Toolkit · Litigation

The Class Action Toolkit

9 U.S.C. §§ 2

Type
Toolkit
Jurisdiction
National
Date
2026-08-04
Topic
Litigation

This toolkit is UpLaw's front door to class action practice. It maps the entire lifecycle — the strategic decision to bring or defend, threshold defenses including arbitration and class waivers and Article III standing, CAFA jurisdiction and removal, pleading, certification under Rule 23(a) and 23(b), the predominance and damages-model fight, expert evidence, Rule 23(f) interlocutory review, notice and opt-outs, Rule 23(e) settlement approval, fees and objectors, trial of a certified class, and the substantive areas where class actions concentrate. Each stage gives you plain-English doctrine, the controlling authority, the traps that decide cases, and a link into UpLaw's deeper guide or checklist. A running example follows one fictional consumer class action from demand letter through final approval. Use it as a research map: skim the lifecycle, jump to the stage you need, and follow the links down into the detail.

The Class Action Toolkit

Litigation > Class Actions | Published 4 August 2026 - Updated 1 September 2026 | UpLaw - uplaw.us

A class action is a procedural device that turns a $40 injury into a $40 million case. Everything that is hard about it — the leverage, the settlement pressure, the ferocity of the certification fight, the judge's unusual duty to protect people who are not in the room — flows from that one fact. This toolkit is the map.

Keywords: class action, Rule 23, class certification, predominance, commonality, CAFA, removal, class waiver, arbitration, TransUnion standing, Comcast, Dukes, Rule 23(f), class notice, opt-out, settlement approval, attorney's fees, objectors

How to read this toolkit

Class action practice is not one problem. It is a chain of twelve or so problems, each with its own doctrine, its own subsection of Federal Rule of Civil Procedure 23 or the Class Action Fairness Act, and its own way of killing a case. A defendant can win at any link in the chain — on an arbitration clause before the complaint is even answered, on Article III standing, on predominance, on the damages model, on the adequacy of the notice plan. A plaintiff has to win at every link. That asymmetry is the single most important structural fact in this area of law, and it explains almost everything about how these cases are actually litigated.

This toolkit walks the chain stage by stage. It is a research map, not a treatise — each stage gives you the plain-English doctrine, the anchor authority, the classic traps, and a link into UpLaw's deeper guide or checklist where the real detail lives. The doctrinal foundation beneath everything below is Class Action Litigation: A Comprehensive Guide to Rule 23; keep it open in the next tab.

Use this document three ways. Read it end to end once, to build the mental model of how a class case moves from demand letter to final approval. Jump to a stage when you have a live question — "can they compel arbitration?", "does Comcast sink my damages model?", "what does a judge look for under Rule 23(e)(2)?" — and follow the deep-dive links. Or use the consolidated gated checklist near the end as an audit of a live case from either side of the caption.

A running example keeps the doctrine concrete. Meet Northgate Nutrition, Inc., a fictional supplement maker whose flagship product, Cleanburn, is labeled "30 grams of complete protein per scoop." An independent lab reports that Cleanburn is nitrogen-spiked with cheap free-form amino acids and delivers roughly 21 grams of actual complete protein. A consumer named Dana Ruiz, who bought twelve tubs at $54.99 each, walks into a plaintiffs' firm. By the end of this toolkit, Ruiz's case will have touched every stage of class action practice.

The class-action lifecycle at a glance

  1. The strategic decision — Is this a class case at all? The settlement-pressure dynamic that drives everything. (Rule 23; Coopers & Lybrand v. Livesay)
  2. Threshold defenses — Arbitration and class waivers, Article III standing, Rule 12. (9 U.S.C. §§ 2–4; Concepcion; TransUnion)
  3. CAFA jurisdiction and removal — Who picks the forum, and how the defendant takes it. (28 U.S.C. §§ 1332(d), 1453)
  4. Pleading a class claim — The class definition, the Rule 23(c)(1)(B) elements, and Twombly/Iqbal.
  5. Certification: Rule 23(a) and 23(b) — Numerosity, commonality, typicality, adequacy; then the (b)(1), (b)(2), or (b)(3) gate. (Dukes)
  6. Predominance, damages models, and experts — Where most certifications are actually won and lost. (Comcast; Tyson Foods; Rule 702)
  7. Rule 23(f) interlocutory review — The 14-day window, and why it is the whole ballgame.
  8. Notice, opt-outs, and due process — Rule 23(c)(2)(B); Eisen; Shutts.
  9. Settlement and Rule 23(e) approval — Preliminary approval, the 2018 four-factor test, and the judge as fiduciary. (Amchem)
  10. Fees, incentive awards, and objectors — Rule 23(h); the common-fund doctrine; CAFA's coupon rules.
  11. Trying a certified class — The rare event, and why "Trial by Formula" is forbidden.
  12. Where class actions concentrate — Consumer, securities, antitrust, employment, data breach, products.

Stage 1 — The strategic decision and the settlement-pressure dynamic

Rule 23 exists because some wrongs are real but individually unlitigable. Dana Ruiz was overcharged, at most, a few hundred dollars. No rational lawyer takes that case alone. Aggregate 900,000 Cleanburn purchasers and the same claim is worth tens of millions — and is worth a defense budget to match.

That aggregation is not neutral. Because a certification order converts a nuisance claim into an existential one, the Supreme Court has long recognized that certification exerts "hydraulic" pressure to settle regardless of merit; in Coopers & Lybrand v. Livesay, 437 U.S. 463 (1978), the Court acknowledged the "death knell" reality even while refusing to make certification orders immediately appealable as of right — a gap Congress and the Rules Committee later filled with Rule 23(f). For the defendant, this means the certification hearing is often the trial. Somewhere between 90% and the overwhelming majority of certified class actions settle; very few are tried. Plan accordingly.

For the plaintiff, the strategic questions before filing are: Is there a common, classwide misrepresentation or practice (as opposed to individualized sales pitches)? Is there a classwide method of proving damages tied to the theory of liability? Is the class ascertainable from defendant's records or objective criteria? Are there arbitration clauses with class waivers in the purchase path? Is there a statute with fee-shifting or statutory damages that carries the case?

For the defendant, the mirror questions are: Can we compel arbitration? Can we knock out standing for most of the class? Can we make liability turn on individualized reliance, exposure, causation, or damages? Can we show that the named plaintiff is atypical or inadequate? And, critically: what is our exit price, and when do we pay it?

Trap. Both sides routinely under-plan the front end. Plaintiffs plead a class definition they cannot certify and then cannot escape without dismissing a named plaintiff. Defendants blow the arbitration argument by litigating for a year before raising it — waiver by conduct is real, and courts have found it where a party substantially invoked the litigation machinery. For the surrounding federal-procedure architecture that a class case sits inside, see The Civil Pretrial Toolkit: From Pleadings to Summary Judgment.

Stage 2 — Threshold defenses: arbitration and class waivers, standing, Rule 12

Arbitration and the class waiver

The single most consequential development in modern class action practice is not in Rule 23 at all. It is the Federal Arbitration Act, 9 U.S.C. §§ 1–16, as applied in AT&T Mobility LLC v. Concepcion, 563 U.S. 333 (2011), which held that the FAA preempts state-law rules conditioning enforceability of an arbitration agreement on the availability of classwide arbitration. A consumer or employee contract with a valid arbitration clause and a class-action waiver will, in the ordinary case, take the dispute out of court and out of the class device entirely.

Three follow-on decisions define the terrain. Stolt-Nielsen S.A. v. AnimalFeeds International Corp., 559 U.S. 662 (2010), held that a party may not be compelled to submit to class arbitration absent a contractual basis for concluding it agreed to do so. Lamps Plus, Inc. v. Varela, 139 S. Ct. 1407 (2019), held that an ambiguous agreement is not a sufficient basis — silence and ambiguity are treated alike. And Epic Systems Corp. v. Lewis, 138 S. Ct. 1612 (2018), held that employment arbitration agreements with individualized-proceeding requirements are enforceable notwithstanding the National Labor Relations Act.

The counter-moves are real but narrow: unconscionability under generally applicable state contract law (which survives Concepcion only if it does not single out arbitration), failure of formation (no assent to browsewrap terms), waiver by litigation conduct, non-signatory defenses, and — in California — the partial workaround for representative PAGA claims recognized in Viking River Cruises, Inc. v. Moriana, 596 U.S. 639 (2022). Note also the phenomenon of mass arbitration, in which plaintiffs' firms file thousands of individual demands and the per-case administrative fees become the leverage the class waiver was meant to eliminate. Drafting is where this fight is won or lost years in advance: see Drafting an Arbitration Clause: A Practical Checklist and Arbitration and ADR: A Toolkit for Choosing a Forum, Drafting Clauses, and Running a Case.

Northgate's clause is on the back of the tub in six-point type and there is no arbitration provision on the e-commerce checkout page. Counsel concludes formation is too weak to press. The case stays in court.

Article III standing — and TransUnion

Standing is now a first-order class action defense, not a formality. Spokeo, Inc. v. Robins, 578 U.S. 330 (2016), held that a bare procedural violation divorced from concrete harm does not confer standing. TransUnion LLC v. Ramirez, 594 U.S. 413 (2021), went further and held that every class member who seeks individual damages must have Article III standing — "No concrete harm, no standing" — and that a risk of future harm that never materialized is not, by itself, a concrete injury for a damages claim. In TransUnion itself, only the roughly 1,853 class members whose misleading credit files were actually disseminated to third parties had standing; the other 6,332 did not.

TransUnion reshaped class definitions. Defendants now argue that a class defined to sweep in uninjured members cannot be certified, or at least cannot recover; plaintiffs narrow definitions to injury-in-fact and litigate how many uninjured members are tolerable. The Ninth Circuit sitting en banc in Olean Wholesale Grocery Cooperative, Inc. v. Bumble Bee Foods, LLC, 31 F.4th 651 (9th Cir. 2022) (en banc), held that a class is not automatically barred by the presence of some uninjured members, so long as predominance is otherwise satisfied and the number is de minimis; other courts are less forgiving, and the D.C. and First Circuits have written pointedly the other way. This is a live, unsettled fault line, and the threshold percentage of uninjured members a court will tolerate is not fixed by any binding authority — check your circuit before you rely on a number.

Rule 12

Everything ordinary applies: failure to state a claim under Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009); preemption; personal jurisdiction (and the open question left by Bristol-Myers Squibb Co. v. Superior Court, 582 U.S. 255 (2017), of whether its due-process limits on mass-tort joinder extend to absent class members in a class action — the circuits are not aligned). Motions to strike class allegations at the pleading stage are permitted but rarely granted; courts usually want a certification record. See Answering a Complaint and Pleading Affirmative Defenses: A Practical Checklist.

Stage 3 — CAFA jurisdiction and removal

The Class Action Fairness Act of 2005, 28 U.S.C. §§ 1332(d), 1453, and 1711–1715, was Congress's response to state-court "magnet" jurisdictions. It grants federal district courts original jurisdiction over class actions where: (1) the proposed class has 100 or more members; (2) there is minimal diversity (any class member diverse from any defendant); and (3) the aggregated amount in controversy exceeds $5,000,000, exclusive of interest and costs. Section 1453 then removes the ordinary obstacles: any defendant may remove without the consent of the others, and the forum-defendant rule does not apply.

CAFA has exceptions that plaintiffs mine hard. The local controversy exception (§ 1332(d)(4)(A)) and the home-state exception (§ 1332(d)(4)(B)) require remand where the class is predominantly in-state and a significant in-state defendant is present; the discretionary exception (§ 1332(d)(3)) permits remand where between one-third and two-thirds of the class and the primary defendants are in-state. CAFA also reaches mass actions — 100 or more plaintiffs proposing a joint trial (§ 1332(d)(11)) — a provision the Supreme Court read narrowly in Mississippi ex rel. Hood v. AU Optronics Corp., 571 U.S. 161 (2014), holding that a state's parens patriae suit is not a "mass action" merely because it is brought on behalf of many unnamed citizens.

Two procedural rules matter enormously. Under Dart Cherokee Basin Operating Co. v. Owens, 574 U.S. 81 (2014), a notice of removal need only include a plausible allegation that the amount in controversy exceeds the threshold; evidence is required only if the plaintiff or court contests it. And under Standard Fire Insurance Co. v. Knowles, 568 U.S. 588 (2013), a named plaintiff cannot defeat CAFA by stipulating to damages below $5 million, because he cannot bind absent class members before certification. Finally, Home Depot U.S.A., Inc. v. Jackson, 139 S. Ct. 1743 (2019), held that a third-party counterclaim defendant is not a "defendant" who may remove under § 1441 or § 1453.

Section 1453(c) also gives the courts of appeals discretionary review of a remand order if applied for within 10 days — a rare and valuable appellate hook.

Northgate is a Delaware corporation headquartered in Illinois; Ruiz sues in California state court on behalf of a nationwide class. Nine hundred thousand tubs at a claimed price premium of roughly $12 puts the amount in controversy far above $5 million, minimal diversity is obvious, and no CAFA exception fits a nationwide class. Northgate removes within 30 days of service.

Stage 4 — Pleading a class claim

The class complaint has to do two jobs at once: state a claim, and lay the groundwork for certification. Rule 23(c)(1)(B) requires the certification order to define the class, the class claims, issues, or defenses, and appoint class counsel — so the complaint should already be drafted toward that order.

Draft the class definition with obsessive care. It must be (1) precise — using objective criteria a court can apply without a merits inquiry; (2) administratively workable — no "all persons who were deceived," which is a fail-safe class that defines membership by the merits and is impermissible in most circuits; and (3) matched to injury after TransUnion. Consider subclasses at the outset (state-law subclasses are the standard fix for a nationwide consumer class governed by fifty different UDAP statutes).

Ruiz's operative definition, after two rounds of amendment: All persons who purchased Cleanburn protein powder in the United States for personal use, and not for resale, between 1 March 2023 and 30 June 2026. Note what it does not say: nothing about reliance, nothing about deception, nothing that requires a mini-trial to determine membership. For the mechanics of the pleading itself, see Drafting a Federal Civil Complaint: A Practical Checklist.

Trap. Many state consumer statutes carry pre-suit notice conditions — California's Consumer Legal Remedies Act, for instance, requires a 30-day demand before damages may be sought. Blowing that notice is a self-inflicted, unfixable wound on the damages claim. The substantive statutes that carry most consumer class actions are mapped in Consumer Protection Law: A Comprehensive Guide to the FDCPA, FCRA, TCPA, and State UDAP Statutes.

Stage 5 — Certification: Rule 23(a) and 23(b)

Certification is a two-step gate. The movant must satisfy all four Rule 23(a) prerequisites and then fit within one of the three Rule 23(b) categories.

Rule 23(a): the four prerequisites.

Prerequisite What it requires Where it breaks
Numerosity — 23(a)(1) Joinder is impracticable Rarely; 40+ is a common rule of thumb, not a rule
Commonality — 23(a)(2) At least one common question capable of classwide resolution Dukes: the answer must be common, not just the question
Typicality — 23(a)(3) Named plaintiff's claims arise from the same course of conduct Unique defenses; plaintiff never saw the challenged representation
Adequacy — 23(a)(4) No conflicts; counsel is competent Intra-class conflicts (Amchem); professional plaintiffs; counsel's own conflicts

Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011), is the hinge. It did two things. First, it raised commonality from a formality into a real requirement: class members' claims must "depend upon a common contention" whose "truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke." A policy of delegating discretion to local managers is not a common answer. Second, Dukes insisted that a court must conduct a "rigorous analysis" that will "frequently entail some overlap with the merits" — killing the old idea that certification is a purely procedural inquiry.

Rule 23(b): the three doors.

  • 23(b)(1) — separate actions would risk inconsistent standards for the defendant (b)(1)(A) or impair absent members' interests (b)(1)(B), the "limited fund" theory. Mandatory (no opt-outs). Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999), narrowed limited-fund classes almost to the vanishing point: the fund must be independently established, its inadequacy demonstrated, and it must be devoted entirely to the claims.
  • 23(b)(2) — the defendant "acted or refused to act on grounds that apply generally to the class," so that injunctive or declaratory relief is appropriate for the class as a whole. Mandatory, no notice required by the rule. Dukes held that (b)(2) does not permit claims for individualized monetary relief — backpay is not "incidental" — because that would deny absent members notice and opt-out rights.
  • 23(b)(3) — predominance and superiority. The damages class. This is where the war is fought, and it is Stage 6.

The full doctrinal treatment, with the circuit splits, lives in Class Action Litigation: A Comprehensive Guide to Rule 23. For the movant's mechanics, see Moving for Class Certification: A Practical Checklist; for the mirror image, Opposing Class Certification: A Defense Checklist.

Ascertainability is a fifth, unwritten requirement in some circuits. The Third Circuit has required a showing of "administrative feasibility" — a reliable, administratively workable way to identify class members. The Seventh, Ninth, Sixth, and others have rejected a heightened ascertainability standard as unmoored from Rule 23's text. In a consumer case with no receipts and no loyalty-card data, this split can be dispositive. Ruiz's team knows this and pleads into a circuit that has rejected the heightened test, while building a claims-administration protocol with affidavits and retailer purchase data anyway.

Stage 6 — Predominance, damages models, and expert evidence

Rule 23(b)(3) requires that common questions predominate over individual ones and that a class action be superior to other methods. Predominance is not a headcount of questions; it is a qualitative judgment about whether the case can be tried on common proof.

Comcast Corp. v. Behrend, 569 U.S. 27 (2013), is the defense's best weapon here. The Court decertified an antitrust class because the plaintiffs' regression model measured damages from four theories of antitrust impact when only one had been accepted for class treatment. The holding: a damages model must be consistent with the plaintiff's theory of liability, and a court must consider whether the model can measure damages on a classwide basis, even if that inquiry overlaps the merits. Comcast is often overread — it does not require damages to be identical across the class, and it does not forbid individualized damages calculations after a classwide liability finding. But it does require that the methodology be classwide and tethered to the certified theory.

Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442 (2016), is the plaintiff's counterweight. The Court permitted the use of representative statistical evidence (a time-study of donning-and-doffing) to establish classwide liability, reasoning that such evidence is permissible in a class action if each class member could have relied on it in an individual suit. Tyson Foods is the cleanest statement that class actions do not lower the evidentiary bar — but they do not raise it either.

The practical consequence: certification is now an expert fight. The plaintiff's economist offers a conjoint analysis, hedonic regression, or benchmark model to compute the classwide price premium; the defense economist attacks its assumptions, its fit to the liability theory, and its treatment of uninjured members. Courts in most circuits apply a full Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), analysis to expert testimony offered at certification — and Rule 702, as amended effective 1 December 2023, now makes explicit that the proponent must show by a preponderance that each requirement is met and that the opinion "reflects a reliable application" of the methodology to the facts. Work the motion with Challenging an Expert Under Daubert and Rule 702: A Practical Checklist and, for the evidentiary architecture generally, The Federal Evidence Toolkit.

The record that feeds all of this is built in discovery — merits and certification discovery are rarely cleanly separable, and the fight over bifurcation is itself strategic. See The Federal Discovery Toolkit: A Roadmap From Rule 26(f) to Sanctions. And because the damages model is the case, read it alongside The Litigation Damages Toolkit: Valuing and Proving Money Awards.

Ruiz's expert runs a conjoint survey and concludes the "30g protein" claim commanded a $9.40 per-tub premium — uniform across the class because every tub bore the same label. Northgate's expert responds that purchasers who bought for taste, price, or brand loyalty suffered no premium at all, and that retailer discounting means many class members paid less than the shelf price. The court certifies, finding the label was uniform, the theory was classwide, and the model matched the theory — with damages individualization reserved for claims administration.

Stage 7 — Rule 23(f) interlocutory review

Rule 23(f) permits a court of appeals, in its discretion, to grant permission to appeal an order granting or denying class certification if a petition is filed within 14 days after the order is entered. That deadline is short, unforgiving, and — as the Supreme Court held in Nutraceutical Corp. v. Lambert, 586 U.S. 188 (2019) — not subject to equitable tolling, because it is a claim-processing rule that the appellate rules expressly make non-extendable.

Rule 23(f) exists because of Coopers & Lybrand v. Livesay: without it, a certification order was effectively unreviewable, since the defendant would settle rather than risk trial and the plaintiff whose class was denied had no final judgment to appeal. And it is why Microsoft Corp. v. Baker, 582 U.S. 23 (2017), matters: plaintiffs denied certification may not manufacture appellate jurisdiction by voluntarily dismissing their individual claims with prejudice. Rule 23(f) is the road, and it is the only one.

Petitions are granted sparingly. The recurring grounds: the order is a "death knell" for one side; it turns on an unsettled legal question of general importance; or it is manifestly erroneous. Draft the petition as if it were the merits brief, because for practical purposes it is. For everything that happens on appeal thereafter, see Federal Civil Appeals: A Comprehensive Guide.

Trap. Calendar the 14 days from the entry of the certification order, not from receipt, and do not assume a motion for reconsideration restarts the clock. Some circuits treat a timely reconsideration motion as tolling; others do not. File protectively.

Stage 8 — Notice, opt-outs, and due process

For a certified (b)(3) class, Rule 23(c)(2)(B) requires "the best notice that is practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort." The 2018 amendments expressly authorize notice by "United States mail, electronic means, or other appropriate means" — email and targeted digital campaigns are now mainstream, not exotic.

Two due-process pillars. Eisen v. Carlisle & Jacquelin, 417 U.S. 156 (1974), held that individual notice is mandatory where members are identifiable and that the representative plaintiff, not the defendant, bears the cost — a holding that has ended more than one class action on economics alone. Phillips Petroleum Co. v. Shutts, 472 U.S. 797 (1985), held that due process requires absent class members in a damages class to receive notice and an opportunity to opt out, and that a forum state may not apply its own law to claims with no significant contacts to that state (the seed of every choice-of-law fight in a nationwide consumer class).

The notice must be written in plain, easily understood language (Rule 23(c)(2)(B)) and must state: the nature of the action, the class definition, the claims and defenses, that a member may enter an appearance through counsel, that the court will exclude members who request exclusion, the time and manner for opting out, and the binding effect of a class judgment.

Trap. Under-inclusive notice plans are a leading cause of reversal on appeal from a settlement approval. If the defendant has email addresses, use them; if it has only geography, buy media. And note the asymmetry: (b)(1) and (b)(2) classes are mandatory — Rule 23 does not require notice or opt-out rights, which is precisely why Dukes refused to let monetary claims ride into a (b)(2) class.

Stage 9 — Settlement and Rule 23(e) approval

Nearly every certified class action ends here. Rule 23(e) makes the district judge a fiduciary for absent class members, because the ordinary adversarial check is gone: at settlement, plaintiffs' counsel and the defendant both want the deal approved.

The modern sequence:

  1. Rule 23(e)(1) — the parties must give the court information sufficient to determine whether it will likely be able to approve the settlement and certify the class. This is "preliminary approval," and the 2018 amendments deliberately reframed it as a likelihood determination, not a rubber stamp.
  2. Notice to the class of the proposed settlement.
  3. CAFA notice — 28 U.S.C. § 1715 requires the defendant to serve the settlement papers on the U.S. Attorney General and the appropriate state officials within 10 days of filing, and no final order may issue until 90 days after that service. Miss it and class members may refuse to be bound.
  4. Fairness hearing and Rule 23(e)(2) final approval.

Rule 23(e)(2), as amended in 2018, sets four factors. The court may approve only after finding the settlement is "fair, reasonable, and adequate" after considering whether:

  • (A) the class representatives and class counsel have adequately represented the class;
  • (B) the proposal was negotiated at arm's length;
  • (C) the relief for the class is adequate, taking into account the costs, risks, and delay of trial and appeal; the effectiveness of the proposed claims process; the terms of any proposed attorney's fee award, including timing of payment; and any Rule 23(e)(3) side agreement; and
  • (D) the proposal treats class members equitably relative to each other.

Settlement classes get special scrutiny. Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997), held that a court may consider settlement in deciding certification (manageability drops out — there will be no trial), but the other Rule 23 requirements still apply with full force, and indeed adequacy and predominance demand more attention, not less, because the settlement-only posture removes the trial-court's ordinary safeguards. The Amchem class failed on both: futures claimants and presently-injured claimants had irreconcilable interests.

Watch for the structural red flags courts and objectors hunt: clear-sailing clauses (defendant agrees not to contest the fee request), kicker/reverter clauses (unawarded fees revert to the defendant rather than the class), coupon relief (see Stage 10), cy pres-only distributions — the Supreme Court in Frank v. Gaos, 139 S. Ct. 1041 (2019), took the cy pres question and disposed of the case on standing without resolving it, so the doctrine remains contested — and claims processes so onerous the take-rate collapses. Work the deal with Settling a Class Action: A Rule 23(e) Approval Checklist.

Northgate settles: a $28,000,000 non-reversionary common fund, $12 per tub with proof of purchase and up to 3 tubs without, plus injunctive relief reformulating the label. CAFA notice goes out. The claims administrator runs email, retailer-list, and social media notice reaching an estimated 78% of the class.

Stage 10 — Fees, incentive awards, and objectors

Fees. Rule 23(h) governs. In a common fund case, the equitable doctrine recognized in Boeing Co. v. Van Gemert, 444 U.S. 472 (1980), lets counsel recover a percentage of the fund from the beneficiaries who share it; the alternative is a lodestar (reasonable hours × reasonable rate, with a possible multiplier), which is the standard method in fee-shifting statutory cases. Percentage awards commonly cluster in the 20–33% range in the federal courts, with a "cross-check" against lodestar — but this is a judicial convention subject to circuit variation and to the size of the fund (mega-fund settlements typically draw declining percentages). Do not represent any of these figures as a fixed rule.

Coupon settlements. CAFA, 28 U.S.C. § 1712, specifically targets them: where a settlement provides coupons, the portion of the attorney's fee attributable to the coupon relief must be based on the value of coupons actually redeemed, not the face value issued. This provision alone has largely ended pure-coupon consumer settlements.

Incentive (service) awards to named plaintiffs are routine in most circuits — a few thousand dollars for the time and exposure of serving. But the Eleventh Circuit, in Johnson v. NPAS Solutions, LLC, 975 F.3d 1244 (11th Cir. 2020), held that incentive awards are flatly prohibited by the nineteenth-century Supreme Court decisions in Trustees v. Greenough, 105 U.S. 527 (1882), and Central Railroad & Banking Co. v. Pettus, 113 U.S. 116 (1885), which permit reimbursement of expenses but not a bounty for serving as a plaintiff. Other circuits have declined to follow. This is a genuine, unresolved circuit split — do not assume an incentive award is available; check your circuit before building one into a settlement structure.

Objectors. Rule 23(e)(5) allows any class member to object; the 2018 amendments require the objection to state whether it applies to the objector only, a subset, or the whole class, and — critically — require court approval for any payment in connection with withdrawing an objection or an appeal, a direct response to serial "objector blackmail." Under Devlin v. Scardelletti, 536 U.S. 1 (2002), an objecting class member who is not a named party may appeal the approval of a settlement without first intervening. A good-faith objector is a gift to the court; a professional one is a tax on the class.

Stage 11 — Trying a certified class

Class trials are rare but not mythical. The single most important constraint comes from Dukes: the Court rejected what it called "Trial by Formula" — the plan to try a sample of claims, calculate the percentage of valid ones, and extrapolate to the class — because it would deprive the defendant of its right to litigate individual defenses, in violation of the Rules Enabling Act, 28 U.S.C. § 2072(b), which forbids interpreting a procedural rule to "abridge, enlarge or modify any substantive right."

Tyson Foods clarifies the boundary: representative evidence is fine where it would be admissible and probative in an individual case; it is not fine where it substitutes for proof the defendant is entitled to contest person by person. The practical middle ground is bifurcation — try classwide liability and a classwide damages methodology to the jury, then resolve individual entitlement and amount through a claims administration process or Rule 23(c)(4) issue certification. Rule 23(c)(4) ("issues" classes) is itself the subject of a circuit split over whether predominance must be satisfied as to the whole case or only the certified issue.

Preparation is ordinary federal trial practice, at scale: see Federal Civil Trial Practice: From Final Pretrial Conference to Verdict. And because summary judgment often follows certification (or precedes it — a defendant may sometimes prefer to win on the merits before a class is certified, though a pre-certification merits win binds only the named plaintiff under the one-way-intervention rule), keep Moving for Summary Judgment Under Rule 56: A Practical Checklist in the file.

Stage 12 — Where class actions concentrate

Rule 23 is a procedural device; its content comes from the substantive law it carries. Six areas generate the overwhelming majority of federal class filings.

A word of caution before you read this section. Rule 23, CAFA, the FAA, the PSLRA, and the federal antitrust statutes are federal law and are uniform across the country. The substantive law they carry very often is not. Consumer protection (state UDAP statutes), tort, products liability, and insurance coverage are overwhelmingly matters of state law, and the states genuinely diverge — on reliance and causation requirements, on whether a private right of action exists at all, on comparative versus contributory fault, on whether design defect is tested by consumer expectations or risk-utility, and on what a policy covers. Nothing in this section states a nationwide rule of substantive liability, because for these areas there is none. That divergence is not a footnote in class practice; it is frequently the whole certification fight, because fifty bodies of law applied to one nationwide class is the most common way predominance dies. Confirm the governing state's rule — and the choice-of-law analysis that selects it — before relying on anything below.

Consumer. False labeling, state UDAP statutes, and the federal FDCPA, FCRA, and TCPA. The certification fight is almost always about a uniform representation (good for plaintiffs), individualized reliance and materiality (good for defendants), and a classwide price-premium model. The UDAP statutes are state law and are not interchangeable: some require reliance, some require scienter, some cap or multiply damages, some have pre-suit notice conditions, and a few bar class treatment outright. A nationwide consumer class therefore lives or dies on Shutts choice of law and on whether state-law subclasses can be made manageable. Start at Consumer Protection Law: A Comprehensive Guide to the FDCPA, FCRA, TCPA, and State UDAP Statutes and The Consumer Protection and Credit Toolkit.

Securities. The most structurally distinctive class action there is: the Private Securities Litigation Reform Act, 15 U.S.C. § 78u-4, imposes a lead-plaintiff selection process, a heightened pleading standard, and an automatic discovery stay. Certification turns on the fraud-on-the-market presumption of Basic Inc. v. Levinson, 485 U.S. 224 (1988), which supplies classwide reliance; Halliburton Co. v. Erica P. John Fund, Inc. (Halliburton II), 573 U.S. 258 (2014), preserved the presumption but held that a defendant may rebut it at certification with evidence of no price impact; Amgen Inc. v. Connecticut Retirement Plans & Trust Funds, 568 U.S. 455 (2013), held that materiality need not be proved at certification; and Goldman Sachs Group, Inc. v. Arkansas Teacher Retirement System, 594 U.S. 255 (2021), held that the generic nature of an alleged misrepresentation is relevant evidence at the price-impact stage. See Securities Fraud Litigation: A Comprehensive Guide to Rule 10b-5 Private Actions and The Securities Litigation and Enforcement Toolkit.

Antitrust. Section 4 of the Clayton Act, 15 U.S.C. § 15, gives treble damages and fees; Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977), confines federal damages to direct purchasers (with many states permitting indirect-purchaser suits under "Illinois Brick repealer" statutes — hence the parallel direct and indirect class structures you see in every cartel case). Predominance turns on classwide proof of antitrust impact across a heterogeneous purchaser class. See Antitrust Law: A Comprehensive Guide to the Sherman, Clayton, and FTC Acts and The Antitrust and Competition Toolkit.

Employment. Note the fork: Title VII and § 1981 discrimination claims proceed under Rule 23 (and Dukes made them hard), while FLSA wage-and-hour claims proceed as opt-in collective actions under 29 U.S.C. § 216(b) — a different statute, a different standard, and a different (much lower) participation rate. Hybrid state-law Rule 23 classes alongside FLSA collectives are common and contested. See Employment Discrimination Law: A Comprehensive Guide and Wage and Hour Law Under the FLSA: A Comprehensive Guide.

Data breach and privacy. The fastest-growing category and the one most transformed by TransUnion: the central question is whether increased risk of identity theft, time spent mitigating, or loss of "benefit of the bargain" is a concrete Article III injury, and the circuits differ sharply. Statutory-damages regimes (Illinois's BIPA, wiretap statutes) sidestep the problem by supplying a concrete privacy interest. See Data Breach Notification Law: A Comprehensive Guide and The Privacy and Data Protection Toolkit: Building and Operating a Compliance Program.

Products liability and mass torts. Personal-injury classes almost never certify — causation and damages are irreducibly individual, which is why these cases run through multidistrict litigation under 28 U.S.C. § 1407 rather than Rule 23. Economic-loss and medical-monitoring classes are the exceptions worth studying. But note why they so rarely certify, because the reason is the state-law point above: tort and products liability are state law, and the elements themselves differ by state. Whether a design-defect claim is tested by consumer expectations, by risk-utility, or by some hybrid; whether the state has adopted the Restatement (Second) of Torts § 402A or the Restatement (Third) of Torts: Products Liability; whether contributory negligence bars recovery or comparative fault merely reduces it; whether medical monitoring is a recognized cause of action without present physical injury — every one of these varies, and a nationwide products class must reckon with all of them at once. There is no national law of products liability to certify. See Products Liability: A Comprehensive Guide to Design Defect, Manufacturing Defect, and Failure to Warn and The Tort and Products Liability Toolkit.

Defendants should also read the coverage side early — and here too the answer is state-specific. Whether a CGL policy's "damages because of bodily injury" or advertising-injury coverage reaches a class claim, whether defense costs erode limits, and how the duty to defend is triggered are questions of state insurance law and of the actual policy wording; they are not settled by any uniform national rule. Tender early, reserve rights carefully, and read the policy. See The Business Insurance and Coverage Toolkit.

The lifecycle as a working checklist

Run this gate by gate. If you cannot clear a gate, do not spend money on the next one.

Gate 1 — Case selection

  • [ ] Is there a uniform, classwide practice or representation?
  • [ ] Is there a classwide method of proving injury and damages tied to the liability theory (Comcast)?
  • [ ] Are there arbitration clauses with class waivers in the transaction path (Concepcion)?
  • [ ] Does every class member have a concrete injury (TransUnion)?
  • [ ] Any pre-suit notice conditions (state UDAP; CLRA) satisfied?

Gate 2 — Forum and jurisdiction

  • [ ] CAFA elements checked: 100+ members, minimal diversity, >$5,000,000 aggregate (28 U.S.C. § 1332(d))?
  • [ ] Any CAFA exception (local controversy, home state, discretionary)?
  • [ ] Notice of removal filed within 30 days, plausibly alleging amount in controversy (Dart Cherokee)?
  • [ ] § 1453(c) petition calendared (10 days) if remanded?

Gate 3 — Pleading

  • [ ] Class definition objective, ascertainable, not fail-safe, matched to injury?
  • [ ] Subclasses for divergent state law?
  • [ ] Claims survive Twombly/Iqbal?
  • [ ] Rule 23(g) class counsel showing assembled?

Gate 4 — Certification

  • [ ] Numerosity, commonality (Dukes), typicality, adequacy (Amchem) — each with record evidence, not argument?
  • [ ] Correct 23(b) door chosen; no individualized money damages smuggled into a (b)(2) class?
  • [ ] Predominance supported by a classwide damages model that matches the theory?
  • [ ] Expert opinions Rule 702-proof; Daubert motions filed?
  • [ ] Ascertainability standard in this circuit identified and satisfied?

Gate 5 — Interlocutory review

  • [ ] Rule 23(f) petition calendared: 14 days from entry, not extendable?
  • [ ] Ground identified: death knell, unsettled question, or manifest error?

Gate 6 — Notice

  • [ ] Best notice practicable; individual notice to identifiable members (Eisen)?
  • [ ] Plain language; all seven Rule 23(c)(2)(B) contents included?
  • [ ] Opt-out mechanism and deadline stated (Shutts)?

Gate 7 — Settlement

  • [ ] Rule 23(e)(1) preliminary-approval record adequate for a likelihood finding?
  • [ ] CAFA § 1715 notice served; 90-day clock run before final order?
  • [ ] Rule 23(e)(2)(A)–(D) addressed with evidence?
  • [ ] Structural red flags cleared: clear-sailing, reverter, coupons, cy pres, take-rate?
  • [ ] Fee request supported (percentage + lodestar cross-check); incentive award lawful in this circuit?
  • [ ] Rule 23(e)(5) objections and any withdrawal payments disclosed and court-approved?

Primary Authority

Rules

Statutes

Cases

A note on currency and variation — read this before you rely on anything above.

Unsettled or circuit-dependent within Rule 23 itself. Several rules above are genuinely contested: heightened ascertainability; the tolerable number of uninjured class members in a certified class (no binding authority fixes a percentage); the availability of incentive awards (an open split following the Eleventh Circuit's Johnson v. NPAS Solutions); whether Bristol-Myers Squibb limits personal jurisdiction over absent class members' claims; whether predominance under Rule 23(c)(4) issue classes is measured against the whole case or the certified issue; and the vitality of cy pres distributions after Frank v. Gaos, which the Supreme Court took and then disposed of on standing without reaching the merits. Fee-percentage conventions and settlement-approval practice also vary meaningfully by circuit and by judge; the 20–33% band described in Stage 10 is an observed convention, not a rule of law.

Federal versus state. Rule 23, CAFA, the FAA, the PSLRA, and the Sherman and Clayton Acts are federal and uniform. The substantive law most class actions carry is not. Tort, products liability, insurance coverage, land use, and state UDAP statutes are state law and diverge materially from jurisdiction to jurisdiction. This toolkit deliberately states no nationwide rule in those areas, because none exists. Identify the governing state law — and the choice-of-law rule that selects it — before you evaluate any claim or defense.

Figures that move. Any dollar threshold, fee convention, filing deadline, or regulatory list referenced in passing here should be confirmed against the current statute, rule, or agency source at the time you use it. Amounts and thresholds in adjacent regulatory regimes — Hart-Scott-Rodino notification thresholds, tariff and duty rates, and OFAC and export-control designations, all of which are revised frequently and some of which change without notice — are especially volatile and must be verified before you rely on them; do not treat any figure recited in a secondary source, including this one, as current.

Secondary Sources

  • Newberg and Rubenstein on Class Actions — the standard treatise; the first place to look on any Rule 23 question.
  • Wright & Miller, Federal Practice and Procedure §§ 1751–1804 (class actions).
  • Manual for Complex Litigation (Fourth) — the Federal Judicial Center's practical guide to managing certification, notice, and settlement; widely relied on by district judges.
  • Advisory Committee Notes to the 2003 and 2018 amendments to Rule 23 — indispensable for the settlement-approval and notice provisions.

Related UpLaw Resources

Related Guides & Articles

Related Checklists

Related Toolkits

This toolkit is general legal information from UpLaw (uplaw.us), not legal advice, and does not create an attorney-client relationship. Laws vary by state and circuit and change over time; class action doctrine in particular is unsettled on several of the points flagged above. Verify current requirements and consult qualified counsel before acting.