We all know companies hold more customer data than ever before, but how can we successfully marry this to primary research? What's the best way to map psychographic data (attitudes, behaviours, responses, etc.) to the data you reliably have for all your customers or prospects? With a mature customer base and a propensity model that had outlived its usefulness, Sky's reward-based 'Introduce A Friend' referral scheme had achieved relatively low recent uptake. Sky wished to understand if there were customers for whom different propositions and contact strategies would be more successful. A 'reverse' segmentation of its customer base, mapping survey data onto existing fields, enabled Sky to tailor and target its referral scheme to specific customer groups, and thereby open up new headroom.
This paper describes a new approach for analyzing customer behavior from data collected in customer relationship management systems. The very short response times of the new approach allows one to browse interactively through the variables describing a customer. This allows discovery and understanding of behavior patterns with little effort. The same approach can also provide customer segmentation into segments of similar behavior without the need to define criteria for similarity in advance. Thirdly the approach is able to make real time predictions about user behavior which can be used to personalize web pages, make caller specific offers in call centers, or to target campaigns. A case study from an online computer magazine is presented.
This chapter will focus on the application of modelling with data and research findings (often otherwise collected) to enable businesses and organisations to improve their effectiveness. Underlying methodology will be touched on where it is considered to influence the interpretation and application of results. The world of packaged goods has traditionally created the data sets required in modelling so it is not surprising that many of the modelling approaches have been developed for and by the packaged goods industries. Modelling has become increasingly popular as the advances in computers have brought the data and software tools to a wider audience. It has also been seen as the answer to How do I make sense of all these data?.
The word panel describes a continuous collection of identical information from a sample which represents a segment of a population to study. It is by definition permanent and repetitive. There are as many varieties of panels as there are different populations and methods for measuring them. The research objective must determine which population will be studied and the nature of the information collected. For example, we can have doctor panels which measure their prescription orders, user panels which measure specific products or services, measures of the media audience and of course shop panels. In this chapter, we will take a particular interest in those studies generally known as consumer panels. Their objective is to specifically record the everyday purchases of different households. These panels were developed many years ago (since 1948 in the United Kingdom, 1962 in France) in most developed countries and are part of the basic tool kit for professionals working in marketing research and the marketing of mass-consumption products. They are syndicated studies, covering a large number of products and/or services, the results of which are shared between all the players within a given market. This characteristic of covering many customers and many product categories is basic to the very idea of a panel.
Non-classical segmentations/Cluster Analysis techniques based on structures of needs have proven very useful in such a context, not only for identifying market segments/target groups, but also for observing their evolution, comparing their maturation processes on the various national markets. The authors present and analyse a few illustrations of such an approach, covering product fields such as Cars or Personal Computers.
Segmentation is one of several multivariate analysis methods which can be applied to market research data; it can be carried out in several ways. This paper discusses briefly some of the possible techniques. The purposes of segmentation which we have experience of are the practical ones of defining marketing objectives, planning campaigns and briefing creatives. We are not interested in statistical 'explanation' as such, but in helping people, often enumerate, to take better decisions. We find segmentation can do this successfully. In discussing techniques we concentrate not on mathematics or programming but on input strategy. We have used clusters created in four different ways, depending on whether we use the whole sample or a special group, and on whether we input descriptive or product data. The example given is from the UK convenience food market, where product data was used on a large sample to create nine clusters. These are described with examples and the use of such data is discussed.
In the broadest sense of the term, market segmentation as a strategy has been with us as long as marketing itself. Essentially, it hinges on the principle that consumers are not all alike, and utilises that principle in the sale of merchandise. Until about ten years ago, segmentation research was based largely upon simple cross classification by demographic characteristics or purchase behaviour. During the 1960's, however, major research interest and emphasis has focussed upon new, more powerful, and hopefully more meaningful ways of partitioning markets based upon attitudes held by consumers.
In order to prepare this paper, we have covered a considerable bibliography, including the conferences at the present seminar. There exist considerable redundancies, and present techniques seem already somehow stabilized, though wide fields of research remain open to mathematical improvements.
If we are going to be successful in our marketing it is necessary to achieve certain standards of profitability. The further removed the customer is from the producing company, the more difficult it is to promote products and services to him without waste . We want, therefore with segmentation analysis, to reduce waste because our efforts in product design, promotional effort and distribution are directed at a clearly defined segment of the market. In marketing we should know that it is impossible to be all things to all men.
The aim of this contribution is the evaluation of clustering techniques from the viewpoint of practical marketing application. Clustering techniques (segmentation methods, grouping methods) are procedures that search for and detect natural groupings of objects (eg. persons) which are described by their values on variables. The requirements that must be met by such methods and the criteria for their evaluation which will be discussed have been shown to be important in the repeated application to practical marketing problems.
As stated in the programme, the aim of this Seminar was to illustrate and discuss recent trends and experiences in segmentation and taxonomic techniques.