Tuesday, May 20, 2014

Recap: Social Media - Conversations Matter More Than Numbers



This morning we sat in on NewMR's presentation by Preriit Souda about social media and how conversations matter more than numbers. It was a very interesting presentation and was quite an eye-opener in the whirlwind of social media that takes place every day. A big thanks to Ray Poynter and NewMR for allowing us to be a Platinum sponsor for all of the events taking place in 2014. Preriit offered many insights on social media and the context for understanding conversations. Read on to see our recap of the lecture as well as some images from his invigorating research.

After the presentation, we were able to speak with Preriit briefly on what inspired this type of research. He explained that it all started when he was writing a paper for CASRO last year that took a totally different approach. That approach was what he calls a bottom up approach. While he was working on it, he started looking at traditional social media analysis which he calls top down approach, which is what he demonstrated here today. In addition to this, Preriit comes from a tech savvy background so he is generally away of things done in computer science on data mining. Therefore, he adapted those techniques into ways that could be used in marketing and business insights.


We converse all the time in real life and on social media. With conversation mapping, we are trying to replicate that same thing. The goal is to understand who is talking to who, when and how much. This issue has been largely unaddressed in social media and how we listen since since it all began. In today's presentation by Preriit Souda, Senior Analyst at TNS Global, he is just scratching the surface on how to mine this data.

Since it's emergence 10-15 years ago, social media has become a huge part of our daily lives. We live in the moment, and new data is created continuously. In fact, everyday Facebook users share 2,460,000 pieces of content every minute. But how do we draw insights from this large amount of data? In the past, most of us have used the volumetric lens. The volumetric lens is used to measure our social media success based on the number of followers, likes, retweets, shares, comments, links, impressions, etc. Although those are important metrics to measure, they do not accurately illustrate the value of conversations. Rather, we are interested in conversation mapping.

Preriit stressed the importance of conversation mapping in regard to social media context. With conversation mapping, we look at the most influential people in respect to the conversation we are having (in this case, with a brand). For example, someone may already be influential but for a specific campaign they may not be influential. On Twitter, influential people relate by influential mediums through retweets and mentions... but are the hubs of influence positive or negative?

Conversation Clustering is another method of viewing a large amount of Facebook data. This method shows Facebook posts based on fan behavior. From this, the brand can take ideas that resonate and use them to their advantage. Preriit used a fun example to illustrate conversation clustering. See the image below for a cool data visualization. He used the Facebook Brand Page of How I Met Your Mother. Here you can see different sized dots that represent various Facebook updates. The bigger the circle, the bigger the conversation.



If you are interested in seeing the whole lecture, in addition to the slides from the presentation, they will be available at http://newmr.org/ in a couple of days. Thanks again to NewMR and Preriit Souda for an exciting and insightful presentation topic today!
iframe {max-width:100%;} .embed{ width: 100%; }