Media Keeps Calm, Carries on During Twitter Outage

Service Dark for More Than an Hour Thursday Afternoon

By Published on .

There's nothing like a Twitter outage to remind media and marketers how dependent they've become on, well, Twitter.

To the chagrin of social-media newsroom producers and marketers everywhere, the site experienced its second major outage in five weeks, leaving users without access for well over an hour.

The outage appears to have begun around 11:15 a.m. EST and service was restored but spotty an hour and a half later. An early afternoon update on the site Twitter uses to post about service issues read in vague terms, "Users may be experiencing issues accessing Twitter. Our engineers are currently working to resolve the issue," but Twitter hasn't responded to request for comment about what caused it to go dark.

Message users saw when they tried to log into Twitter.
Message users saw when they tried to log into Twitter.

While the service was down, users saw a blank screen with the large bolded text, "Twitter is currently down for <%= reason >.", instead of the "fail whale" (an image of a whale in a net weighing down a flock of Twitter birds that are trying to lift it skyward) that often materializes during downtime and has become a well known, if not exactly beloved, icon in its own right. Twitter later confirmed service was restored by 1:25 EST.

Twitter last experienced a severe outage on Thursday, June 21, when there was intermittent downtime throughout the day. It was officially attributed to a "cascading bug" that affected various software elements.

A recurrence of outages is potentially troublesome for a company looking to court brand advertisers and potentially go public in the not-so-distant future. Then again, Twitter isn't alone with its uptime issues: Facebook was down or slow to load for some users for roughly two hours this May, according to TechCrunch.

"What this probably reinforces is the importance of having a multiplatform approach so that we don't put all of our eggs in one basket, so to speak," said Scott Monty, Ford's head of social media.

Of the media publications Ad Age contacted to ask about how social-media managers had responded to Twitter going dark, People magazine confirmed that People.com had experienced a drop in referrals during the primary hour of the outage, "but it was more of just a blip, nothing significant," according to a spokeswoman. Ad Age 's own referrals from social media were down 50% from a typical day.

A spokesperson for the Huffington Post said bit.ly shortened links (which the Huffington Post uses to create the huff.to links that it shares on Twitter and other social networks) dropped 50% between 10 and 11 a.m. and 70% by noon. Clicks returned to about 10 a.m. levels by 1 p.m.

On the other hand, Buzzfeed reported little impact on traffic. "I checked with our research team and it looks like there was minimal effect on our Twitter traffic today and our Twitter traffic is similar to yesterday's," a spokesperson said. "There was no effect on Facebook."

Nor, as it turns out, did Forbes. "When you build social traffic that benefits from all social networks and you have content that is shared and linked to by thousands of websites, it's just a piece of the puzzle," said Lewis DVorkin, chief product officer at Forbes Media.

Individual reactions to the outage when Twitter was back up and running were bolder. Writer/comedian Andy Borowitz tweeted, "You learn a lot when #Twitter is down. For example, my wife left me four years ago" from @BorowitzReport.

Meanwhile, CNBC built a Storify of the migration of snarky comments from Twitter to Facebook.

Most difficult for Twitter devotees was not having Twitter to complain about Twitter being down.

Update: Twitter gave some additional details on the cause of the outage: "The cause of today's outage came from within our data centers. Data centers are designed to be redundant: when one system fails (as everything does at one time or another), a parallel system takes over. What was noteworthy about today's outage was the coincidental failure of two parallel systems at nearly the same time."

In this article:
Most Popular