A report that Instagram's user numbers were plummeting spooked Facebook investors so badly that the stock dropped 3 percent this morning.
So how did this happen, and what do we actually know about Instagram's usage?
It's a vital question, because Facebook's acquisition of Instagram has been seen as a major victory, keeping Twitter and other competitors from grabbing an incredibly fast-growing photo-based social network. Instagram now has substantially more mobile usage than Twitter.
But Instagram's now seen as vulnerable after a badly handled rollout of new terms of service which scared users into thinking Instagram was going to use their photos in advertisements. Flickr and Twitter's rollout of new mobile photo-sharing features seemed opportunely timed to take advantage of the fuss.
So the narrative that Instagram had taken a big hit felt intuitively appealing.
This much everyone agrees on: An app activity-counting service, AppData, run by Inside Network, showed a sudden, marked drop in Instagram users connected to Facebook around Christmas.
That's not all Instagram users—just the ones who have linked their Instagram and Facebook accounts and actively connect the two by logging in, sharing photos or likes from Instagram to Facebook, or other actions that bridge the app and the social network.
AppData tracks activity on a daily, weekly, and monthly basis. On a monthly basis, Instagram has grown 15 percent from Thanksgiving, when it had 38 million monthly active users who posted a record-setting 10 million photos. It now has 43 million monthly active users connected to Facebook.
AppData's weekly numbers are steadily up, too. They grew by 2.3 million users to 29.4 million in seven days.
Those are not numbers you'd associate with a troubled app.
It's only in the daily numbers where we see a drop, from 15.9 million to 12.4 million users.
If you're familiar with AppData numbers, you know that they occasionally show short-term glitches, which the service ascribes to changes in how Facebook reports the underlying app-usage numbers it relies on. These usually work themselves out over time as Facebook improves how it reports app activity, but they bedevil app developers, especially social game makers, for whom AppData numbers constitute something like TV's Nielsen ratings.
Around the holidays, of course, a lot of people got new smartphones and tablets, with new operating systems.
In its latest version for iPhones and iPads, Instagram introduced support for a new feature built into Apple's mobile operating system that connects apps to Facebook. Facebook-connected apps worked on iPhones before, but the new OS feature makes things a lot smoother.
This is just a theory, but what if that direct, OS-level connection changed the way Instagram's app interacts with Facebook in a way that causes less apparent activity on the service?
One other explanation: Instagram made some change to the app that upset a lot of real users, who tend to care about things like how the app works and how their pictures look much more than they do about poorly worded legalese.
Either of those theories strikes us as more plausible than 25 percent of Instagram's user base abandoning it overnight even as it adds millions of new users at breakneck speed.
We asked Inside Network, the operator of the AppData service, for comment on our theories, but haven't heard back yet.
Facebook disputed AppData's numbers but didn't offer details.
"This data is inaccurate," a Facebook spokesperson told us. "We continue to see strong and steady growth in both registered and active users of Instagram. We don’t have anything more to share beyond this statement."
More From Business Insider