Customer segmentation is one of the most discussed concepts in customer-relationship management. There have always been customers with different needs and attitudes. Segmentation is intended to support organisations in achieving differentiation through customer-relationship management that is optimally tailored to the respective segment.
Customer segmentation is understood as splitting the customer base into homogeneous groups that are heterogeneous among themselves. It forms the basis of differentiated customer-relationship management.
In practice, however, customer-relationship management is still implemented according to the ‘watering-can principle’ or the ‘shotgun principle’. Customers are offered the same (cheap) experience, and whoever takes it up may have won or lost. Customer-relationship management is carried out undifferentiatedly, and usually the efficiency for the organisation and the enthusiasm for the customers suffer under this approach. But why is that?
In our view, customer segmentation is understood too one-sidedly. Customer segmentation has a dual orientation. This duality needs to be understood precisely. A customer-centric organisation should fundamentally use a segmentation for customer-relationship management. So segmentation, as set out in the many explanations, is an essential element of customer centricity. Here, customer centricity is oriented towards the value for the customer and for the organisation (customer-firm value). Customer-relationship management should make investments in the customer segments that can enable the organisation the greatest increase in value in the future. This premise is the basis for the further explanations. Segmentation is not done because it is possible (in the sense of: because customers are different), but because segmentation supports increasing the customer-firm value.
Market capture as one side of customer segmentation
As already explained, the acquisition and use of customer insights are important elements of customer centricity. With reference to customer segmentation, this is referred to as market capture.
Market capture denotes the goal of understanding customer behaviour, in the sense of customer insights, as well as possible in relation to the organisation’s objective.
Market capture can be used to delimit the relevant market, determine the relevant market segments, and find market gaps. So customer data, or the acquisition of customer insights, is in the foreground of this part of customer segmentation. In the past, it was usually sufficient to draw on simple customer data such as age and gender to identify different market segments. There were clear brand preferences of the genders, and age correlated with purchasing power in most offering categories. By using age and gender (sociodemographic data), organisations could differentiate themselves from the competition and increase their value. Today, age and gender usually no longer help to determine stable market segments. More and more women buy supposedly ‘male’ car brands, and more and more men buy supposedly ‘female’ ones. Some young people spend all their money on cars (especially in Aargau), and more and more older people forgo a car. There are still differences between age and gender, but these are increasingly blurring. In addition, most organisations have the possibility of evaluating sociodemographic data and using it for customer-relationship management. As a result, studies show that offerings and brands increasingly resemble each other, also because the same data underlay the respective decisions. If everyone uses the same data, the decisions, or the results, will be more or less identical. Meanwhile, for market capture, there is the possibility of using higher-order segmentation approaches based on customer behaviour and customer-firm value. These segmentation criteria are helpful for strengthening differentiation and value increase. As an example, the difference between NPS® (attitude) and actual referral (behaviour) can be identified. An organisation that knew, for every individual customer, how often this customer recommends — and, even better, to whom they recommend — its own organisation could use this for more efficient customer-relationship management. At this point, a big ‘but’ is appropriate. The vast majority of organisations do not know this, or have no data regarding the actual referral of every individual customer.
Market cultivation as one side of customer segmentation
The second side (action side) of customer segmentation is market cultivation.
Market cultivation denotes the use of the developed segmentation(s) for customer-relationship management through the deployment of the instruments.
In contrast to market capture, it is not the acquisition of customer insights but their use that is at the centre of market cultivation. As with the example of the referral made, it would be valuable for an organisation to understand which customer actually refers. In practice, this information is usually not available at an individual level. So market cultivation, in contrast to market capture, has to limit itself to the existing customer data in the organisation.
The high importance of customer-data management for improving customer centricity becomes clear. If an organisation only has sociodemographic data about customers, then only exactly this can be used for market cultivation. Although the organisation can, in the context of market capture — e.g. by means of a customer survey — determine the customer purchase process or the actual number of referrals, these insights can only be used in market cultivation with very great effort and with a usually too high error term. The error term arises because only a small proportion of customers participate in a customer survey. Afterwards, ‘crutches’ such as age need to be determined that connect a different strength of referral. The actually high data quality on the market-capture side is lost again in the attempt, due to the missing individual data about every customer on the market-cultivation side. The explanations show the central problem of customer segmentation in practice. In our view, too many of those responsible are not aware of this problem at all. Another challenge that is closely connected with customer segmentation, but likewise does not always seem to be clear, is the requirement for value increase. While the vast majority of organisations still don’t use any segmentation in everyday business, other organisations develop too many segments. This inevitably leads to high complexity in customer-relationship management and the challenge of working on the individual defined segments in such a way that the high effort also leads to a profit increase. The number of segments has a big influence on the costs for content creation. For campaigns, offerings and touchpoints to be aligned to the respective segments, investments are necessary. Too large a number of segments may lead to a value increase for customers, but at the same time to value destruction for the organisation. So I recommend, at the beginning, considering how many segments make sense at all from a financial point of view, before segmenting too ‘finely’. Every organisation must decide which segmentation strategy should be used. Ideas are now circulating about addressing the customer as individually as possible, or using individual segmentation. But this presupposes high competencies in the area of customer-data acquisition and use, which the vast majority of organisations do not yet possess. So in practice, work is usually done with two to four segments — or with no segmentation at all.
The segmentation process
The process is shown in the figure above.
It is important, as a first step, to determine the objective precisely. Should communication be improved, should customer acquisition be improved, etc. Depending on the objective, different segmentations may be advisable. Customer-centric organisations usually do not have just one segmentation; rather, several segmentations are in use for the respective area of application. It is important that all are connected with one another via the dimension of value development. For example, it is not helpful if the communication department segments on the basis of lifestyles and there is no understanding of customer value. On the other hand, sales uses only a customer-value segmentation and completely ignores customers’ lived reality in its activities. After delimiting the relevant market, the relevant segmentation criteria need to be determined against the backdrop of the requirements for market capture and market cultivation. Then the relevant segments need to be determined. The information gained about the individual segments needs to be used for decision-making in the context of market capture. Finally, the individual segments are worked on based on the decisions made.
Again and again we hear that the greatest challenge in customer segmentation lies in measurability. Why does this impression persist so stubbornly? By now, everything and everyone is measurable. Many measurements, such as online customer surveys and online analytics of one’s own website, are not even expensive. The actual challenge in segmentation, against the backdrop of multi-optional purchasing behaviour, lies in behavioural relevance. Does customer segmentation really allow me to understand and influence my customers’ future behaviour optimally? For that, the data must be available at an individual level, and this can quickly violate the requirement of economic viability (although many possibilities now exist to reduce the costs, which are, however, hardly used). Besides these requirements, when selecting the segmentation it is above all important to consider that market capture and market cultivation need to be integrated in a goal-appropriate way. The actionability of the customer segmentation is at the centre of the assessment. In practice, too many organisations still segment with existing data on sociodemographics, offering purchase, etc. So, as a first step, it is important to have ‘better’ data about customers, since in strongly differentiated markets the customer (not in the sense of age and gender) should be moved to the centre of the segmentation.
