By all accounts, it’s been a rough year for Twitter, and an even tougher quarter. With hopes of an acquisition effectively quashed and a new round of layoffs announced this week, the social media platform is still struggling, despite today’s earnings, which actually beat Wall Street expectations.

The company has successfully launched several high-profile media streaming partnerships – in September alone, Twitter began livestreaming the NFL’s Thursday Night football games as well as all three presidential debates. The debate coverage was a particular success in the ad revenue department, with a completely sold-out inventory, as shared by Twitter’s investor relations team. But amid executive shakeups, a wayward overall product strategy, and an increasingly visible problem with trolls, Twitter’s ad revenue numbers (and relatedly, its user engagement and retention rates) are coming under particular scrutiny.

Twitter’s External Traffic Problem

But there’s a bigger problem for Twitter: the great disparity between its external traffic numbers and its logged-in user base. As we explained in our previous coverage of the presidential debates, the distinction between external traffic and logged-in users is especially important for a platform like Twitter, which relies so heavily on ad revenue. Briefly put, external traffic includes people who are exposed to Twitter content with being logged in to the site, such as those who see feeds embedded on sites and services (such as news sites that feature a scrolling Twitter feed in the sidebar). It could also include those who watch Twitter’s livestream content, which does not require a Twitter account to view. But most importantly, external traffic cannot be tracked in the way that logged-in user can be tracked – and that’s a crucial difference for brands and advertisers who use Twitter as an ad platform.

As shown below, Verto Analytics data shows that Twitter’s U.S. external traffic is nearly double that of its logged-in monthly users (among adults, ages 18+). And those external users are spending far more time on the site than logged-in users: in September, the same month that Twitter debuted its NFL and debate livestreams, external users spent an average of 1 hour 47 minutes on Twitter, while logged-in users only spent 52 minutes on Twitter.

Is An Audience You Can’t Track Worth Advertising To?

For Twitter, these high-profile livestreaming deals have resulted in a slight uptick in user traffic (among both external and logged-in users), although the time that these users spend on Twitter has actually dropped from previous months. Regardless, Twitter’s next challenge is to convert some of its massive external traffic numbers into active Twitter users: clearly, the company has succeeded in engaging this external audience with content; now, it must create an incentive for these users to become active users of the platform, and to keep them engaged.

At the same time, Twitter also needs to prove the value of its traffic – external and otherwise – to advertisers. While advertisers have flocked to livestream content, it’s unclear what sorts of tracking, targeting, or engagement metrics Twitter can provide for its external traffic – data that every advertiser relies upon to gauge the effectiveness of their campaigns. If Twitter cannot prove an effective way to monetize its various audiences, its ad revenue (and the entire business model) based around it is in big trouble.

Interested in our data about audience measurement and social media? Subscribe to our newsletter for the latest insights, or follow us on Twitter and LinkedIn.