For the second time, a California federal court dismissed a Telephone Consumer Protection Act lawsuit against Facebook, ruling that the plaintiff failed to adequately allege that text messages from the social networking site were sent using an automatic telephone dialing system (ATDS).

Noah Duguid originally filed his complaint in 2015, alleging that Facebook sent repeated text messages that his account had been accessed. One problem: Duguid didn’t have a Facebook account. Instead, he had purchased a cellphone with a recycled phone number. So when Duguid received texts stating “Your Facebook account was accessed by an unknown browser,” for example, he could not log in to the account to opt out of security messages because it was not his account.

His first complaint was dismissed by U.S. District Court Judge Jon S. Tigar last March, after the court determined that Duguid relied upon conclusory allegations that Facebook used an ATDS without providing evidence that the defendant sent text messages en masse to randomly or sequentially generated numbers. The court dismissed the suit without prejudice, allowing the plaintiff leave to amend.

But Duguid failed to improve upon his first attempt, Judge Tigar recently concluded, and dismissed the complaint with prejudice. Even recognizing that it may be difficult for a plaintiff to identify the specific type of dialing system used without the benefit of discovery, the court said Duguid’s allegations suggested direct targeting by Facebook that was inconsistent with the random or sequential number generation required for an ATDS.

“Courts generally rely on the message content, the context in which the message was received, and the existence of similar messages to assess whether an automated dialer was utilized,” the court wrote. Duguid repeated the central elements of an ATDS and asserted that Facebook’s system possesses all of them.

Duguid made additional allegations in his second complaint, specifically that Facebook uses a “computerized protocol for creating automated text messages programmed to appear customized to the user” through a template-based process and employed a two-factor authentication system for login approval that requires users to enter a code sent to mobile phones via text message when they log into Facebook from a new or unrecognized device. But these were insufficient for the court.

“It is unclear why Duguid believes these facts would strengthen the inference that Facebook sent text messages en masse using an ATDS,” Judge Tigar said. “To the contrary, allegations of customizable protocols and unique codes only further suggest, in line with Duguid’s other allegations, that the messages were sent through direct targeting.”

Even with the new allegations, Duguid again failed to allege the existence of an ATDS, the court said. “At best, his allegations are conclusory, given that he merely asserts that Facebook ‘maintains a database of phone numbers on its computer’ and ‘transmits alert text messages to selected numbers from its database using its automated protocol,’ without offering any factual support for this claim,” the court wrote. “At worst, this claim contradicts the variety of other allegations offered by Plaintiff, which suggest that Facebook does not dial numbers randomly but rather directly targets selected numbers based on the input of users and when certain logins were attempted.”

Given that Duguid’s allegations “strongly suggest direct targeting rather than random or sequential dialing,” Judge Tigar dismissed the suit with prejudice.

To read the order in Duguid v. Facebook, Inc., click here.

Why it matters: The decision is a victory for Facebook and other TCPA defendants, emphasizing that plaintiffs will not be able to survive a motion to dismiss on bare assertions that a text message was sent using an ATDS. Further guidance may be forthcoming on the issue, as Duguid has already filed a notice of appeal with the U.S. Circuit Court of Appeals for the Ninth Circuit.