You may have read my article last week on the differentiation between profiling, modeling and segmentation. Hopefully you were able to learn a lot from that entry. Here is a more detailed breakdown of types of segmentation and their uses.
When building a customer segmentation, it is important to determine up front whether the segmentation will be used for prospecting, customer campaigns (active and/or lapsed), or both. This is important because, in order to leverage the segmentation in marketing campaigns or operations, variables used to create the segmentation must also be available on your target audience. For example, if you build the segmentation on a group of prospects using appended data, you can use that for your prospects and your customers (because you can append the same data). If you build the segmentation on customer data, such as transactional data, you cannot apply it to a group of prospects because you do not have transactional data for them. That is not to say that it isn’t worth building a customer campaign on customer data though as customer data is often very powerful in segmentation and predictive strength. In addition, depending upon the needs of the campaign, financial services can do consumer or business segmentation.
Here are some types of segmentation and what they mean:
A Priori Segmentation
A priori segmentation refers to segments that are pre-determined and then people are assigned a segment based on certain criteria. Common examples of this would be High Net Worth Customers or generational cohorts.
Statistical segmentation methodology follows a three-phased process framework. I would recommend using clustering methodology with K-Means clustering technique for segmentation. Clustering is an undirected method and does not require the selection of target variables.
Clustering is preferred for the following reasons:
Cluster analysis is based on data patterns to build segments that are distinct, i.e., homogenous within and heterogeneous between the segments. That is, all similar customers showing the same behavior would be clustered into one segment.
Where there are large numbers of independent explanatory variables, clustering is a preferred technique: it allows for data reduction, providing information on specific subgroups within a population.
Allows for the partition of the universe into similar clusters.
Allows for identification of relationships and profiling of segments within a defined universe.
Allows for better understanding of different types or groups of stores.
The K-Means cluster Analysis technique is preferred, considering the large data sets and the efficiency of the K-Means clustering technique with large data sets. It provides superior results on new data sets in identifying the cluster seed points.
Dimension – A single variable, element, or attribute.
Cluster – A single cell representing a grouping of like stores. cluster=segment=cohort.
Segmentation – A group of clusters identified using a clustering methodology or a priori assignment.
Segmentation Schema – The entire segmentation approach, which could include multiple segmentations.
Segmentation Matrix – A matrix with each cell cross sections of the segmentations being used for an initiative. All cells will not necessarily be used.
I hope this was helpful to better understand segmentation terminologies. Stay tuned for my next entry, where I will further explore the three phases of segmentation methodologies.