TWTR Tweets Tell the Tale Getting Ahead of a Story Using Social Media Analytics

At about 3:07 PM EDT on April 28, financial data firm Selerity sent out a tweet disclosing that Twitter would miss their earnings projections:

This in itself isn’t extraordinary, but the problem was that the earnings call wasn’t supposed to take place for about another hour. Rumors swirled that this information had somehow been “leaked” and Twitter’s share price dipped about 5% until a trading halt was placed on the stock in order to determine what had happened. When the halt was lifted a little less than 20 minutes later, the results were staggering: TWTR had fallen by over 20% in the course of a few minutes.

How did something like this happen? Despite early claims of a hack or leak, it instead turned out just to be a case of misplaced information, as Selerity later clarified:

Since then, the Internet has been inundated with many different stories about why this happened and what it means for Twitter. While the fact that Twitter’s stock price was hurt by a flood of information broadcast on Social Media is an ironic twist, the more important point to focus on is that this entire story played out in real-time on Twitter. Tools that can quickly identify market trends via social media and reliably turn that information into actionable intelligence are extremely valuable in situations like these.

Here is a chart that tracked the social sentiment around TWTR as the news began to develop:

Social Sentiment Declines Sharply After the Initial Tweet

Social sentiment tracking utilities were able to filter through the noise of Social Media at large and pick up this negative trend well before the information would have been made available by any traditional news sources. As noted above, once the news became more widespread and Twitter requested that trading of their stock be halted, 12 minutes had already elapsed since the swell of negative social sentiment was identified. Being able to quickly find, process, and react to this type of information is invaluable during these few otherwise uncertain minutes.

Another measure that can be useful in identifying actionable events is the amount of buzz a given security is getting on Social Media. Are people mentioning this stock more than usual? How does the current amount of activity compare to the relative baseline of activity for this security? In this case, Social Alpha’s volatility indicators picked up on the huge increase in activity around TWTR at this time:

Buzz Around TWTR Nearly Triples in a Few Minutes

The number of mentions that TWTR received on their own platform almost tripled in the span of a couple minutes, more than enough of a shift to set off some alarm bells.

The final interesting aspect here is just how accurately social sentiment can be as a leading indicator of stock price:

Price of TWTR Stock Followed the Trend in Social Sentiment

As can be seen in the chart above, sentiment begins to dive, and the price soon follows. Also noticeable is that after trading resumed and the stock price plummeted, sentiment then took a slight positive turn, and after a few minutes, the price rebounded a bit.

For one thing, this story shows why it’s important to protect information in a world where Social Media has dramatically increased the speed with which that data can be shared. But equally as importantly it shows how, using the right tools, the power of social media can be harnessed to provide insight and risk management during those crucial few minutes when a story is still breaking.