In the latest chapter of the Lavigne v. First Community Bancshares saga, the District of New Mexico approved a very broad notice plan based on the notoriously unreliable reverse-lookup method. TCPA class-action practitioners are intimately familiar with this “methodology.” See No. 1:15-cv-00934, 2019 U.S. Dist. LEXIS 37724 (D.N.M. March 7, 2019). The plaintiff’s class identification expert runs a list of phone numbers through a few different third-party databases that claim to show the subscriber and users of cell phone numbers at points in time. The identification expert then uses a similar method to attempt to locate a current address for the individual identified in the reverse lookup. There are many problems with this method, including the fact that the third-party databases often produce different results for the same data.

Reverse-lookup is particularly unreliable in wrong number classes like Lavigne. As the defendant pointed out, the methodology will simultaneously fail to identify many members of the class while at the same time falsely identifying people are not actually in the class. That is because in a wrong number class, the parties rarely know other identifying information about proposed class members, such as a name, address, or date of birth, that can be used to verify the results or obtain more accurate data. Generally, the more inputs you have, the more accurate the output. But that information is lacking in a wrong number class, which is kind of the point. If the defendant knew the information, it would have known that the number was incorrect and would not have called it. So to identify who the number actually belonged to, the court and parties have to engage in a process that all involved admit will produce many incorrect results.

To its credit, the Lavigne court admitted “there may be errors in identifying class members,” but claimed that is not a reason “to deny class notice.” (Emphasis in original). And that may be true. But it is – or at least should have been – a reason to deny class certification.

One of the more disturbing trends among some circuits is to move away from requiring plaintiffs to show ascertainability at the certification stage. These circuits instead punt ascertainability questions to later, post-certification proceedings. Lavigne is a prime example of the problems with that approach. The plaintiff gets their certified class, but then has to give notice. Without any real options to identify class members accurately, notice will go out to several people who shouldn’t get it, and others who might actually be part of the class are left out. That prompts another fight altogether on whether the class should be de-certified. De-certification requires additional briefing, additional discovery, and additional hearings. Meanwhile, both parties are racking up considerable fees and expenses and taking up valuable court time. All of this can be avoided by simply requiring the plaintiff to show that a class is ascertainable at the certification stage.