The Telecommunications Sector and Social Network Analysis
http://www.itweb.co.za/index.php?option=com_content&view=article&id=142937
An old dog, the telecommunications sector, is learning a new trick, social network analysis (SNA). Social network analysis has actually been around for about 80 years, but only recently have companies started to realize the great potential associated with utilizing this type of analysis. A common misconception of SNA is that it is performed using social networks like Facebook, Twiiter, or Instagram. In reality, SNA is a subset of data mining that can explain individuals’ behavior as it relates to their social ties. The telecommunications sector is taking advantage of SNA for two primary reasons: the churn problem and the availability of data. The churn problem involves companies responding to some of their customers being unlikely to continue their subscriptions and these customers influencing their social networks to do the same. Individuals who are more likely to influence their social network in this way are labeled as having higher “churn influence.” Additionally, telecommunications companies have a wide array of data available to them because of who their customers interact with via telecommunications systems. Companies specifically use call data records, or CDRs, to develop and prepare these social networks for analysis.
Standard churn models, which are created with the purpose of predicting the likelihood that a given individual will opt to cancel their subscription to a given service, do not account for these individuals relation ships. The problem is that they view customers as individuals in a vacuum. This is simply incorrect. Humans are inherently social in nature, and their decisions and opinions can and will be altered and influenced by those in their social network. Therefore, SNA is an extremely valuable tool for improving these existing churn models. Companies can use social networks to adapt to changing market conditions, allowing them to focus on customer retention, which is becoming more and more necessary as the telecommunications sector becomes more saturated. It is important to note that SNA is not a replacement for standard churn models, but rather an analysis tool that can be used to improve the existing models. SNA has quickly become established as an indispensable method for the analysis of complicated real-world networks, demonstrating that the aggregate of an individuals connections, his or her social network, can provide eye-opening insights in seemingly unrelated circumstances.