The advent of digital data has put innumerable demands on the marketing analysts and the database professionals. The need is to model the data in a way that it starts to make sense after getting derived from all these complex sources.
When we talk specifically in terms of traditional marketing data, it consisted of those streams that were easily understood to everyone. Variables such as gender, income and education were categories that were easy to identify by all. Attitudes could be attached to them in order to match business opportunities. Relatively when we talk in terms of digital data, it is not so forthcoming. This is mostly because it comprises stream of facts rather than discrete atomic data. There is a specific meaning which is attached to these streams and it only after understands what the pages mean that it becomes possible to infer what was being attempted to be accomplished by the visitor. For years now, the technique of segmentation has been the most impactful technique of data segregation in the field of marketing. The traditional categorization of visitors is done on the basis of business relationships. At the highest level is the category of customer vs prospects vs non-qualified. Even when we talk in terms of digital data, this kind of segmentation does not go away. The firms are still concerned about the kind of relationship that the customers have with them. But the concern of digital data aggregation remains a concern here. To add to it , digital data is most of the times completely anonymous leaving the analysts in a state of fix. Two tiered segmentation does help in dealing with the problem of digital data aggregation to an extent if not completely