Customer Segmentation

It is common for marketers to break up their market into identifiable groups in which individuals are highly similar within each segment, while the segments are highly differentiated from one another. The intent of market segmentation is to determine on which segment to concentrate marketing efforts, as well as optimizing the marketing approach to each segment of interest by fine-tuning the approach to what is most relevant to each segment. For instance, segmentation can be used to uncover a part of the market that is not well-served with the panoply of brands that are currently available. More generally, segmentation can also be used to gain an understanding of a market, as it is highly unlikely that an entire market thinks and behaves alike.

There are typically four different segmentation research methods:

1. Behavioural segmentation: In this form of segmentation, consumers’ purchasing volume of a certain product/service, or their inclination to buy one brand over others, are measured using either primary or secondary research. Consumers are then split into different segments, such as a Pareto split (80/20), deciles, or terciles.

2. Attitudinal segmentation: This form of segmentation can only be achieved via primary research. Using a survey tool, respondents provide their level of agreement to attitudinal statements. This can be done in a traditional Likert scale rating exercise, or with more novel approaches such as Maxdiff and conjoint exercises. Using cluster analysis statistical techniques (such as K-Means or Latent-Class), respondents are then segmented in such a way that each segment holds highly similar views, while each segment holds a wide variation of opinions from the other segments.

3. Needs-based segmentation: As with attitudinal segmentation, this form of segmentation can only be achieved with primary research, and can be implemented via the usage of either Likert scale rating, a Maxdiff exercise, or a conjoint exercise. However, rather than measuring consumers’ psychographics, the market is segmented based on what consumers look for in the products/services they purchase. Cluster analysis statistical techniques would also be used here to segment the respondents.

4. A priori segmentation: Perhaps the most common form of segmentation used by marketers, a priori segmentation applies an already-deduced segmentation. For instance, one segment consists of single renters between the ages of 25 to 34 years old, while another consists of empty-nesters over the age of 55 years old. As with behavioural segmentation, this form of segmentation can be performed either via primary or secondary data.

Across all four types of segmentation, we would observe the differences across these segments on all other available information (such as demographics, media usage, geodemographic tools, etc.) to further optimize the marketing mix for each segment. Whereas a priori segmentation already has a pre-defined set of segments, the other three forms of segmentations can result in as many or as few segments as needed. It is therefore recommended to run different sets of segment solutions (2-segment solution, 3-segment solution, etc.) to assess which solution works best within the marketer’s needs and limitations.

In the case of attitudinal and needs-based segmentation, a segment classifier tool can be designed post-research. For instance, consumers can answer a few simple questions to be classified into a particular segment to receive a communication that best fits them.

© 2014 SmartPoint Research Inc. All rights reserved.