top of page



A useful and well-done segmentation is often easy to recognize because valuable segments can be characterized by having strong internal homogeneousness and strong external heterogeneousness, meaning:

  1. The individuals within the segments will agree on many topics and they share a significant set of values and beliefs (homogeneous).

  2. The different segments will disagree on different topics and have significant different values and beliefs (heterogeneous).


Overall there are 3 types of segmentations, being:

  1. Segmentations based on demography and geography.

  2. Segmentations based on general values. For example, Gallup Compass or Minerva.

  3. Segmentations based on values ​​and attitudes related to a specific category.


The three methodologies each have their strengths and weaknesses. Generally, method 1 and 2 are good as operational models in e.g. media buying strategies – while a value based segmentation (method 3) is a strategic model that provides a deeper understanding of the segments and where actual archetypes can be created in detail providing crucial consumer insights to the business strategy. The latter is recommended as method applied to the current study and will also be the one that is most likely to have the strongest internal and external variance.


The basic theory of method 3 is that any individual’s decision to engage in a particular behavior is based on pre-existing values and beliefs towards a specific consumer category and life in general.


In other words, a waitress in Berlin and a CEO from München may have far more in common than one might think because they share a common set of values and beliefs.


A value based segmentation will have the following benefits:


  1. A more refined interpretation of the market research results, because the category perception will most likely be more dependent on beliefs than demography.

  2. Better strategic recommendations enabling more actionable insights.

  3. The advantage of using the segments in other research studies in other markets. (the segments can always be replicated via a reduced questionnaire).




To develop a solid value based segmentation a large set of values or beliefs are hypothesized followed by the creation of 30 – 40 different 5-point likert scale questions. The questions are asked to a representative sample of the population followed by an iterative cluster analysis. The goal of the cluster analysis is to group all the respondents in a given number of segments based on their answers to the 30 – 40 questions.


They key to get the most out of the value based segmentation is to ask the right questions and that the questions asked are relevant and cover all aspects of the respondents’ values and attitudes towards the category.

bottom of page