In the

    United States Court of Appeals
                For the Seventh Circuit
                    ____________________
No. 24-2574
MONICA RICHARDS, individually and on behalf of all other
similarly situated individuals,
                                       Plaintiff-Appellee

                                v.

ELI LILLY & COMPANY and LILLY USA, LLC,
                                    Defendants-Appellants.
                    ____________________

        Appeal from the United States District Court for the
         Southern District of Indiana, Indianapolis Division.
     No. 1:23-cv-00242-TWP-MKK — Tanya Walton Pratt, Judge.
                    ____________________

    ARGUED JANUARY 28, 2025 — DECIDED AUGUST 5, 2025
                ____________________

   Before HAMILTON, KIRSCH, and LEE, Circuit Judges.
    KIRSCH, Circuit Judge. The Fair Labor Standards Act au-
thorizes similarly situated employees to collectively sue em-
ployers for violations of the Act. 

(b). District
courts take an active and early role in the management of
these mass actions—commonly known as collective actions—
including by issuing notice to potential plaintiffs so that they
may opt to join the collective. At issue in this interlocutory
2                                                  No. 24-2574

appeal is the showing necessary to procure this court-issued
notice.
    Absent meaningful guidance from the FLSA or higher
courts, district courts have been left to fashion their own
standards. Today, we clarify this area of the law and provide
district courts in our circuit with a uniform, workable frame-
work for assessing the propriety of notice to a proposed col-
lective. As we explain in greater detail below, district courts
must consider both parties’ evidence with respect to similar-
ity and may issue notice to potential plaintiffs when the
named plaintiffs have raised at least a material factual dispute
as to the similarity of potential plaintiffs.
                               I
                               A
    To ensure broad and robust enforcement, the Fair Labor
Standards Act of 1938 (FLSA) permits employees to bring so-
called collective actions to sue employers for violations of the
FLSA on behalf of themselves and other similarly situated
employees. 

(b). The Age Discrimination in Em-
ployment Act of 1967 (ADEA) incorporates this enforcement
provision, permitting employees to band together in collec-
tive actions when suing an employer for age discrimination.

 § 626(b); Hoffmann-La Roche Inc. v. Sperling, 

,
167–68 (1989). The statutory text authorizing collective actions
is sparse, however, and explains neither what it means to be
similarly situated nor by what standard or at what stage in
the proceedings courts must assess the similarity of the col-
lective:
       An action … may be maintained against any
       employer … by any one or more employees for
No. 24-2574                                                      3

       and in behalf of himself or themselves and other
       employees similarly situated. No employee
       shall be a party plaintiff to any such action un-
       less he gives his consent in writing to become
       such a party and such consent is filed in the
       court in which such action is brought.

(b).
    Collective actions are a unique enforcement mechanism.
Though, at first blush, they resemble a Federal Rule of Civil
Procedure 23 class action, the two actions are “fundamentally
different” in practice. Genesis Healthcare Corp. v. Symczyk, 

 (2013). In a class action for damages under Rule
23(b)(3), a named plaintiff represents the rights of absent non-
party plaintiffs who will be bound by the disposition of the
case unless they opt out of the class. See 7B Wright & Miller’s
Federal Practice and Procedure § 1807 (3d ed. 2005). As we
stated in Vanegas v. Signet Builders, Inc., 

 (7th Cir.
2024), collective actions are, by contrast, “a consolidation of
individual cases, brought by individual plaintiffs.” 

(quotation omitted). Plaintiffs must affirmatively opt in to join
a collective action. And unlike in a class action, each plaintiff
who joins the collective enjoys full party status. See Wright &
Miller, supra, § 1807.
   “The twin goals of collective actions are enforcement and
efficiency.” Bigger v. Facebook, Inc., 

, 1049 (7th Cir.
2020). Litigating collectively furthers these goals by lowering
the “individual costs to vindicate rights” and enabling com-
mon legal issues to be resolved in one fell swoop. Hoffmann-
La Roche, 

. But due to their opt-in nature, the
benefits of collective actions can only be effectively realized if
other similarly situated employees are made aware of the
4                                                     No. 24-2574

pending action. Acknowledging this fact, the Supreme Court
has held that federal district courts may issue notice of a
pending collective action to “potential plaintiffs” so that they
may make “informed decisions about whether to participate.”

 at 169–70, 172–73. The Court declined, however, to define
who qualifies as a potential plaintiff or what showing plain-
tiffs must make to secure notice, confirming “the existence of
the trial court’s discretion” to facilitate notice but “not the de-
tails of its exercise.” 

 at 169–70.
    With minimal guidance from Congress or the Court, dis-
trict courts have largely been left to devise their own stand-
ards for facilitating notice to similarly situated employees.
Over the last few decades, most district courts—including
those in our circuit—have followed the Lusardi approach,
which originated in a New Jersey district court. See Lusardi v.
Xerox Corp., 

 (D.N.J. 1987). Though it has
many variations, Lusardi generally divides the management
of a collective action into two steps.
    At the first step, a plaintiff seeking notice to a proposed
collective must make a “modest factual showing sufficient to
demonstrate that they and potential plaintiffs together were
victims of a common policy or plan that violated the law.”
Strait v. Belcan Eng'g Grp., 

 (N.D. Ill.
2012) (quotation omitted). A plaintiff’s evidentiary burden at
this stage is minimal, and many courts refuse to “make merits
determinations, weigh evidence, determine credibility, or
specifically consider opposing evidence presented by a de-
fendant.” Iannotti v. Wood Grp. Mustang, 

,
653 (S.D. Ill. 2022) (quotation omitted). Once a plaintiff makes
this “modest” showing, the court issues notice to prospective
plaintiffs, who may then opt in to the collective action by
No. 24-2574                                                      5

filing written consent with the court. Occasionally, if signifi-
cant discovery has already taken place, courts will impose a
higher level of scrutiny—referred to as a “modest-plus” or
“intermediate” standard—before issuing notice. See, e.g.,
Hawkins v. Alorica, Inc., 

 (S.D. Ind. 2012);
O'Neil v. Bloomin' Brands Inc., 

, 776 (N.D.
Ill. 2023).
    Step two occurs once opt-in and discovery are complete,
at which point the defendant typically moves to challenge
whether the collective is similarly situated. Iannotti, 603 F.
Supp. 3d at 654. Now, with the benefit of more specific infor-
mation about the collective’s membership, the court engages
in a more rigorous review to determine whether the plaintiffs
are, in fact, similarly situated. If not, the individual plaintiffs’
claims may be severed. See Alvarez v. City of Chicago, 

 (7th Cir. 2010) (“When a collective action is decerti-
fied, it reverts to one or more individual actions on behalf of
the named plaintiffs.”).
    Lusardi’s two steps are commonly but misleadingly re-
ferred to as conditional certification (step one) and decertifi-
cation (step two). These stages are not to be confused with
class certification under Rule 23, however. See Clark v. A&L
Homecare & Training Ctr., LLC, 

, 1009 (6th Cir.
2023) (rejecting the use of this terminology in the collective
action context). Conditional certification does not produce a
legal class or join any parties to the action. Rather, “[t]he sole
consequence of conditional certification is the sending of
court-approved written notice to employees.” Genesis, 

. Thus, at step two, there is technically no collective
to “decertify.”
6                                                            No. 24-2574

    In recent years, Lusardi’s lenient notice standard has be-
come the object of increasing scrutiny. Because most collective
actions settle rather than proceed to trial, the issuance of no-
tice, though it has no legal effect, carries significant practical
implications for parties. For plaintiffs, broad and early notice
helps to increase the size of the collective. This lowers costs,
improves plaintiffs’ bargaining position, and makes it easier
to recruit suitable counsel. See Hoffmann-La Roche, 

. Expeditious notice is particularly important in FLSA
cases. Under the FLSA, the statute of limitations is not auto-
matically tolled to the date of the first filing for plaintiffs who
have yet to opt in. 

(b). This means that delays
in notice risk limiting or even running out the clock on puta-
tive plaintiffs’ claims. 1 Notice also matters greatly to defend-
ants, who worry that overly permissive notice standards will
allow plaintiffs to artificially expand the size of a collective,
“increas[ing] pressure to settle, no matter the action’s merits.”
Bigger, 947 F.3d at 1049. Notice to employees who are ineligi-
ble to join a collective may also risk stirring up further litiga-
tion against an employer, especially if court-issued notice is
perceived as an invitation for “those employees to bring suits
of their own.” Clark, 68 F.4th at 1010.
   Consensus as to the proper standard for notice remains
elusive. Several circuits have approved the use of some ver-
sion of Lusardi’s two-step approach. See Harrington v. Cracker
Barrel Old Country Store, Inc., 

, 681–83 (9th Cir.


    1 This issue is not present in ADEA collective actions, at least in this

circuit, because we have interpreted the ADEA to permit opt-in plaintiffs
to piggyback on the timely filed claim of the named plaintiff, so long as
the named plaintiff alleges class-wide discrimination. See Anderson v.
Montgomery Ward & Co., 

, 1016–17 (7th Cir. 1988).
No. 24-2574                                                               7

2025) (citing Campbell v. City of Los Angeles, 

,
1108–10 (9th Cir. 2018)) (holding that a district court did not
abuse discretion by following the two-step process); Thiessen
v. Gen. Elec. Cap. Corp., 

, 1105 (10th Cir. 2001)
(finding “no error” with a district court’s choice to use the
two-step approach); Myers v. Hertz Corp., 

, 554–55
(2d Cir. 2010) (describing the two-step approach as “sensible”
but not required); Hipp v. Liberty Nat'l Life Ins. Co., 

 (11th Cir. 2001) (same).
     Recently, two circuits have expressly rejected Lusardi’s
modest notice standard in favor of a heightened burden of
proof. The Fifth Circuit now holds that notice may only issue
if, at the outset of the case, plaintiffs can demonstrate that no-
tice recipients are “actually similar to the named plaintiffs,”
ostensibly by a preponderance of the evidence. Swales v.
KLLM Transp. Servs., L.L.C., 

, 434 (5th Cir. 2021).
Taking a slightly different tack, the Sixth Circuit has con-
cluded that notice may issue only where plaintiffs can show a
“strong likelihood” that employees are similarly situated.
Clark, 68 F.4th at 1011. Until now, our circuit has yet to directly
address the issue. 2
                                     B
   In 2022, Monica Richards applied for a promotion to be-
come District Sales Manager for one of Eli Lilly’s Boston-


    2 In Bigger, we passingly observed that while some courts in our circuit

follow the Lusardi approach, we have never required it. 947 F.3d at 1049
n.5. Richards relies on language in In re New Albertsons, Inc., No. 21-2577,

 (7th Cir. Sept. 1, 2021), to suggest that we have rejected
critiques of Lusardi, but such non-precedential orders do not “constitute
the law of the circuit.” 7th Cir. R. 32.1(b).
8                                                 No. 24-2574

based sales teams. A six-year veteran of the company in her
early fifties, Richards had already been occupying the role on
an interim basis for nearly six months. When the promotion
was given instead to a much younger employee with less sales
experience, Richards sued Eli Lilly & Company and Lilly
USA, LLC, (collectively “Eli Lilly”) in federal court, alleging
age discrimination in violation of the ADEA, 

et seq., and the Massachusetts Anti-Discrimination Law,
Mass. Gen. Laws ch. 151B, § 4(1B) (2024).
    On her ADEA claim, Richards moved to conditionally cer-
tify a collective action, asserting that the unfavorable treat-
ment she experienced was part of a broader pattern of age dis-
crimination against Eli Lilly’s older employees. To alert other
potential plaintiffs to her suit, Richards requested that the
court send notice of the action to all “Eli Lilly employees who
were 40 or older when they were denied promotions for
which they were qualified, since February 12, 2022.” The
members of this proposed collective were similarly situated,
Richards alleged, because they had all been plausibly harmed
by a companywide initiative to support and retain “Early Ca-
reer Professionals” (i.e., workers with less than two years of
postgraduate experience) at the expense of older, more expe-
rienced employees.
    Though the parties debated the appropriate standard for
notice, the district court ultimately followed Lusardi and re-
fused to consider Eli Lilly’s evidence opposing the similarity
of the proposed collective. Concluding that Richards had
made the requisite modest factual showing, the district court
conditionally certified the collective and agreed to issue no-
tice. Recognizing, however, that we have not yet articulated a
standard to govern the issuance of notice, the district court
No. 24-2574                                                      9

certified the question for interlocutory review under 

(b). The proper facilitation of a collective action is
largely committed to a district court’s case management dis-
cretion, see Hoffmann-La Roche, 493 U.S. at 170–73, and condi-
tional certification is not a final appealable decision under 

. Accordingly, we rarely review a district court’s
decision to issue notice. In light of the Fifth and Sixth Circuit’s
recent decisions rejecting Lusardi’s “modest showing” stand-
ard, we agreed to take this interlocutory appeal. This should
not signal, however, that we intend to accept frequent inter-
locutory appeals from notice decisions or micromanage the
notice process.
                                II
    We review a district court’s decision to issue notice of a
collective action for abuse of discretion but review the legal
conclusions underlying that decision de novo. Bigger, 947 F.3d
at 1048.
    The time has come for us to offer clearer guidance on the
proper facilitation of notice. As more courts have begun to
question Lusardi, the standard for issuing notice in our circuit
has begun to vary by court. See, e.g., McColley v. Casey's Gen.
Stores, Inc., 

, 776 (N.D. Ind. 2021) (observ-
ing that Lusardi “remains the dance of this circuit—a[t] least
for the time being”); Laverenz v. Pioneer Metal Finishing, LLC,

, 613–16 (E.D. Wis. 2024) (rejecting Lusardi
and applying a preponderance of the evidence standard); Fil-
lipo v. Anthem Cos., No. 22-cv-00926, 

, at *2
(S.D. Ind. Dec. 30, 2022) (applying “the law of Rule 23 class
certification” to evaluate whether conditional certification is
appropriate). Such inconsistency leaves parties in the dark
with respect to a critical aspect of the collective action process.
10                                                    No. 24-2574

And if left unchecked, this disuniformity may eventually
jeopardize broader principles of equal treatment, creating the
appearance of a system that permits unpredictability and ar-
bitrariness. See Antonin Scalia, The Rule of Law as a Law of
Rules, 

, 1182 (1989).
    Richards resists, arguing that we cannot impose any uni-
form standard for notice because the management of collec-
tive actions is committed solely to the discretion of district
courts. No doubt, district courts generally enjoy “wide discre-
tion to manage collective actions.” Alvarez, 

.
And appellate courts cannot create prophylactic rules that
“circumvent or supplement legal standards” set out by Con-
gress or the Supreme Court. United States v. Tsarnaev, 

, 314–16 (2022). But, here, there is no legal standard to dis-
place. In Hoffmann-La Roche, the Court confronted only the
“narrow question” of whether district courts may play “any
role” at all in facilitating notice of a collective action. 

. Richards relies heavily on a single line from Hoffmann-
La Roche: “We confirm the existence of the trial court’s discre-
tion, not the details of its exercise.” 

. But this remark
is hardly an express delegation of sole and open-ended au-
thority to district courts. Indeed, even as it left the “details” of
how notice is facilitated for another day, the Court empha-
sized that the district court’s discretion with respect to notice
is not “unbridled.” 

.
    Imposing guardrails on the exercise of a district court’s
discretion is the normal business of appellate courts. When
we review discretionary district court decisions, we typically
do so deferentially but against defined legal standards. In the
class action context, for example, we have concluded that the
prerequisites for class certification must be proven by a
No. 24-2574                                                     11

preponderance of the evidence, see Messner v. Northshore
Univ. HealthSystem, 

 (7th Cir. 2012), even
though Rule 23 confers “broad discretion” on district courts
to determine whether certification is appropriate, Arreola v.
Godinez, 

 (7th Cir. 2008) (quotation omitted).
More to the point, we have already outlined some contours of
a district court’s discretion to issue notice to a collective. See
Bigger, 947 F.3d at 1050 (describing steps district courts must
follow to investigate and resolve arbitrability disputes prior
to issuing notice); Woods v. N.Y. Life Ins., 

 (7th
Cir. 1982) (holding that the district court’s power to regulate
notice does not include the power to forbid sending notice al-
together). The issues presented in this case—who qualifies as
a potential plaintiff under the FLSA, what evidence may be
considered to make this showing, and the requisite burden of
proof—are exactly the sorts of legal questions that ordinarily
fall within our purview. Cf. Ramirez v. T&H Lemont, Inc., 

 (7th Cir. 2016) (a district court’s decision to dis-
miss a case for misconduct is discretionary but “distinct from
the standard of proof by which the underlying facts must be
proven”).
                                III
    Having established the need for a uniform notice stand-
ard, we turn now to the task of defining that standard. Be-
cause the text of the FLSA is silent with respect to notice, our
analysis begins with Hoffmann-La Roche, from which we ex-
tract three core principles to guide the proper facilitation of
notice. The first is the importance of ensuring that notice is
timely and accurate—otherwise, the Court cautioned, the
benefits of collective actions cannot be effectively realized.
Hoffmann-La Roche, 493 U.S. at 170–71. Second is the principle
12                                                          No. 24-2574

of judicial neutrality: a district court’s involvement in the no-
tice process must remain “distinguishable in form and func-
tion from the solicitation of claims,” and courts must facilitate
notice in such a way as to “avoid even the appearance of judi-
cial endorsement of the merits of the action.” Id. at 174. Third,
district courts must use their discretion to “govern[] the con-
duct of counsel and the parties” to prevent abuses and ensure
that the joinder of additional parties “is accomplished in an
efficient and proper way.” Id. at 171–73 (quotations omitted).
    Below, we consider the various notice standards proposed
by Richards and Eli Lilly. But because we find each incon-
sistent with the Court’s instructions, in the end, we chart a
different path that more closely adheres to the principles con-
tained in Hoffmann-La Roche.
                                    A
   Richards argues that district courts should be permitted to
continue applying Lusardi’s modest notice standard. We rec-
ognize that the majority of district courts have opted to apply
Lusardi. 3 It is difficult to know, however, whether Lusardi’s
ubiquity reflects genuine consensus that it is “an effective
tool,” Hipp, 

, or is merely the product of inertia
or “anchoring bias,” Clark, 68 F.4th at 1008 (quotation omit-
ted). Indeed, even as courts routinely employ the “modest” or
“modest-plus” standards from Lusardi, those standards have

     3 We don’t, however, attribute significance to the fact that the Court

in Hoffmann-La Roche affirmed a lower court decision that followed Lu-
sardi. See Sperling v. Hoffmann-La Roche, Inc., 

 (D.N.J.
1988). Hoffmann-La Roche dealt only with the narrow question of whether
court-issued notice was permissible at all; the Court never discussed the
two-stage certification process nor considered whether Lusardi was a per-
missible exercise of the district court’s discretion.
No. 24-2574                                                   13

been repeatedly criticized as unclear and difficult to apply.
See, e.g., McColley, 559 F. Supp. 3d at 775 (“The court has re-
searched deeply the origins of the ‘modest’ and ‘modest plus’
standards but thus far has not found or been presented with
a case that defines them.”); Swales, 985 F.3d at 440 (describing
Lusardi as an “amorphous and ad-hoc test [that] provides little
help in guiding district courts in their notice-sending author-
ity”). When the district court in this case certified Eli Lilly’s
interlocutory appeal, for instance, it noted that existing case
law does not resolve “what scrutiny should apply … at con-
ditional certification” nor “how the court should precisely de-
fine that scrutiny.” Richards v. Eli Lilly & Co., No. 1:23-cv-
00242-TWP-MKK, 

, at *5 (S.D. Ind. May 10,
2024) (alteration in original) (quotations omitted).
    Undoubtedly, Lusardi’s permissive notice standard has
practical advantages. It enables courts to expeditiously send
notice without weighing competing evidence, engaging in
protracted discovery, or delving into difficult merits issues.
But these rigid limits on what arguments and evidence courts
may consider favor timely notice at the expense of both judi-
cial discretion and neutrality. With their discretion curtailed
at stage one, courts applying Lusardi must typically wait to
sever a collective until after full opt-in and discovery are com-
plete—even if the dissimilarity of the collective could have
been readily established earlier. In such a case, issuing notice
is deeply inefficient, inviting “futile attempts at joinder” that
undermine judicial economy and unnecessarily increase liti-
gation costs. Bigger, 947 F.3d at 1050.
   A lenient and virtually unrebuttable notice showing also
threatens judicial neutrality. Sending notice to plaintiffs who
are ineligible to join the collective can generate significant
14                                                    No. 24-2574

discovery costs, incentivizing defendants to settle early rather
than attempt to “decertify” at step two. Such notice may also
be seen as “solicitation of those employees to bring suits of
their own,” Clark, 68 F.4th at 1010, transforming what should
be a neutral case management tool into a vehicle for stron-
garming settlements and soliciting claims. See Swales, 985 F.3d
at 442. Because its overly permissive notice standard places “a
judicial thumb on the plaintiff’s side of the case,” Bigger, 947
F.3d at 1050, we join the Fifth and Sixth Circuits in concluding
that the modest level of scrutiny commonly employed under
Lusardi inevitably conflicts with a district court’s obligation to
“maintain neutrality and to shield against abuse of the collec-
tive-action device,” id. We do not, however, mean to cast
doubt on the general proposition that a two-step approach to
notice can be an appropriate exercise of a district court’s dis-
cretion nor categorically disapprove every decision that per-
mitted notice with a citation to Lusardi.
    For its part, Eli Lilly argues we should adopt the Fifth Cir-
cuit’s preponderance of the evidence approach, Swales, 985
F.3d at 434, or, at the very least, the Sixth Circuit’s strong like-
lihood of similarity standard, Clark, 68 F.4th at 1011. We de-
cline to do either.
    Requiring plaintiffs to prove their similarity by a prepon-
derance of the evidence to secure notice is unworkable and
inconsistent with Hoffmann-La Roche. While some factual dis-
putes about the similarity of a proposed collective may be de-
finitively resolved prior to notice, others cannot be because
the evidence necessary to establish similarity resides with yet-
to-be-noticed plaintiffs. See Clark, 68 F.4th at 1010 (“[A]n em-
ployer’s records might show that an employee worked less
than 40 hours per week during a certain period; the employee
No. 24-2574                                                   15

might be ready to testify that she worked more.”). In such
cases, this heightened requirement for notice functions as an
insurmountable barrier for even meritorious collective ac-
tions. This result is incompatible with the Court’s express in-
struction that the “broad remedial goal of the [FLSA] should
be enforced to the full extent of its terms.” Hoffmann-La Roche,

.
    Recognizing the impracticality of a preponderance of the
evidence standard, the Sixth Circuit has endorsed a less de-
manding standard for notice: strong likelihood of similarity.
Clark, 68 F.4th at 1011. But when relevant evidence of similar-
ity is in the hands of individuals who are not yet parties to the
action, we doubt that plaintiffs will be able to show a strong
likelihood of similarity before notice any more than they
would be able to show it by a preponderance of the evidence.
Moreover, both heightened standards may foster delay and
inefficiency, requiring courts to facilitate pre-notice discovery
even when it’s readily apparent that such discovery is un-
likely to help resolve the factual dispute at hand.
     Though the notice frameworks outlined in Lusardi, Swales,
and Clark strike different compromises between timely notice
and judicial neutrality, they share a common weakness: each
imposes an inflexible notice standard that all but eliminates
judicial discretion. In doing so, they not only disregard a core
principle of Hoffmann-La Roche but also leave district courts
ill-equipped to efficiently resolve the varied factual disputes
that can arise when defining the scope of a collective action.
In our view, Hoffmann-La Roche requires something different.
16                                                                No. 24-2574

                                        B
    The plain text of the FLSA states that plaintiffs must be
similarly situated if they are to proceed collectively. 

(b). 4 Accordingly, once opt-in is complete, it’s sensible
that plaintiffs bear the burden of moving to certify their col-
lective action, at which point the court will assess whether
plaintiffs are, in fact, similarly situated and may proceed col-
lectively. And in the absence of statutory language dictating
otherwise, we presume that plaintiffs must establish their
similarity at the certification stage by a preponderance of the
evidence. See E.M.D. Sales, Inc. v. Carrera, 

, 52 (2025)
(“Faced with silence, courts usually apply the default prepon-
derance standard.”). Indeed, we have applied this default
standard to other procedural and case management decisions.
See Bell v. PNC Bank, Nat’l Ass'n, 

 (7th Cir.
2015) (noting that while we review class certification orders
for abuse of discretion, the party seeking class certification
“bears the burden of demonstrating that certification is
proper by a preponderance of the evidence”).
    Whether a plaintiff can reasonably be expected to make
this showing before notice, however, is a different question


     4 Though we have described the opt-in process in general terms as a

“system of permissive joinder,” Vanegas, 113 F.4th at 725 (quotations omit-
ted), we agree with our sister circuits that the similarly situated require-
ment distinguishes a collective action from an ordinary multiparty suit
achieved through Rule 20. See Waters v. Day & Zimmermann NPS, Inc., 

, 96 (1st Cir. 2022) (“The FLSA’s ‘similarly situated’ limitation for
collective actions displaces Rule 20 and limits the range of individuals
who may be added as opt-in plaintiffs by requiring that they be ‘similarly
situated.’”); Scott v. Chipotle Mex. Grill, Inc., 

, 520 (2d Cir. 2020);
Campbell, 

.
No. 24-2574                                                    17

altogether and turns largely on the nature of plaintiffs’ allega-
tions. For example, if the common thread connecting plain-
tiffs’ claims is an employer’s informal policy of requiring
work off the clock, it may be impossible to prove which em-
ployees were subject to that policy until opt-in plaintiffs are
identified. See, e.g., Smith v. Fam. Video Movie Club, Inc., No.
11 C 1773, 

, at *4 (N.D. Ill. Feb. 22, 2012) (ex-
plaining that an official corporate policy on working hours
might be rebutted by testimony from opt-in plaintiffs that
they were ordered not to adhere to the policy). But, in other
cases, it may be readily proven prior to notice that a chal-
lenged policy was limited in scope—for example, to only a
particular type of worker or geographic location. See, e.g.,
Jonites v. Exelon Corp., 

, 725–26 (7th Cir. 2008) (de-
scribing a collective action as “hopelessly heterogeneous”
where it included employees who did not work the day shift
and thus had no conceivable claim based on the employer’s
daytime lunch policy).
    With this in mind, we conclude that to secure notice, a
plaintiff must first make a threshold showing that there is a
material factual dispute as to whether the proposed collective
is similarly situated. By this, we mean that a plaintiff must
produce some evidence suggesting that they and the mem-
bers of the proposed collective are victims of a common un-
lawful employment practice or policy. A plaintiff’s evidence
of similarity need not be definitive, but defendants must be
permitted to submit rebuttal evidence and, in assessing
whether a material dispute exists, courts must consider the
extent to which plaintiffs engage with opposing evidence. We
acknowledge that, at this stage, the relevant evidence will
likely come in the form of affidavits and counter-affidavits,
including those that a plaintiff may wish to submit after
18                                                   No. 24-2574

seeing a defendant’s rebuttal evidence. That’s fine; courts rou-
tinely assess whether competing affidavits create a material
factual dispute.
    This threshold showing imposes a meaningful yet appro-
priate evidentiary burden on the plaintiff, reflecting both the
preliminary stage of the proceedings and the practical signif-
icance of notice to both parties. A lower standard would per-
mit notice based on little more than allegations, conflicting
with the Court’s warning against encouraging “the solicita-
tion of claims.” Hoffmann-La Roche, 

. By con-
trast, a substantially higher standard risks leaving some
plaintiffs in limbo, unable to make the required showing
without access to evidence held by individuals who are not
yet parties to the case.
     We stress, however, that a plaintiff is not automatically en-
titled to notice upon establishing the existence of a material
dispute as to similarity. Once the district court is satisfied that
there is at least a material dispute, the decision to issue notice
will depend on its assessment of the factual dispute before it.
And here, as Hoffmann-La Roche instructs, we rely on the
sound discretion of the district court to facilitate notice in an
efficient way that strikes an appropriate balance between
timely notice and judicial neutrality.
    If the district court is persuaded that the evidence neces-
sary to resolve a similarity dispute is likely in the hands of
yet-to-be-noticed plaintiffs, it may proceed with a two-step
approach—that is, it may issue notice to the proposed collec-
tive while postponing the final determination as to whether
plaintiffs are similarly situated until plaintiffs move for certi-
fication after opt-in and discovery are complete. Alterna-
tively, if the court is confident that a similarity dispute can be
No. 24-2574                                                  19

resolved by a preponderance of the evidence before notice, it
may authorize limited and expedited discovery to make this
determination and tailor (or deny) notice accordingly. This
flexibility promotes efficient case management, enabling a
district court to tailor its approach depending on the issues
and complexities of the case before it.
    This need not be an all-or-nothing determination. A dis-
trict court might decide that a subset of issues relating to the
similarly situated analysis are capable of definitive resolution
and narrow the scope of notice accordingly, even as it recog-
nizes that other disputes cannot be resolved until later in the
proceedings. And if a plaintiff fails to produce evidence that
establishes a material factual dispute, a district court might
deny the motion for notice without prejudice, subject to pos-
sible reconsideration if the plaintiff comes forward with fur-
ther evidence. The watchword here is flexibility, with respect
for the principles outlined in Hoffmann-La Roche and the reme-
dial goals of the FLSA and ADEA.
    This framework is consistent with Bigger, in which we held
that courts may not issue notice to an employee if a defendant
can show that a valid arbitration agreement bars the em-
ployee from participating in the action. 947 F.3d at 1050. As
Bigger recognizes, whether some members of a proposed col-
lective are bound by arbitration agreements is the type of fac-
tual dispute that can be readily resolved prior to notice and
opt-in. Our decision today builds on this logic and empowers
courts to recognize other similarity disputes that can likewise
be resolved swiftly and conclusively before notice.
   When district courts do decide to authorize pre-notice dis-
covery, we trust that they will carefully supervise this process
20                                                              No. 24-2574

to prevent abuses and unnecessary delay. 5 And while such
discovery should be conducted expeditiously, we are mindful
that even minimal pre-notice discovery risks running out the
clock on putative plaintiffs’ FLSA claims. To that end, district
courts should decide as soon as practicable whether notice is
appropriate and are empowered to use their equitable tolling
authority to ensure that plaintiffs are not unfairly disadvan-
taged by any delays in discovery. See Clark, 68 F.4th at 1014
(Bush, J., concurring) (“Equitable considerations support the
use of tolling for FLSA collective actions.”). Indeed, the equi-
table tolling of FLSA claims is already a familiar practice in
our circuit, see, e.g., Iannotti, 603 F. Supp. 3d at 657; Koch v.
Jerry W. Bailey Trucking, Inc., 

, 799 (N.D.
Ind. 2020), and that is a fair and appropriate measure, espe-
cially while disputes about notice are being resolved.
    Finally, we caution that this pre-notice discovery should
remain narrowly tailored to the similarly situated inquiry—
in other words, to the question of whether common issues of
fact or law make it more efficient to resolve plaintiffs’ claims
together. This inquiry should not devolve into an early adju-
dication of the merits. That said, courts need not refuse to con-
sider merits issues altogether. Some factual disputes about
similarity inevitably overlap with merits issues. A defendant
might, for example, present evidence that the proposed

     5 And, of course, even after notice and opt-in are complete, a district

court retains broad discretion with respect to the scope of discovery. Be-
cause a collective action might comprise tens, hundreds, or even thou-
sands of opt-in plaintiffs, a district court’s discretion includes carefully de-
limiting any additional discovery mindful of its obligation under Federal
Rule of Civil Procedure 1 to “secure the just, speedy, and inexpensive de-
termination” of the action and the “twin goals of collective actions … en-
forcement and efficiency.” Bigger, 947 F.3d at 1049.
No. 24-2574                                                      21

collective includes employees who are exempt from FLSA
protections. If true, this fact would certainly affect the merits
of those plaintiffs’ claims. But it is also directly relevant to as-
sessing whether the collective is similarity situated—that is,
“whether the merits of other-employee claims would be sim-
ilar to the merits of the original plaintiffs’ claims” such that
“collective litigation would yield ‘efficient resolution in one
proceeding.’” Clark, 68 F.4th at 1012 (quoting Hoffmann-La
Roche, 

). Accordingly, a district court need not
ignore evidence of dissimilarity at the pre-notice stage simply
because it touches on a merits issue. Rather, it is for the district
court to decide whether and to what extent a particular merits
question affects the similarly situated analysis.
                         *       *       *
    Our decision today realigns the issuance of notice in our
circuit with the core principles of Hoffmann-La Roche. It recog-
nizes the need for workable standards to guide district courts
in the process of issuing such notice and empowers them to
use their discretion to strike the proper balance between
timely notice and judicial neutrality. Because the district court
below did not have the opportunity to evaluate the propriety
of notice under the framework we adopt today, we vacate and
remand for further proceedings consistent with this opinion.
                                       VACATED AND REMANDED
22                                                    No. 24-2574

    HAMILTON, Circuit Judge, concurring in part and concur-
ring in the judgment. I am pleased to join almost all of Judge
Kirsch’s opinion for the court. His opinion provides im-
portant and appropriate guidance for issuing notice to people
who may be situated similarly to lead plaintiffs and who may
have similar claims under the FLSA or ADEA. His opinion
also preserves important flexibility and discretion for district
judges in managing these cases.
    My only point of disagreement is with the paragraph on
page 16 addressing what happens after notice is issued in
FLSA or ADEA cases to people who may be similarly situated
to the lead plaintiff. The draft opines without citing relevant
authority (a) that the plaintiff bears the burden of moving to
“certify” a collective action and (b) that similarity must be
proven by a preponderance of the evidence. We would do bet-
ter not to opine on either point in this case, where this passage
amounts to obiter dictum.
    Neither question is actually presented in this narrow in-
terlocutory appeal. We agreed to review only the standard for
issuing notice in the first place. And the majority’s answers to
these two extra questions may not be correct. They seem to
me a little too close to the Rule 23 process and standard for
certifying a true class action.
    As I see it, proceeding in a collective action may be better
understood as closer to presenting Rule 20 or 21 questions of
permissive joinder or misjoinder or Rule 42 questions of con-
solidation of cases for trial. The Supreme Court has twice de-
scribed § 216(b) as a “joinder process.” See Genesis Healthcare
Corp. v. Symczyk, 

, 70 n.1 (2013); Hoffmann-La Roche
Inc. v. Sperling, 

, 170 – 71 (1989) (district court has
“managerial responsibility to oversee the joinder of
No. 24-2574                                                    23

additional parties”). This court and other circuits have used
the same characterization. E.g., Vanegas v. Signet Builders, Inc.,

, 725 (7th Cir. 2024) (“Congress chose to ‘create a
system of permissive joinder rather than creating so-called
class actions.’” (internal quotation marks omitted)), quoting
Fischer v. Fed. Express Corp., 

, 379 (3d Cir. 2022);
Alvarez v. City of Chicago, 

 (7th Cir. 2010)
(whether FLSA plaintiffs are “permitted to pursue their
claims in one action or several is committed to the sound dis-
cretion of the district court, but misjoinder of parties is never
a ground for dismissing an action”), citing Fed. R. Civ. P. 21;
see also Waters v. Day & Zimmermann NPS, Inc., 

, 96
(1st Cir. 2022) (“The FLSA’s ‘similarly situated’ limitation for
collective actions displaces Rule 20 and limits the range of in-
dividuals who may be added as opt-in plaintiffs by requiring
that they be ‘similarly situated.’”); Scott v. Chipotle Mexican
Grill, Inc., 

, 520 (2d Cir. 2020) (“[T]he FLSA not
only imposes a lower bar than Rule 23, it imposes a bar lower
in some sense even than Rules 20 and 42, which set forth the
relatively loose requirements for permissive joinder and con-
solidation at trial.”), quoting Campbell v. City of Los Angeles,

 (9th Cir. 2018); Campbell, 903 F.3d at 1104–
05 (“The natural parallel [with opt-in plaintiffs] is to plaintiffs
initially named or later added under the ordinary rules of
party joinder.”).
   If a district court decides to exercise its authority under
Hoffman-La Roche to send notice to potential plaintiffs, those
plaintiffs may opt into the case by filing notices with the court.
Merely filing those notices makes them parties to the case. See

(b) (employee becomes “party plaintiff” when
he files written consent “in the court in which [the] action is
brought”). That becomes the status quo, and it would
24                                                  No. 24-2574

ordinarily require a motion (by the defendant) to change that
status, though the court could also act on its own initiative.
The remedy, as the panel notes, should be severance—allow-
ing the actions to continue, albeit individually or perhaps in
smaller sub-groups.
    Further, being similarly situated is not an element in a
plaintiff’s case that must be proven by a preponderance of the
evidence. It is only an issue of joinder or consolidation, which
are matters of case management. That’s consistent with our
and other courts’ decisions holding that FLSA and ADEA col-
lective actions are not “representative” actions. E.g., Vanegas,
113 F.4th at 723 (collective actions are just amalgamations of
parties, requiring personal jurisdiction over each plaintiff);
Canaday v. Anthem Cos., 

, 402 (6th Cir. 2021) (“A Rule
23 class action is representative, while a collective action un-
der the FLSA is not.”); Campbell, 

 (“A collective
action, on the other hand, is not a comparable form of repre-
sentative action. Just the opposite: Congress added the
FLSA’s opt-in requirement with the express purpose of
‘bann[ing]’ such actions under the FLSA. … A collective ac-
tion is more accurately described as a kind of mass action, in
which aggrieved workers act as a collective of individual
plaintiffs with individual cases—capitalizing on efficiencies
of scale, but without necessarily permitting a specific, named
representative to control the litigation, except as the workers
may separately so agree.”). It’s also consistent with the Su-
preme Court’s emphasis in Hoffmann-La Roche that the collec-
tive action mechanism is designed to benefit the judicial sys-
tem through “efficient resolution in one proceeding of com-
mon issues of law and fact arising from the same alleged dis-
criminatory activity.” 

.
No. 24-2574                                                  25

    We review Rule 20 decisions granting or denying permis-
sive joinder with a deferential standard, abuse of discretion.
E.g., UWM Student Ass’n v. Lovell, 

, 863–64 (7th
Cir. 2018). To my knowledge this court has never said one side
or the other bears a burden of going forward or persuading
on those questions. We have left those matters to the discre-
tion of the district court to choose a process and to answer the
case-management question. Those decisions will most often
be based on factors that have more to do with efficiency, fair-
ness, and expense than with disputed issues of material fact.
See 7 Wright, Miller & Kane, Federal Practice & Procedure
§ 1653 at 435 (2019) (“language in a number of decisions sug-
gests that the courts are inclined to find that claims arise out
of the same transaction or occurrence when the likelihood of
overlapping proof and duplication in testimony indicates that
separate trials would result in delay, inconvenience, and
added expense to the parties and to the court”); Eclipse Mfg.
Co. v. M and M Rental Center, Inc., 

(N.D. Ill. 2007); see also 9A Wright, Miller & Kane § 2383, 26
(“consolidation of separate actions presenting a common is-
sue of law or fact is permitted under Federal Rule 42 as a mat-
ter of convenience and economy in judicial administration”).
That should be the case for § 216(b) actions too. Whether the
claims proceed to trial on a collective basis should depend on
a case-specific determination of whether plaintiffs are suffi-
ciently similarly such that their claims could be more effi-
ciently resolved together.
   Accordingly, with respect for my colleagues’ different
views, I would leave open the burden of moving forward and
the burden of persuasion and save those issues for an FLSA
or ADEA appeal where the answers would make a difference.