Can you tell by the menu prices that a restaurant will provide you with a great experience? Will reading the final score substitute for the experience of the game you could not attend? Some analytics practitioners would answer “yes” to these questions, especially practitioners who specialize in the field of Social Media Research. They say that transactional data, such as a number of re-tweets, Facebook “likes” or link clicks, provide meaningful insights into human behavior or intentions.
Wikipedia defines analytics as
“… the discovery and communication of meaningful patterns in data. Especially valuable in areas rich with recorded information, analytics relies on the simultaneous application of statistics, computer programming and operations research to quantify performance.”
Which data patterns should be considered “meaningful” depend entirely on the definition of “performance.” Complaints about the practical usefulness of research arise when business process owners shy away from or fail to define the performance they need to quantify. In other words, it is best to first figure out the actions you could take if you had the kind of information Social Media research can deliver. A successful example of this approach is extending Customer Services processes by using Social Media listening and monitoring to combine transactional data with rudimentary sentiment analysis. This approach quickly gained a lot of traction with many companies.
Yet it is not as easy to find truly actionable research applications as commonly adopted by marketing business processes, as these rely more on contextual than transactional data. Such contextual information is usually provided by traditional market research methodologies that solicit feedback from consumers and customers. However, these techniques are much better suited for confirming existing hypothesis than discovering meaningful new patterns, so they are not the same as contextual analytics. Since meaningful contextual analysis of Social Media is relatively new, very few marketing organizations have existing business processes that are immediately extendable.
Below is just one example of the distinction between transactional analytics and contextual analytics (borrowed from the previously published article):
New processes are evolving that use contextual analytics of Social Media to focus on improving market segmentation, marketing communications effectiveness and customer-driven new product development. These processes depend on the aggregation of transactional, contextual and operational analytics to produce substantially higher revenue growth.