Behavioural segmentation systems: A perspective
- PDF / 256,990 Bytes
- 8 Pages / 598.216 x 767.28 pts Page_size
- 48 Downloads / 189 Views
Brian Birkhead is the Managing Director of Macon Consulting Ltd, a company he founded in 1992. After gaining his MSc, he was a senior research fellow at the University of Warwick and in 1983 moved to a teaching and research position in the Department of Statistical Science at University College London. Brian’s first commercial post was as Manager of Technology for the TSB Group from 1987–89.
Abstract Customer Segmentation is a fundamental tool in relationship marketing, providing a strategic planning framework, which allows tailoring based on differentiating pieces of data and information. There are many forms of segmentation and several techniques employed in their derivation. This paper discusses behavioural segmentation and clustering methods specifically, which are becoming more commonly employed in the industry. Some of the beneficial applications of a behavioural solution are discussed along with the relative merits of the clustering approach compared with more exact, rule-based systems.
Brian Birkhead Managing Director, Macon Consulting Ltd, 70 Grafton Way, London W1P 5LE. Tel: ⫹44 (0)20 7554 5100; Fax: ⫹44 (0)20 7383 4522; e-mail: brian.birkhead @maconconsulting.com
EXACT VERSUS CLUSTER-BASED SYSTEMS The prevalence of database warehouses and the resultant quality and scope of customer product and transactional data have led to an increase in the development of behavioural segmentation systems using multivariate methods (such as cluster analysis). These systems are characterised by their use of a large number of behaviour-related data fields to define a relatively small number of roughly homogeneous segments, each of which is distinct in terms of its dominant characteristics — a distinction which permits the development of segment-specific marketing strategies. Such systems are quite different from traditional segmentation schemes which are largely intuitively (rather than data) led, and are defined in terms of very few
Henry Stewart Publications 1350-2328 (2001)
Vol. 8, 2, 105–112
data fields. (Recency-Frequency-Value (RFV) systems are typical of this class.) — Systems like RFV allow exact allocation to segments of customers who share the same values of the small number of defining fields — Clustering systems are based on a large set of behavioural fields, and individuals within the same cluster are grouped together on similarity. As the number of fields considered important for segmentation increases, the number of segments in an exact system increases inordinately quickly to make this approach infeasible. For example, a system based on three fields (each with, say, three values) creates 27 segments, while one with only seven fields (each with three values) creates a system with over 2,000 segments! In contrast, most
Journal of Database Marketing
105
Birkhead
cluster-based systems use many more than ten fields, often to create fewer than ten segments. This highlights a severe limitation of traditional exact systems. If ten fields were indeed necessary to capture customer behaviour and only three
Data Loading...
