On February 2, 2021, the Eleventh Circuit weighed in on the “ascertainability” debate raging in the federal courts – specifically, whether plaintiffs must show that it would be “administratively feasible” to identify class members before the class can be certified. The term “ascertainability” is not in the text of Federal Rule of Civil Procedure 23. Some courts, however, view ascertainability as an implicit requirement of a properly defined class. Other courts take it a step further and embrace a “heightened ascertainability” standard – i.e., “administrative feasibility” – and deny certification when plaintiffs fail to prove that the process for identifying absent class members will be administratively feasible.

In Cherry v. Dometic Corp.,[1] the Eleventh Circuit rejected the “heightened ascertainability” standard and held that administrative feasibility is not a requirement for class certification. In so doing, the court joined the Second, Sixth, Seventh, Eighth, and Ninth Circuits on this issue. Nevertheless, the Eleventh Circuit held that administrative feasibility remains relevant to the “manageability” balancing test of Rule 23(b)(3)’s superiority requirement, which is applicable to damages classes.

Courts that consider administrative feasibility a precondition to class certification are more likely to deny class certification than are courts that consider it as part of the manageability balancing test under Rule 23(b)(3). For the former, the lack of an administratively feasible class is a dead end on the road to certification. But for the latter, it can be merely a roadblock that leaves open other alternate routes to manage the class.

Ascertainability: The Implicit Requirement

Federal Rule of Civil Procedure 23 requires class action plaintiffs to meet specific, enumerated requirements before a court can certify the class. The rule requires that plaintiffs prove numerosity, commonality, typicality, adequacy, and – for a class seeking damages under Rule 23(b)(3) – predominance of common questions and superiority of the class action device. But there is another, implicit requirement that continues to spark division among the circuits: ascertainability.

Ascertainability refers to a court’s ability to identify the individuals in the class. In general, a court must be able to ascertain who the members are by using objective criteria—for example, all purchasers of a certain product within a certain geographic area. Some courts consider “ascertainability” to relate only to the class definition—whether the class is defined clearly and objectively. If the class definition is too vague or subjective, then the court will not be able to “ascertain” the class members.

But other courts have adopted a “heightened ascertainability” standard and held that proof of ascertainability relates not only to the class definition, but also to the administrative feasibility of identifying class members. These courts generally require the plaintiff to prove that the process of identifying class members is reliable and manageable and does not require individual factual inquiries. Plaintiffs counter that this reading of Rule 23 is strained and sets the bar too high; they argue that they need prove only that class members are capable of determination using specific, objective criteria.

Courts have found both arguments compelling, which has created a circuit split.

Cherry v. Dometic Corp. Sides with the Majority

In Cherry v. Dometic Corp., the Eleventh Circuit joined five other circuits in holding that Rule 23 does not require plaintiffs to prove administrative feasibility.

In Cherry, owners of allegedly defective refrigerators made for use in RVs filed a class action against the manufacturer. Class representatives sought to certify a class of “all persons who purchased in selected states certain models of Dometic refrigerators that were built since 1997.”

The plaintiffs moved for class certification, and the manufacturer challenged the motion by arguing that the class was not ascertainable. The plaintiffs had proposed using a combination of DMV records, manufacturer’s records, and class member affidavits, if necessary, to identify the class members. The manufacturer argued that this method was not administratively feasible because those records could not readily and accurately identify class members, and because relying on class member affidavits would be unworkable. Relying on two unpublished decisions from the Eleventh Circuit, the District Court for the Southern District of Florida agreed with the manufacturer and denied class certification.

First, the district court found that the DMV records would identify only the owners of RVs – not necessarily owners of RVs with Dometic refrigerators, who would make up the class. Second, the court found that the manufacturer’s records (from a previous recall campaign) could be helpful only for notifying potential class members but not for identifying them. The recall campaign had targeted all potential owners, which meant that those records would include 1.4 million individuals who were not class members. The court explained that using these records to identify class members would also require three steps: send out notices to all individuals, wait for those actually in the class to self-identify after receiving notice, and then wait again for those who self-identified to reach back out to the court. This process, the court determined, was not feasible. Finally, the court noted that the plaintiffs had failed to provide a detailed plan for how class members could self-identify by affidavit. The court denied certification, and the class representatives appealed.

The Eleventh Circuit vacated and remanded, holding that plaintiffs need not prove administrative feasibility to be entitled to class certification. The court rejected the “administratively feasible” standard for ascertainability, determining instead that a “class is ascertainable if it is adequately defined.” In doing so, the Eleventh Circuit joined five other circuits that follow the so-called “weak” version of ascertainability. Using Webster’s dictionary definition of “ascertain,” the court reasoned that something is ascertainable if it is “capable of determination.” Because the class was capable of determination, even if it was not capable of convenient determination, the Court of Appeals vacated the district court’s denial of class certification.

The circuit split over ascertainability now sits at six to three, with circuits trending away from the heightened standard. The First, Third, and Fourth Circuits require class representatives to meet the heightened ascertainability standard for class certification, while the Second, Sixth,[2] Seventh, Eighth, Ninth, and (now) Eleventh Circuits do not.[3] The Fifth Circuit, in an unpublished decision, also appears to have rejected the heightened standard.[4] The Tenth and D.C. Circuits have not addressed the issue; district courts in the D.C. Circuit appear to reject the heightened standard,[5] while district courts in the Tenth Circuit are divided.[6]

What’s the Difference and Why Does It Matter?

How and when courts consider whether a class is ascertainable can lead to starkly different outcomes. Plaintiffs have greater difficulty getting a class certified in a circuit that requires the class to be “administratively feasible.” In contrast, classes may be more readily certified in circuits on the other side of the split, which evaluate ascertainability later in the analysis – as part of the superiority/manageability requirement in Rule 23(b)(3).

From a practical standpoint, it can be more difficult in heightened-ascertainability jurisdictions to bring class actions in cases involving low-cost goods or services or purchases made long ago. These consumers are less likely to have records or receipts, and the defendant’s records are usually unable to identify these consumers, especially if sales were made through a third party.

Meanwhile, courts that do not require administrative feasibility focus on whether the class is defined adequately. If the class is capable of determination, it may be certified, provided that the other Rule 23 requirements are satisfied. These courts then consider administrative feasibility when evaluating the manageability of a class action under Rule 23(b)(3).

Rule 23(b)(3) imposes additional requirements on plaintiffs seeking damages. For example, plaintiffs must show that a class action is the superior method for adjudicating the controversy, which involves considering the manageability of the class. If the class is not manageable, then the court may deny class certification, or it may direct further subdivision of the class or require plaintiffs to give detailed plans for notifying class members and managing the litigation.

When courts consider administrative feasibility in the context of Rule 23(b)(3), the costs and benefits of the class action device are also evaluated. This means that if the class is capable of determination, but the identification process is not administratively feasible, these courts may still certify the class, although they may direct some of the management actions noted above.

Key Takeaways

Parties to a class action can expect different interpretations and applications of ascertainability depending on the circuit in which the case is brought. In the First, Third, and Fourth Circuits, classes without “administratively feasible” methods for determining members may never survive the certification stage, but those same classes might be certified and go all the way to settlement or trial in the Second, Sixth, Seventh, Eighth, Ninth, and Eleventh Circuits.

Ultimately, there are two possible resolutions to this circuit split: either the U.S. Supreme Court will resolve it or the circuits may fall in line with each other. Shortly after the death of Justice Scalia in 2016, the Supreme Court declined the opportunity to address the issue. Now with three new Justices and the circuits split six-to-three, resolution could be on the horizon.


[1] 986 F.3d 1296 (11th Cir. 2021).

[2] Although in 2015 the Sixth Circuit rejected the Third Circuit’s “heightened ascertainability” approach, the Sixth Circuit has not been entirely clear on the applicable standard, and at times has appeared to adopt the “administrative feasibility” standard. See Lyngaas v. Curaden AG, 992 F.3d 412, 428 (6th Cir. 2021) (“the class definition must be sufficiently definite so that it is administratively feasible for the court to determine whether a particular individual is a member of the proposed class”).

[3] Cherry, 986 F.3d at 1302 (identifying circuit split).

[4] See Seeligson v. Devon Energy Production Co., 761 F. App’x 329, 334 (5th Cir. 2019).

[5] See, e.g., In re McCormick & Company, Inc., Pepper Prods. Mktg. & Sales Practices Litig., 422 F. Supp. 3d 194, 242-43 (D.D.C. 2019) (rejecting “heightened ascertainability” and finding the Seventh and Ninth Circuits’ approach to be “persuasive”).

[6] Compare Cline v. Sunoco, Inc. (R&M), 333 F.R.D. 676, 688 (E.D. Okla. 2019) (requiring that the “class description [be] sufficiently definite so that it is administratively feasible for the court to determine whether a particular individual is a member”) with In re EpiPen Mktg., Sales Practices & Antitrust Litig., No. 17-md-2785-DDC-TJJ, 2020 WL 1180550, at *9-12 (D. Kan. Mar. 10, 2020) (rejecting the “administratively feasible” standard and predicting that the Tenth Circuit will, too).