The Four Threshold Requirements
No class exists until four prerequisites are established. They are short to state and heavily litigated, because each one is proved with evidence rather than pleaded, and a court must be satisfied of all four before it reaches the question of which kind of class is sought.

The rule in short
Rule 23(a) permits representative litigation only where the class is so numerous that joinder is impracticable, questions of law or fact are common to the class, the representatives' claims or defenses are typical, and the representatives will fairly and adequately protect the interests of the class. A court must conduct a rigorous analysis, which frequently overlaps with the merits, and the party seeking certification bears the burden on each element.
Every class action begins with the same four questions, and none of them concerns the merits of the claim. The rule permits one or more members of a class to sue or be sued as representative parties only if the class is so numerous that joinder is impracticable, questions of law or fact are common to the class, the representatives' claims or defenses are typical, and the representatives will fairly and adequately protect the interests of the class.
Numerosity as a practical judgment
The text asks about impracticability of joinder rather than about a number, and that framing controls the analysis. A group of several hundred geographically dispersed purchasers of a small consumer product presents a stronger case than a group of the same size composed of institutional counterparties in a single city, because the practical burden of bringing the latter before the court is manageable. Courts also weigh whether members can be identified, whether individual claims are large enough to be worth pursuing alone, and whether members are known to each other.
The requirement is rarely dispositive on its own but it is not a formality. Where a class is defined so narrowly that membership cannot be estimated, or where the proponent offers nothing beyond an assertion that the group is large, the element fails for want of proof. Reasonable inference from records is permitted; speculation is not, and courts have declined certification where the size of the group rested on assumption rather than data.
Commonality and the common answer
The text requires questions of law or fact common to the class, and modern practice reads that phrase as demanding more than a shared grievance. The question must be capable of classwide resolution, meaning its determination will resolve an issue central to the validity of each claim in one stroke. What matters is not the raising of common questions but the capacity of the proceeding to generate common answers apt to drive the resolution of the litigation.
The practical consequence is that a claim depending on many individualized decisions rarely produces a common answer, however uniform the legal theory sounds. A challenge to a written policy applied identically to everyone generates one; an allegation that many supervisors exercised discretion badly, without a common mode of exercising it, usually does not. Framing therefore matters, and a proponent who defines the common question at a high level of generality invites the objection that the answer will resolve nothing.
Commonality asks whether a common question exists that can be answered classwide. Predominance asks whether such questions outweigh the individual ones. A case can satisfy the first and fail the second, which is why a class certified for injunctive relief may be refused for damages on the same facts. The distinction is examined in the treatment of predominance, and conflating the two is the most common error in briefing certification.
Typicality and the representative's position
Typicality compares the representative's claim with the claims of the class and asks whether they arise from the same course of conduct and rest on the same legal theory. Perfect identity is not required; differences in the amount of damages, or in the precise circumstances of each transaction, do not defeat it. What defeats it is a difference in kind, particularly where the representative's claim depends on facts peculiar to that person.
A related problem arises where the representative's own transaction differs structurally from most of the class. A named plaintiff who bought through a channel that carried different terms, or who was covered by a superseding agreement, may share the legal theory while occupying a different factual position on the element that decides the case. Courts sometimes cure this by substituting a representative or by adding one, which is why proponents commonly plead more than one named plaintiff at the outset.
Defenses are where the element usually turns. A representative subject to a unique defense, such as an individually negotiated release, an arbitration agreement that others did not sign, or a limitation problem that does not affect the class, will spend the litigation defending a fight that is not the class's fight. Courts treat that risk as a typicality failure and sometimes as an adequacy failure, since the two elements overlap at exactly this point.
| Prerequisite | Question asked | Usual proof | Common failure |
|---|---|---|---|
| Numerosity | Is joinder impracticable | Records showing group size and dispersion | Size asserted rather than evidenced |
| Commonality | Can a question be answered classwide | The uniform policy, contract or practice | Discretionary decisions with no common mode |
| Typicality | Does the representative's claim match | Transaction documents and the theory pleaded | A defense unique to the representative |
| Adequacy | Will the interests of absentees be protected | Declarations of the representative and counsel | Intra-class conflict or an unengaged representative |
Adequacy in two parts
Adequacy divides into the representative and counsel. As to the representative, the inquiry is whether that person's interests conflict with those of absent members and whether the person is sufficiently engaged to direct the litigation rather than lend a name to it. As to counsel, the rule provides a separate mechanism: an order certifying a class must appoint class counsel, and the court considers the work done to identify the claims, counsel's experience, knowledge of the applicable law and the resources to be committed.
Because absent members will be bound by a judgment they did not participate in, adequacy is the element with constitutional weight behind it, and a failure of representation is the ground on which a judgment can later be attacked collaterally. The particulars of both halves, including the credibility challenges that defendants raise against named plaintiffs, are set out in adequacy of the named plaintiff and counsel.
The rigor the analysis demands
Certification is not a pleading exercise. The party seeking it must be prepared to prove that the requirements are in fact satisfied, and the court must conduct an analysis rigorous enough to be satisfied of that, which frequently entails some overlap with the merits of the underlying claim. Where evidence conflicts on a point material to a prerequisite, the court resolves the conflict for that purpose rather than deferring it.
Timing is set by the same rule. The court must determine whether to certify at an early practicable time after a person sues or is sued as a class representative, which in practice means after discovery targeted at the certification questions rather than at the outset of the case. An order granting or denying certification may be altered or amended before final judgment, so a class certified on one record can be narrowed, divided or decertified as the evidence develops.
Two limits keep the inquiry from swallowing the case. The court considers merits questions only to the extent they bear on whether the rule is satisfied, and findings made at this stage do not bind the trier of fact at trial. Once the prerequisites are established, the analysis moves to which category of class is available, which is the subject of the three routes to certification, and, for damages classes, of predominance and superiority.
Points to carry away
- Numerosity asks whether joinder is impracticable, not whether a fixed number of members exists.
- Commonality requires questions capable of classwide resolution, so that a common answer drives the litigation.
- Typicality compares the representative's claim and the defenses against it with those of the class.
- Adequacy addresses conflicts within the class and the qualifications of counsel.
- The analysis is rigorous and may require resolving factual disputes that overlap with the merits.
Questions readers ask
Is there a minimum number of class members?
The rule states no number. It asks whether the class is so numerous that joinder of all members is impracticable, which is a practical judgment about the burden of bringing everyone before the court. Courts weigh the size of the group alongside the geographic dispersion of members, the ease of identifying them, the size of individual claims relative to the cost of suing, and whether members are able to bring their own actions. Small dispersed claims satisfy the requirement more readily than large concentrated ones.
How does the class definition itself affect certification?
The order granting certification must define the class and the class claims, issues or defenses, so a workable definition is a condition of the order rather than a preliminary matter. A definition drawn by reference to the ultimate legal conclusion, such as all persons injured by the defendant's conduct, requires a merits determination before membership can be established. Definitions tied to objective criteria, such as transactions of a described type in a described period, avoid that circularity.
Can a court decide disputed facts at this stage?
It can, and often must. The prerequisites are matters to be proved rather than pleaded, so a court may weigh competing evidence and resolve disputes bearing on them even where the same facts also bear on the merits. What the court is not doing is deciding whether the class will win. The permitted inquiry extends only as far as necessary to determine whether the rule's requirements are satisfied, and findings made for that purpose do not bind the trier of fact later.
Sources
- Federal Rule of Civil Procedure 23 — Class ActionsStates the four prerequisites and the certification order requirements.
- Federal Rule of Civil Procedure 23.1 — Derivative ActionsShows the separate representative device that Rule 23 does not govern.
- Federal Rules of Civil Procedure — current textThe judiciary's own publication of the operative rules.
- 28 U.S. Code § 1332 — Diversity of citizenship; amount in controversyContains the class action jurisdictional provisions that determine the forum.
- 28 U.S. Code § 1453 — Removal of class actionsGoverns removal of class actions and the review of remand orders.
- Federal Judicial Center — Manual for Complex Litigation, FourthThe judiciary's reference work on managing class and complex cases.
Pinnacle Law Review is a publication, not a law firm. This article states general rules and cites its sources; it is not advice about any particular case, and the law differs by state and changes over time.
More in Class Actions
Predominance and Superiority in a Damages Class
Certification of a damages class requires findings that common questions predominate over individual ones and that a class action is superior to other available methods of adjudication. Predominance is a qualitative weighing of the issues that would be tried, not a count of them. Superiority considers members' interests in separate control, litigation already begun, the desirability of the forum, and the manageability of a class trial.
Expert Evidence at the Certification Stage
Because certification requirements are proved rather than pleaded, courts receive expert opinions on class membership, common impact and damages methodology. The evidentiary rule requires that expert testimony rest on sufficient facts, reliable principles and methods, and a reliable application of those methods. Circuits differ on how completely that screening must be performed at certification, but a model that does not match the certified theory of liability fails regardless.
The Three Routes to Certification
Rule 23(b) supplies three categories. The first covers cases where separate actions would risk incompatible standards for the party opposing the class or would impair the interests of absent members. The second covers cases where the opposing party acted on grounds applying generally to the class so that final injunctive or declaratory relief is appropriate. The third covers cases where common questions predominate and a class action is superior.


