Investors in Twitter Inc. drove down the company's share price nearly 15% on Wednesday after a disappointing earnings report a day earlier.
Persistent questions regarding the size of Twitter's user base came back into full view after the company reported its worst three-month growth rate to date. It added two million new users, or 0.7%, for a total of 304 million in the quarter ended June 30. (That tally excluded six million new users, mostly in emerging markets, who accessed Twitter only through text-messaging services, known as SMS users.)
But a less frequently disclosed metric added to Twitter's woes: Regular users are signing in less often. Twitter finance chief Anthony Noto said on Tuesday the share of users who visited daily fell to 44% in the latest quarter, down from 48% during the first three quarters of 2014, the only comparable figure available. Those figures count only users?including those who post via SMS?located in the company's 20 biggest markets.
In comparison, about 65% of Facebook's users visited the social network daily in the second quarter.
The two companies define a daily active user slightly differently. Facebook includes not only users who sign in directly but also those who share content from third-party websites or apps like Spotify. Twitter doesn't.
Twitter shares closed down 14.5% at $31.24, and they hit a 52-week low of $31.06 intraday.
The share of Twitter users who take advantage of the service daily is important because the more the service is used, the more ads it can serve each day, which is the primary way the company generates revenue. While revenue beat expectations in the latest quarter, Twitter's difficulty in boosting the bases of regular users could eventually limit its growth potential.
Twitter has offered little visibility into how active its users are during its short life on the public market. It stopped disclosing its main measure of engagement, which it called "timeline views," in the March quarter because it had become irrelevant, the company said.
The company acknowledged as early as its first public earnings report in February 2014 that timeline views were becoming less valuable due to change in the way the service operates. A timeline view was counted each time a user visited Twitter and refreshed the timeline to view more tweets or conducted a search. But tweets that form part of a conversation are now threaded together, diminishing the need for users to search and refresh their timeline.
Twitter has not introduced a measure of engagement to replace timeline views. Instead, it has largely offered general statements that user engagement was rising around newer features like private chat. On Tuesday, for instance, Mr. Noto said the number of searches is "actually growing well" and the number of direct messages "accelerated."
At Twitter's Analyst Day in November, Mr. Noto implied the percentage could reach 51% in the "intermediate term." Since then, it has gone south.
On a call with analysts on Tuesday, Mr. Noto blamed the decline in daily activity on the fact that monthly active users have grown faster than daily users.
"We have not historically focused on driving daily active user growth," he said. "And that's something in 2016 that we will consider more."
Even so, it remains that less than half Twitter's regular users check the service every day. That reality adds to the uphill battle not only signing up new users but also convincing them to use the service on a regular basis. Executives admitted that's not likely to happen anytime soon.
Moves like increasing the number of ads shown per user can only drive serious revenue growth for so long, RBC Capital Markets analyst Mark Mahaney wrote in a post-earnings report. In the end, Twitter will need more users. "That's why user and usage metrics matter. And that's why hitting metrics growth walls really matters."
Write to Yoree Koh at email@example.com