Chris Moody tells us Gnip is founded on the idea that social data represents unlimited innovation. “This full room proves that there will be unlimited, exciting applications,” he says. 100 billion pieces of data are created each month and Gnip is going to understand what makes each set of data unique. Over 90% of Fortune 500 gets content indirectly from Gnip. The variety of people in attendance at Big Boulder is representative of the dense social data ecosystem.
There are many different data sources, which is good news and bad news. New coverage platforms added by Gnip this year inclue WordPress, IntenseDebate, Disqus, Tumblr, and Sina Weibo. So how does an entrepreneur or company prioritize and organize the data that is important to a specific set of customers? This is what Gnip does.
Chris and his team focus on two different dimensions of social data:
- Reaction Time: ranging from ultra-fast (example: Twitter at events) to slow (example: the comments on a blog post or a shared video.)
- Depth: Data is about the what. Twitter is concise and deals with the immediate, “This is so adjective!” It’s difficult to get into the reasons in only 140 characters. Platforms like YouTube and Tumblr are at the other end of the spectrum and tend to boast deep, personal content. When you begin to overlay business use cases over these different data sources, you see the social cocktail. On one side is public relations and crisis management. If you have a client looking to manage a crisis, the priority is your speed in rectifying the situation. If a person complains that their cell phone is on fire, you don’t need to ask them how they feel about it. It just needs to be immediately addressed. On the other side is brand management: what does the collective universe think about not only you but also your competitors? If you’re the cell phone provider with the faulty and explosive device, what was said about you and what was your response?
Chris brings up an example slide of the market response to Netflix earnings last year. The world’s reaction to the opening bell on Twitter was very fast and factual. It demonstrated information based around how correlated the stocks were performing, or the what. Blogs had a half-life on the curve, well past the end of the trading day when blog content and a variety of opinions began to go public, or the why. Comments to this kind of content peaked much later by comparison, the next day in this case.
There are two noticed patterns in social data: expected and unexpected. It’s also important to observe whether the occurrence is a routine or an event. Generally speaking, no one expects a hurricane. Likewise, the social data around a natural disaster spikes drastically as soon as it occurs. We’re continuing to understand this action because new sources are constantly adding new conversations. As a second example, JP Morgan’s surprise trading loss illustrated a strong story in the comments of its articles, animated political .gifs for humor, and theories. The story shared was factual and the reaction to the story was highly narrative and emotional. As a third example, a recently shared image of the new Urban Outfitters line went viral not only because of Urban Outfitters’ huge market but also because the image was so easily sharable in a micro-blog format.
The message being driven home at Big Boulder today and tomorrow is, “Where is this industry going?” In an interview style format, Big Boulder is Gnip’s first conference and panels are conducted around the data science of what we collectively think and feel.