This paper details how to marry the science of segmentation with the art of understanding the organization's business to deliver actionable and usable insights. It discusses why science or statistics alone are necessary but not sufficient to provide an actionable segmentation solution. The paper also demonstrates an alternate analytical approach to avoid some of the common pitfalls of segmentation studies that arise through the use of standard factor-cluster approaches. Though the paper focuses on segmentation, it has broader application by demonstrating best practice in how to balance the use of statistical techniques to deliver usable insights. Successful insight activation through an effective, continued dialogue between research and the business, is also highlighted in the paper.
New products are the lifeblood of marketing. Even the most successful products have a life stage, and so marketing companies are continually looking to develop new products and re-stage existing ones. It is almost universally true that the best performing companies are the ones that have the strongest track record of bringing new products to market. It is also true to say that new product development is one of the most exciting and interesting areas, both for the marketer as well as the researcher. In fact the authors believe that it is sometimes too interesting and can produce an over-concentration of effort in less profitable ventures. The inherent profitability of many marketing companies is heavily dependent on the success of a relatively small number of well-established brands. While it is certainly correct that they should always be looking for new opportunities, the less glamorous task of looking after these profitable brands will pay back considerably more in the short to medium term. It can often be the case that the brighter and more dynamic marketing executives are focused on the more stimulating area of new product development to the detriment of some existing brands and also short-term profitability. Having said this, however, there is a great need for product development within companies, and it is an area in which they constantly look to market research to help them. The rest of the chapter looks at the various stages of the product development process and the role of market research within them.
This paper deals with brand extensions. It looks at one of the largest brands in the UK and how it has been extended in four separate ways. It looks at the rationale for each extension and the problems that each one faced. Using the research conducted at the time, and the subsequent results in the market place the paper tries to draw some conclusions about the conditions necessary for successful extensions. It also examines the impact that such extensions can have for the parent brand.
This paper looks at the growth of computer interviewing over recent years. It discusses our experience with respect to computer interviews and looks at where it has proved successful and where problems have arisen. The paper concentrates on the use of computers for face to face interviewing and does not discuss in any detail the more popular use of computers for telephone interviewing (CATI). The first section of the paper deals with some of the more practical issues involved in computer interviewing such as the implications for research costs, timing and training of interviewers. It will also look at how computer interviewing can be usefully used for multi-national surveys. The second section of the paper is concerned with more sophisticated uses of computer interviewing, where advantages to the user come not from consideration of costs etc. but the ability to conduct types of interview that would not be possible using traditional pencil and paper methodologies. Throughout the paper reference is made to a variety of software that we have used in the course of our work on computer interviewing. The paper is in no way an attempt to evaluate these different programs scientifically but rather reports on our experience with them to date.
A multitude of predictive systems have been developed for FMCG products, many with a fairly reliable track record. The same cannot be said for durables. In this paper, the authors discuss a range of factors that differentiate both the research process and the predictive systems in the two areas and suggest that in many cases the FMCG models cannot simply be adapted for durables. The paper concludes with some suggestions for successful durable modelling.
This paper has attempted to demonstrate the benefits of integrating two quite distinct modelling processes, each based on a micro-modelling philosophy. In particular, it has shown how early quasi test-market volume predictions (such as provided by a Microtest type analysis) can be modelled at a variety of prices in addition to the core price contained within the concept proposition mix, and how valuable this can be to identify optimal pricing points. At the same time, the value of MicroTest type trial and volume data can make traditional Brand/Price Trade-Off modelling considerably more sensitive.
In this paper, following a review of the historical development of research methods for predicting volume sales and brand shares of new products, a new model (MicroTest) is described which uses information gathered in a concept/product test for volume prediction. The model makes use of brand related parameters (such as advertising and distribution), altitudinal predispositions (e.g. experimentalism), and circumstantial factors as input to the model, and these are described, together with the method of integrating these for predicting at the individual respondent level. Individual results are then accumulated across a sample of individuals, and grossed up to provide national sales estimates. The paper describes the various development stages undergone in the construction of the model, and the techniques used to assist this process. In particular, the way in which Artificial Intelligence techniques such as rule induction was used is discussed. Finally, the paper discusses the way in which the basic model may be extended, and some recent work which used the model to generate a measure of cumulative penetration.