Twitter is increasingly being used as a source of breaking news. This is a good source of content but the level of context does not paint the full picture. For example, a person can say something that is positive or damaging to your brand, however, viewing this on a twitter feed of a hashtag could result in the trend setter being lost as the conversing community grows. More importantly, it is not always obvious how influential participants are coming into the conversation (within their specific social networks), nor how influential they are during a specific conversation.
Deloitte South Africa is able to overcome both of these problems with a visualisation of a twitter feed. The person that started a conversation will appear in the center of a map of the conversation. As the conversation grows, the people that contribute the most to the discussion (hash tag) will move to the centre of the map. In addition, you are able to quickly identify who the most influential participants are (represented by the size of the node) as they join a conversation. Finally, one is able to map multiple conversations at the same time using multiple hashtags (represented by point of entry and node colour).
To illustrate this, Deloitte created a visualisation of the twitter conversation during the South Africa – Fiji rugby match on Saturday 8am (GMT+2). The hashtags used for the search were #RWC2011 and #BOKKE. The video below shows a small clip of how the twitter conversation started. Visuals are richer than plain text, so we will explain what the video represents. Each node that appears is a unique user on twitter. The purple nodes are the users that used the #BOKKE tag and will enter the screen from the left. The light green nodes are users that tweeted with the #RWC2011 tag and will enter the screen from the right. A line between the nodes represents a tweet from one person to the other. This is so that we can visualise a tweet where one person mentions multiple people with multiple lines. The size of the node is the Klout score. Klout is an online tool to estimate numerically the level of influence a person has. During the video I will be able to rotate the map along the z-axis and you will be able to see a difference in the height of the nodes. The difference in the height is due to the number of mentions that particular user has. The greater the number of mentions, the higher the level of the node.
An important analysis of the map is that the people in the center of the map are visually and literally the center of the conversation. This is intuitive as the nodes in the center are usually bigger, implying a high Klout score.
The graph below shows the number of tweets that arrived during the match. The green line is the #RWC2011 tag and the blue line is the #BOKKE tag. We noticed that twitter becomes more frenzied as a point is scored. People also make more comments after a match.
The application for this technology is that you can map a conversation to understand the creators of the content, their relative level of influence, and the context in which a conversation is flowing. This enhances traditional Online Reputation Management. Businesses are now able to understand both historical and real time influence within conversations and manage the viral spread of a breaking topic with far greater precision using these solutions.
To find out more information about how this technology can give you an advantage please contact Andre Hugo @maxrsa or Mithun Kalan @mithunkalan on twitter.
Filed under: Executive Leadership, Technology, Media & Telecoms (TMT), #RWC2011, Bokke, Conversations, Data Analytics, ORM, Twitter, Visualisation Graphs
Deloitte SA 