Sie haben Fragen?

Gerne beraten wir Sie telefonisch oder per E-Mail.

+49 8142 5069-0

Sie haben Fragen?

Gerne beraten wir Sie telefonisch oder per E-Mail.

+49 8142 5069-0


Dynamic Segmentation in the Travel Industry


Even if consumers know what they want, they are often unable to select the product option that promises the highest expected utility, simply because their individual product conceptualization is incompatible with the prod­uct-centred "language" the provider is employing. One consequence of this confusion is a cautious buying be­haviour most prominent in industries such as tourism, financial services or telecommunications. This paper pre­sents a conceptual framework and a pragmatic imple­mentation of an online matching algorithm based on an efficient self-correcting customer profiling technique that is empirically superior to many other approaches. The essence of the algorithm is that matching per se does not rely on a complex and industry-specific match­ing equation, but builds on a rather simple feedback bop of customer profile, selected offer and resulting satisfac­tion that can be easily transferred to many other indus­tries where the product variety and/or complexity goes beyond the customer's cognitive capacity. In those cases, the customers are often unable to correlate the formal product features that are advertised by the provider with his individual needs that shall be satisfied with the offer.

Basics and objectives of customer segmentation

Segmenting customers involves identifying homogene­ous groups from the totality of (potential) customers and treating them in accordance with their needs and cus­tomer value. The general objective is to more profitably address and take care of customers over the long term, for example by reducing the time and money expended on marketing, sales or service as well as losses due to non-selective advertising.

In order to do this, segmentations must fulfill two key prerequisites: firstly, they must identify variables that define to what segment a specific customer belongs to (="classification"). Secondly, these segments must have practical implications regarding what specific kind of marketing mix is most attractive to a certain customer segment (="translation").

Classic segmentation procedures and their limitations

Customers can be segmented in along a wide variety of dimensions (for instance, customer value, sociodemo­graphics. psychographics. or geography). There is no "correct" segmentation solution. In practice, however, the various solutions definitely differ in terms of how useful they are. The easier the classification variables can be measured and the clearer the practical implica­tions that can be derived from the fact that a certain customer belongs to a specific segment, the better the segmentation. In fact, the key empirical basis of mean­ingful segmentation solutions is that there has to be a reliable link between the classification (independent variable) and the measures to be deduced from it (dependent variable), since purely circular segmenta­tions ("We'll best reach customers who listen to the radio a lot (= classification) via radio advertising (=translation into measures)") are just as meaningless as non-differen­tiating measures ("Customers in every segment read the same newspapers - so we'll reach every segment equally well by using an advert in a specific newspaper").

Dieser Text stammt aus einem Fachartikel, den Sie im Folgenden kostenlos per E-Mail erhalten können.


ESOMAR World Research Conference
Umfang: 11 Seiten

Bitte senden Sie mir das Dokument per E-Mail.