Target Group Segmentation
Do you know "the customers"? Presumably your market consists of a large variety of different customer wishes. The "average customer" is a theoretical construct that can be geared towards performance though it is ultimately not preferred by real individual people.
The solution lies here in dissecting the market into individual homogeneous segments, which correspond to their preferences, wishes, attitudes or other characteristics.
Products can then be optimized with regard to these segments, communication concepts can be examined for their effect in view of different target group segments, the pricing policy can be differentiated according to price sensitivity and so on.
This helps in the discovery that it is not the "market" or abstractly defined target groups which "demand" something, but that various products / concepts are preferred by different segments at any one time. It can then be decided whether more segments are served by product differentiation or whether a target group is selected which is either particularly attractive or, "suits" the product the best on the basis of the brand core.
Cluster analysis is a classic method for segmentation. By using cluster analysis segments are established according to product requirements or to the perception of different outlines of the concept. The developed segments can then be described by means of additional background variables (residential area, income, general attitudes or something similar).
Discrimination analysis or logistical regression can:
- Uncover influence levels that particularly contribute to segment formation or which particularly differentiate these segments respectively.
- Develop classification models to which other people can be retrospectively assigned and therefore enable the analysis of new data with classifications that have already been found.
Alternatively so-called classification trees can be used. Segments are shown through a combination of explanatory sizes, which are very homogeneous in respect of target sizes. The distinction between private and business customers is additionally important, for different product characteristics are relevant to their product decision.
Mapping techniques (e.g. simple and multiple correspondence analysis) are a further possibility that allow segments or, for example, socio-demographic characteristics and products or individual needs like comfort, information on price, information on quality) to be portrayed in a mutual space.
