law—as to the standard under which I evaluate Defendants’ jurisdictional assertion. Defendants argue that they must “put forth sufficient evidence to demonstrate that it is [ ] more likely than not that CAFA jurisdiction exists” and that Plaintiff, in turn, must disprove jurisdiction beyond a legal certainty. Opp. Br., at 10. This is plainly incorrect. As an initial matter, “the defendant’s amount-in-controversy allegation should be accepted when not contested by the plaintiff.” Dart, 574 U.S. at 87-88. “[P]lausible allegations regarding the amount in controversy will [generally] suffice” in this regard. Leff v. Belfor USA Grp., Inc., No. 15-2275, 2015 WL 3486883, at *2 (D.N.J. June 2, 2015); Grace v. T.G.I. Fridays, Inc., No. 14-7233, 2015 WL 4523639, at *3 (D.N.J. July 27, 2015) (stating that, usually, “[n]o evidentiary support is required”). If, on the other hand, the plaintiff contests the defendant’s jurisdictional assertion, as here, then “removal . . . is proper . . . if the district court finds, by a preponderance of the evidence, that the amount in controversy exceeds the jurisdictional threshold.”3 28 U.S.C. § 1446(c)(2)(B); Dart, 574 U.S. at 88 (“Defendants do not need to prove to a legal certainty that the amount in controversy requirement has been met . . .”) (quoting H.R. Rep. No. 112-10, at 16 (2011)). Relatedly, Judon, 773 F.3d 495, does not determine the standard here, despite the parties’ assertions to the contrary. See, e.g., Pl. Br., at 6-7, 9 n.2; Opp. Br., at 10-11. The Third Circuit decided Judon three days before the Supreme Court decided Dart. The holding in Judon was threefold. First, where a party challenges the amount in controversy but “no evidence or findings in the trial court addressed that issue,” the party alleging jurisdiction has to justify its allegations by a preponderance of the evidence. 773 F.3d at 504 (citing McNutt v. Gen. Motors Acceptance Corp. of Ind., Inc., 298 U.S. 178, 179-80 (1936)). On the other hand, where jurisdictional facts are not contested, the party challenging jurisdiction must show that the plaintiff cannot recover more than amount in controversy to a legal certainty. Id. at 505 (citing Samuel-Bassett, 357 F.3d at 397- 98). The Third Circuit also discussed the so-called “third test” under CAFA, set forth in Morgan, 471 F.3d at 474, and cited with approval in Frederico, 507 F.3d at 195-97: where a plaintiff expressly limits the amount in controversy, a removing defendant must prove “to a legal certainty” that the amount in controversy exceeds the statutory minimum, because the plaintiff is the master of her own complaint. Judon, 773, F.3d at 504 n.8. 3 This standard derives from “removals sought on the basis of the jurisdiction conferred by section 1332(a),” but the Supreme Court has “assume[d] without deciding” that it also applies “to cases removed under § 1332(d)(2),” such as the present one, Dart, 574 U.S. at 88 n.1 (internal quotations omitted), because “there is no logical reason why [a court] should demand more from a CAFA defendant than other parties invoking federal jurisdiction.” Frederick v. Hartford Underwriters Ins. Co., 683 F.3d 1242, 1247 (10th Cir. 2012) (internal quotations omitted) (quoted in Dart); Leff v. Belfor USA Grp., Inc., No. 15-2275, 2015 WL 3486883, at *2 (D.N.J. June 2, 2015) (quoting Dart for the proposition that the defendant in a CAFA case must prove its jurisdictional assertion by a preponderance of the evidence, and explaining that “evidence establishing the amount in controversy is required” only when federal jurisdiction is disputed). Dart clearly “calls into question the soundness of the reasoning for applying the legal certainty test rather than the preponderance of the evidence test,” Grace, 2015 WL 4523639, at *5, as well as Judon’s standards for post-removal challenges to the amount in controversy. Rather, the preponderance of the evidence standard applies whenever a plaintiff disputes jurisdiction. Dart, 547 U.S. at 89. This, together with the fact that a plaintiff’s complaint cannot bind the court as to the amount in controversy—for example by remaining silent—so as to prevent removal, see Knowles, 568 U.S. at 592-93, renders the legal certainty test “unnecessary under the present circumstances.” Grace, 2015 WL 4523639, at *6. In sum, Plaintiff challenges Defendants’ jurisdictional assertion, so I cannot rely on plausible allegations in Defendants’ Notice of Removal as to the amount in controversy. See Quick v. Kramer, No. 15-5845, 2015 WL 7737347, at *2 (D.N.J. Nov. 30, 2015) (“Plaintiffs’ motion to remand contests Defendant’s assertion that the jurisdictional amount has been met. If jurisdiction is challenged, Defendant must establish that the jurisdictional threshold has been satisfied by a preponderance of the evidence; a plausible, good faith allegation of the amount in controversy is not sufficient.”) (internal citation omitted). Instead, I must apply the preponderance of the evidence standard to determine whether Defendants’ evidence, e.g., the Wunder Declaration, is sufficient to meet the jurisdictional threshold.4 See Dart, 574 U.S. at 89; Grace, 2015 WL 4523639, at *6 (“[T]he Court will proceed to analyze the allegations in the Complaint and TGIF’s submitted 4 I note that Plaintiff need not submit her own evidence in order to contest Defendants’ jurisdictional assertion under Dart. See, e.g., Mejia v. DHL Express (USA), Inc., No. 15-890, 2015 WL 2452755, at *2 (C.D. Cal. May 21, 2015) (“Defendant contends that the Supreme Court's reference to ‘both sides’ [in Dart] submitting proof means that Plaintiff should have produced affirmative evidence showing that the amount in controversy is not satisfied with his Motion. We recently rejected a similar argument . . . . While Plaintiff may rebut Defendant's evidence with his own evidence, he need not do so in order to prevail in his Motion.”); see also Rosenblatt, 2016 WL 3546579, at *4 n.1 (quoting Mejia). proofs in deciding whether the amount in controversy exceeds $5,000,000 by a preponderance of the evidence.”). Finally, Defendants misconstrue the burden of proof, see Opp. Br., at 10, which remains at all times with them, as is true of any removal case, whether under CAFA or § 1332(a). See Frederico, 507 F.3d at 193; Morgan, 471 F.3d at 473 (“Under CAFA, the party seeking to remove the case to federal court bears the burden to establish the amount in controversy is satisfied.”). In other words, “the Supreme Court’s Dart [] decision seemingly leaves undisturbed the canon that— even in CAFA removal cases—the party asserting federal jurisdiction . . . bears the burden” of proving it exists. Rosenblatt v. Nuplexa Group, Inc., No. 16-1064, 2016 WL 3546579, at *4 (D.N.J. June 29, 2016) (internal quotation omitted); Leff, 2015 WL 3486883, at *2 (“As the removing party, [Defendant] bears the burden of demonstrating that this action meets the minimum amount to establish that this Court has subject matter jurisdiction.”); Grace, 2015 WL 4523639, at *2 (same); Edginton-Steward v. Wal-Mart Corp., No. 11-4686, 2011 WL 3625599, at *2 (D.N.J. Aug. 17, 2011) (“It is well-settled that a removing defendant carries the burden of establishing the requisite amount in controversy.”). Defendants therefore have the burden of proving by a preponderance of evidence that the amount in controversy exceeds $5,000,000. C. Whether the Wunder Declaration Is Reliable Evidence The crux of the parties’ dispute as to the amount in controversy is whether the Wunder Declaration is reliance evidence, i.e., whether it provides a sufficient factual basis to support Defendants’ claim that damages exceed $5,000,000. Plaintiff argues that the Wunder Declaration fails because it is speculative and conclusory. See Pl. Br., at 9-10. Defendants argue that it is reliable, partly because Plaintiff has not disproved jurisdiction to a legal certainty, which is incorrect as a matter of law, see supra, but also because “[n]o reasonable interpretation of the evidence can support a finding that only 641 or fewer putative class members exist.” Opp. Br., at 11. Ascertaining the amount in controversy is “relatively straightforward.” Portillo, 169 F. Supp. 3d at 597. Defendants must multiply the number of sales covered by the class definition and occurring during the class period, i.e., all sales following an in-home precipitation test between April 2013 and March 2019, by the amount of damages per sale, treble that product under the NJCFA, then add attorneys’ fees. The parties agree on this formula for calculating the amount in controversy, see Pl. Br., at 12-14; Opp. Br., at 9 n.3, which results in damages of about $7,800 per consumer, but disagree on the number of sales included in the class definition. The number of sales, or the class size, is therefore “the most important variable[] involved in the necessary calculation.” Toribio v. ITT Aerospace Controls LLC, No. 19-5430, 2019 WL 4254935, at *3 (C.D. Cal. Sept. 5, 2019). Defendants originally asserted 4,626 class members and an amount in controversy exceeding $74,000,000.5 See ECF No. 1. I found that this was not “an appropriate metric for determining the amount in controversy,” since Defendants “repeatedly asserted that no documents exist that show whether a precipitation test preceded the sale of any Rainsoft product to any New Jersey consumer.” See ECF No. 34. To remedy this deficiency, Defendants produced the Wunder Declaration, which purports to establish that a precipitation test preceded fifty percent of “at least 1,200” sales from 2018-2019. Because Wunder’s Declaration does not attest to sales during the entire class period, however, Defendants use warranty data to support that AWP sold 2,901 water 5 Defendants arrived at this number by assuming that each class member overpaid by $2,000, as here. See ECF No. 1. They then multiplied that by a factor of five to account for punitive damages under the New Jersey Punitive Damages Act, see N.J.S.A. § 2A:15-5.14(b), and added a thirty-percent attorneys’ fee. They subsequently agreed that punitive damages are not available. See Pl. Br., at 13; Opp. Br., at 2. treatment systems between April 2013 and March 2019. See Opp. Br., at 9. They then rely on Wunder’s assertion that a precipitation test occurred half of the time to deduce that the putative class includes 1,451 members. Compare ECF No. 47, Ex. 6, ¶ 8, with Opp. Br., Ex. 2., at 43:13- 15. Because the parties otherwise agree that “$7,800 is potentially at-issue for each putative class member—$2,000 in compensatory damages, trebled to $6,000 by statute, plus $1,800 in attorneys’ fees [per consumer],” Opp. Br., at 2, Defendants conclude that the amount in controversy is $11,310,000. Id. at 12-13, 17. Defendants insist that the Wunder Declaration is reliable because Wunder’s conclusion is based on his 23 years of experience as president, secretary, and treasurer of AWP, see id., Ex. 2, at 40-41, 43, 45-46, and his “hands-on” management style. Id. at 64-65. Plaintiff disagrees, pointing to various statements made by Wunder himself that belie his assertion as to the test-to- sales ratio: (1) AWP does not track precipitation tests in any formal manner, id. at 73; (2) Wunder has never conducted an investigation to determine how many times a salesperson performed a precipitation test prior to a sale, id. at 42-43; (3) to know whether a salesperson performed a precipitation test, Wunder would have to be present during the sale, id. at 74, but he did not attend any sales except for one month in 2013, id. at 24-25; (4) Wunder did not provide training to AWP’s in-home salespeople, id. at 73; (5) Wunder stated that “[t]he exact demonstrations performed, including any precipitation test . . . , are up to each salesman,” id. at 16; and (6) in any given sale, a precipitation test “may be skipped, and that’s a decision that’s based on an individual basis.” Id. at 15-16, 43. Not only is the Wunder Declaration insufficient evidence, but I find that it is also unreliable. It is important to reiterate that Wunder attests to just 1,200 sales between 2018-2019 in his Declaration. To show the potential class size from 2013-2019, Defendants resort to extrapolating from the warranty data. Even assuming that is an appropriate starting point, Wunder never attests to the test-to-sales ratio during the entire class period, but only the last year. Defendants offer no evidence whatsoever as to the ratio from 2013-2017. As such, there is no basis for the inference that the fifty percent rate should apply to sales not made in 2018-2019, even if I take it (as well as the warranty data) at face value. More importantly, notwithstanding that deficiency, Defendants have not come close to “bridg[ing] the gap” between the Declaration and the “speculative conclusion[]” that fifty percent of AWP’s sales—whether 2,901 or 1,200, i.e., during whatever time period—followed a precipitation test. Toribio, 2019 WL 4254935, at *2. The problem is that Defendants fail to explain at all, much less “in detail,” exactly how Wunder arrived at a fifty percent test-to-sales ratio or, “more importantly, why [that ratio is] realistic and reliable,” Wilder v. Roma Food Enterprises, Inc., No. 14-8123, 2015 WL 6164032, at *4 (D.N.J. Oct. 19, 2015), leaving me “mystified” as to Wunder’s factual basis for this assertion. Id. Defendants all but “pull” the test rate “out of thin air,” a problem they try but fail to overcome by asserting—again without evidence—that this rate is “conservative.” Toribio, 2019 WL 4254935, at *3; Opp. Br., at 3. In other words, regardless of whether warranty data or the Wunder Declaration is the starting point for the class size, Defendants provide neither an objective basis from which to infer that half of AWP’s sales followed a precipitation test, such as a record kept in the regular course of business, see Portillo, 2016 WL 1029854, at *8 (“[A] defendant must justify its jurisdictional assertions with some objective, factual basis.”), nor a subjective basis from which I could draw a test rate, such as an affidavit from one of the twelve salespeople who performed in-home tests for Wunder. Cf. Brill v. Countrywide Home Loans, Inc., 427 F.3d 446 (7th Cir. 2005) (crediting an affidavit from the employee who personally sent 3,800 junk faxes which, when multiplied by the statutory damages per fax, exceeded $5,000,000). Wunder’s Declaration also fails because it shows that he cannot “provide” a number that should “readily” exist, Hoffman v. Teleflora LLC, No. 15-4810, 2016 WL 438962, at *1 (D.N.J. Feb. 3, 2016), and which he is in the best position to “simply come forward with.” Rosenblatt, 2016 WL 3546579, at *5. Rather, Defendants have derived the class size based on a series of inferences, an amalgamation of data, and a narrow reading of the Wunder Declaration itself, whereby they use Wunder’s test-to-sales ratio without question but employ a different sales figure. Wunder’s explanation as to why “[the test rate] could not be provided” in a straightforward manner, without calculation or inference, further undermines Defendants’ argument: AWP has never tracked the rate of precipitation tests. Far from “a general search of its records yield[ing] an overly broad result,” Leff, 2015 WL 3486883, at *4, which is still insufficient to establish jurisdiction by a preponderance of the evidence, no records exist to search here. This renders the Wunder Declaration not just “wholly conclusory,” id., but “perfunctory,” id., “entirely speculative,” Hoffman, 2016 WL 423648, at *3, and “tenuous.” Edginton, 2011 WL 3625599, at *2. Portillo demonstrates just how far Defendants are from meeting their jurisdictional burden. In that case, the party asserting jurisdiction “review[ed] the records” of thirty-four out of one-hundred independent contractors “encompassed by the proposed class” and provided a “lengthy spreadsheet, i.e., a factual basis, substantiating this assertion.” 169 F. Supp. 3d at 597; see also Wilder, 2015 WL 6164032, at *4. In short, Wunder lacks personal knowledge of the test-to-sales ratio during 2018-2019, and provides “no factual support for [his] assertion.” Rosenblatt, 2016 WL 3546579, at *3. Defendants then assume that the same test rate applies to all sales during the class period, and “delve[] into a structured calculation that seems designed to “conveniently arrive[] at” an amount in controversy exceeding $5,000,000. Hoffman, 2016 WL 438962, at *1. To nevertheless assume that a precipitation test occurred in fifty percent of 2,901 sales is to simply take Defendants’ word for it, because I cannot objectively assess the reliability of the test rate in 2018-2019 nor, assuming it is accurate, whether it obtained in 2013-2017. Although CAFA favors federal court jurisdiction, Defendants, relying on Wunder’s “estimate,” stretch it too far. Accordingly, I find that Defendants have failed to establish the amount in controversy by a preponderance of the evidence.6 IV. CONCLUSION Because it lacks a sufficient factual basis, I find that the Wunder Declaration is not reliable evidence to establish the amount in controversy. As such, Defendants have failed to prove by a preponderance of the evidence that jurisdiction exists under CAFA. Plaintiffs’ Motion to Remand is therefore GRANTED and this matter is remanded to New Jersey Superior Court, Law Division, Hunterdon County, for further proceedings. DATED: December 28, 2020 /s/ Freda L. Wolfson Freda L. Wolfson U.S. Chief District Judge 6 This is not to say that Defendants could not under any circumstances prove a class size that would meet the amount in controversy. My findings here are focused on the deficiencies in their proof. Indeed, Defendants could have supported the jurisdictional amount in any number of ways. For instance, (1) Wunder could have testified to total sales and the test-to-sales ratio during the entire class period rather than just 2018-19, or Defendants could have (2) sought an affidavit from at least one of Wunder’s twelve salespeople describing his/her personal experience administering precipitation tests, (3) obtained declarations from other Rainsoft dealers in New Jersey, which, according to Defendant without any substantiation, sold 1,725 water treatment systems during the class period, as to their sales numbers or test rates, or (4) provided some documentary support for the test-to-sales ratio Wunder alleged in his Declaration. Defendants did not take any of these steps, even though all evidence necessary to prove the amount in controversy is within their possession and control.