The answer is nowhere near as simple as you might think. Everyone knows that a court is supposed to conduct a “rigorous analysis,” but what that means in practice is not quite as clear.

For example, last year, the Supreme Court (in dicta) made a statement about Rule 23 that many, including myself, thought would reverberate throughout class action cases in the years to come. In American Express Co. v. Italian Colors Restaurant, it said that Rule 23 “imposes stringent requirements for certification that in practice exclude most claims.” A year later, only one court has mentioned the Supreme Court’s characterization, let alone applied it. By contrast, in that same time, a number of courts have continued to follow the Second Circuit’s dictate that Rule 23 be given a “liberal rather than restrictive construction,” or held that courts should “err in favor” of certification.

Similarly, the degree to which a “rigorous analysis” may require further factual inquiry still varies by jurisdiction. The Southern District of Texas has decided that Justice Scalia’s admonition that Rule 23 requires affirmative evidence applies to every facet of the Rule, even those, like adequacy, that courts have traditionally enforced only loosely.  Most courts would not go that far.

The fact is that the standard applied on the ground for class certification still varies wildly from court to court. Some of these variations come from simple error, such as the handful of district courts in the past year that still treated allegations in pleadings as true for certification.

Penn law professor Tobias Barrington Wolff believes that the variation comes from the fact that courts are granted wide discretion to decide class certification. In his article Discretion in Class Certification he comprehensively surveys the various ways in which courts have employed that discretion, both to certify classes, and to deny certification even when there is no textual requirement to do so. Some of the examples he provides stretch his thesis a little too far. Courts certifying a class under Rule 23(b)(3) have every reason to consider manageability; it’s built into the text of the Rule. Similarly, while Judge Posner’s opinion in In re Rhone-Poulenc Rorer and the majority opinion in Castano (two cases he contends declined certification despite the lawsuit having met all of Rule 23’s requirements) both discuss the difficulty of certifying classes in “immature torts,” they also both rely on the predominance requirement as spelled out in Rule 23.

But while Professor Wolff may have overstated the degree to which courts exercise their discretion, he makes a cogent argument as to why that discretion is necessary, given the rapid advancement of litigation technologies compared with the slower pace of changes to legal procedure. He argues that this discretion is a necessary “safety valve” for Rule 23, allowing courts to deny certification in individual cases rather than forcing rulings that would turn Rule 23 upside down on a regular basis.

And it’s true, as I’ve written elsewhere that appellate and trial courts have a number of strategies for cabining class action rulings with which they disagree.

But that doesn’t obscure the real lesson here. And that lesson is: that boilerplate about class certification standards at the start of your brief? Treat it as more than just boilerplate, because this is an area where it’s possible to lose the certification debate before you’ve even started.