Case Study: How Content Diffuses Through Different Social Networks

Scott Hendrickson

Posted on April 5th 2013

Case Study: How Content Diffuses Through Different Social Networks

I work at a company that captures and manages social media data. Gnip captures all social media signals from major social networks including Twitter, Instagram, WordPress, Facebook, Google+ and Tumblr.

One question we’re interested in is how content evolves along the different paths it takes on different social networks.

Because users behave differently depending on the social media platform they’re using, news travels in different pulses depending on what the news is and where it is traveling. This is a quick look at how content is diffused via different social media publishers.

Expected vs Unexpected Events

Let’s take a look at expected versus unexpected events, which each have very different pulses in the form and speed information travels around these events.

Expected events drive significant volume on social media because people know they’re coming and plan to comment, observe, banter, trash talk and analyze. Twitter volumes surge during the World Cup, the Super Bowl, Video Music Awards, etc.

However, theses events are characterized by a gradual growth and decay around the event rather than abrupt changes. A hurricane is an example of an event that people are anticipating, so you’ll see a gradual growth and decay around the event rather than abrupt changes. The bump in volume may last from a few hours to a few days.  It is often somewhat symmetric in its growth and decay, although decay is typically faster than growth.

On the other hand, unexpected events result in abrupt spikes in volume. On Twitter, these spikes may reach tens of thousands of related tweets per minute within five minutes of an event. An earthquake is the perfect example of an unexpected event. Social data volume around unexpected events usually grows rapidly until the networks for related users are saturated with the information, then the volume decays exponentially. These spikes have well-defined growth, peak and decay half-lives.

Understanding patterns of activity help us understand how to monitor better and what to expect around different events. For example, the figure above shows the common unexpected social media pulse function (orange) and shows how the Twitter traffic diverges from this standard model after large news outlets pick up the story--without a model of the social media pulse, we might not realize that the story had changed at about 18:08. 

Different Social Media Networks Have Different Paths

Each social media platform reacts a little differently to events. To show how this works, let’s dive into an example of the evolution of a single story across a mix of publishers. This will provide some intuition into how the social cocktail works when examining a real-world event— in this case, the JPMorgan-Chase $2+ billion loss announcement.

On May 10, 2012, immediately after market closing, JPMorgan-Chase CEO Jamie Dimon held a shareholder call to announce a $2 billion trading loss. While traditional news agencies reported the call announcement late in the afternoon, Twitter led the way with reports from call participants who started tweeting while on the call a few minutes after it started:

The reaction on Twitter was fast and concise immediately following the shareholder call, with a second spike once the news stories were published about the loss.

The chatter on StockTwits was similarly fast and concise but with a more focused audience since StockTwits consists mainly of professional investors. The immediate drop-off of information is due to saturation of the news from the highly connected StockTwits audience.

Tumblr’s reaction was unique, with slow momentum building during the first few hours after the shareholder call, but quickly speeding up when as people created “re-bloggable” content about the news. Rather than an event-response reaction such as Twitter, or a considered reaction, as with blogs, the reaction of the audience on Tumblr accelerates as the type of content Tumblrs reblog appears in the network.

With blogs, there was an initial reaction to the news breaking and then more considered and in-depth analysis taking place over time. Because there was a lot of mainstream media attention to the event, news stories started to appear soon after the conference call, and these attracted comments almost immediately. Then the commenting process continued on blogs for some time.

Twitter is often a radar for breaking news but by ignoring other social media publishers you forgo depth and context that cannot be found on Twitter alone.

Toyota Brands on Different Social Networks

Different social media networks obviously attract different audiences, and brands should be aware of the nuances between the different social media publishers.

Gnip conducted a brief analysis of the Toyota family of brands--Toyota, 4Runner, Camry, Highlander, Lexus, Prius, Rav4, Scion, Sequoia, Tacoma, Tundra--on multiple social media platforms. We looked at brand mentions on Tumblr, Twitter, WordPress and WordPress comments during the period of Oct. 15 to Nov. 15, 2012:

As you would expect, Toyota was the most frequently mentioned brand on each social platform, with one enormous exception – Tumblr. Lexus had 5 times as many mentions on Tumblr as Toyota.

This highlights how aspirational brands do exceptionally well on Tumblr where niche communities of fans often form around brands. (Attention brand managers: this happens whether your company is involved or not).

A central component of Tumblr is visual content, which also plays well with aspirational brands. Furthermore, Tumblr content is both extremely viral and has a long shelf life--meaning that content shared on Tumblr can be shared for longer periods of time and jump to more diverse sub-groups within the network than other social networks. During the month we tracked mentions, Lexus received more than 200,000 mentions while Toyota received 40,000.

Measuring and understanding true sentiment in social signals is an endlessly complex and nuanced challenge for any brand.  But through large-scale reach and deep analytics across the range of social networks we can develop meaningful insights to drive product and marketing strategy that can be used across all communications. 

Scott Hendrickson

Dr. Scott Hendrickson

Data Scientist, Gnip

I'm Data Scientist at Gnip in Boulder, CO. Gnip is the leading provider of Social Media Data to the enterprise. In my role as Data Scientist I demonstrate the opportunities and value of mining realtime social media conversations. 

See Full Profile >

Comments

I'm curious why you didn't include the second largest social network? Does Google not make the necessary data available?