Read Analytics and Data Analysis Blog News

Social network marketing analytics may group mobile (eg. iPhone, BlackBerry) phone users based on geographic networks with whom they text. Online marketers might then target mobile phone social groups with ads for products and services of interest: customized based on text and clustering algorithm analysis of text messages. A similar form of analysis can group loyalty program members by spending habits or identify strong associations between retail products that sell well together. Tablet PC and smart phone Apps – and other built-in functionality for remaining mobile users – may be used to obtain consent from subscribers to analyze their data. Obtaining this permission in accordance with Canada’s information privacy laws might alleviate privacy concerns that Facebook and Google have been confronted with.

Much of the talk about social network analytics has focused on analyzing social media data: Twitter tweets, Facebook shares and blog post comments. This analysis might reveal brand perceptions that must be addressed or customer segments around specific products or services (eg. travel, entertainment). The following article writes about how clustering data analysis may be applied to Facebook or MySpace social network user information: Applying Euclidean Distance Clustering to Social Network Data. It was written a few years ago when the term “social network analytics” was still in it’s infancy. Yahoo!’s Livestand personalized, digital Tablet magazine iPad App uses a form of social network analytics to identify which video, image and text content that Yahoo! readers prefer. But mobile text messages likely also contain online marketing insights if analyzed with text and clustering data mining algorithms: SAS, IBM-SPSS or Angoss would likely support this assertion.

Canadians sent about 6.7 billion text messages in September 2011 according to the Canadian Wireless Telecommunications Association (CWTA) industry stats page: Nine-Month Text Message Total Reaches 57 billion. This country’s over 26 million mobile phone subscribers send an average of just under 9 text messages per day – based on mobile subscriber data from the February 29, 2012 MobileSyrup blog post: Total estimated Canadian wireless subscribers reaches 26,383,417. In short, there are a lot of texts between Rogers, Bell, Telus, and other mobile phone customers – about 9,103,000 of whom are smart phone (eg. iPhone, BlackBerry) users according to quoted comScore research in the February 23, 2012 MobileSyrup blog post: Apple is poised to take the lead in Canada’s smart phone market share.

Social network analytics algorithms may be used to analyze Canadian’s text message data and segment mobile subscribers according to ‘Geo-Groups’ of others with whom they text. These are like the algorithms that help researchers conduct an online library book search or perform a Google Internet keyword search. Mobile text message data might be segmented with clustering the way customer loyalty product purchase behaviour has been for years. Message data might be analyzed similar to the way in which market basket analysis has successfully searched for retail products that sell well together. Search engine optimization (SEO) experts may use a similar technique to find keyword combinations that make it easier for a Google Internet searcher to find an optimized website.

As far as the Canadian mobile market is concerned, RIM’s BlackBerry presents the best opportunity to carriers (eg. Telus, Rogers, Bell) who wish to target mobile subscribers with customized mobile ads. All three stakeholders – manufacturer, carrier and mobile user – are based in Canada making it easier to address privacy laws in co-operation with the Office of the Privacy Commissioner of Canada. Mobile carriers might also develop their own location-based mobile ad systems for retail partners: such as iSign Media Solutions’ recently announced mobile ad partnership with Canada’s 1,400 Mac’s and Couche-Tard convenience stores.

Mobile text message segments could potentially be used as independent variables to predict such things as mobile subscriber loyalty, marketing campaign response or usage time. While much of the social network marketing analytics hype has revolved around Twitter and Facebook data, Canadian usage data is owned by U.S.-based corporations. And that may make it difficult for the clustering analysis described in the above-mentioned KDNuggets article to be performed.

Rogers, Bell, Telus, and other mobile phone carriers have an opportunity to apply clustering and text mining algorithms to more accessible (BlackBerry) text message data. Segments from this analysis might be combined with other retail point-of-sale (POS), customer service and website activity mobile subscriber account data: doing so may yield more accurate predictions of mobile subscriber loyalty, airtime usage and attrition. And this might be accomplished in a way that is sensitive to the privacy concerns of Canada’s wireless (BlackBerry) mobile phone subscribers.

Mobile Phone Marketing Analytics: Digital Privacy ...

Social Network Marketing Analytics: Smartphone Con...