This paper reviews the contribution that a computer mouse can offer with respect to enhancing Computer Aided Personal Interviewing (CAPI). The paper briefly refers to the new advantages and problems that CAPI has introduced, and goes on to discuss the role of analogue scales in particular. The benefits of the mouse are then illustrated via a controlled test example - showing that there can be both quality and cost benefits in using a mouse. The paper then reviews other uses of mice with CAPI, including simple selection type questions and for more advanced two-dimensional positioning work. The paper goes on to indicate the criteria which indicate when it can be beneficial to use a mouse and when not. In general the paper draws the conclusion that mice are more suitable for longer interviews and those which require greater sensitivity in the data. The mouse is less helpful in shorter studies or where sensitivity is not an issue. Finally the paper reviews the other common non-keyboard data entry devices and assesses their potential in a market research context. This section concentrates particularly upon the benefits that should accrue to the market research industry over the next decade arising out of the growth and development of stylus based systems.
This paper consists of three parts. In the first part some developments and the present situation concerning advertising are discussed, in particular in relation with communication research. Manufacturers appear to still have a somewhat conventional attitude towards communication research, despite the developments in that area. In pretesting of communication instruments (advertisements, commercials, logos, packaging, etc.), the instruments are often judged on only one aspect, namely the creative and emotional one. Unfortunately nice is not always effective. The Triad model has been developed to supply more objective criteria; it combines emotional with objective elements. In the second part of the paper the Triad model is presented, which is developed by professor Poiesz of Tilburg University (Economic Psychology). The Triad model says communication instruments have to satisfy three conditions if the information in the instruments is to be processed in the consumers brain: motivation, ability and opportunity. The theoretical background and development of the model is presented. Hereafter the Triad model is discussed in detail. In the last part of the paper is mentioned how the Triad model is used in market research, i.e. communication research. Attention is paid to the operationalization of the model, the design of Triad research, en the (type of) results from Triad research. MarketResponse has linked the Triad model to CAGI (Computer Assisted Graphical Interviewing): the CAGI/Triad instrument. CAGI is a new method of computerized data collection, especially for communication research concerning visuals. This part contains an in-depth discussion of the possibilities and application of CAGI and the CAGI/Triad instrument in communication research.
The Danish Target Group Index provides the user with information at product level, and there is a need to improve media planning with information at brand level. If the existing questionnaire were to include information on more than a 1000 brands there would be an extreme bias in the respondents that would answer. Accordingly, it was decided in October 1990 to conduct an experiment with fusion of brand information database into the Danish Media Index. The results of the experiment were so promising that it was decided to go ahead with the actual fusion. The actual fusionend database was available on May 11991, and it has been very well received. After less than one year all development costs have been paid by the use of the database. The control of the fusion shows that the fused information on market share and demographic structure within target groups defined by brand usage or brand knowledge is very similar to the information in the original database.
The general trend towards decentralizing data processing is further emphasized by the specific requirements and structures of information existing in the various departments of a large company. What Steirerbrau needed most, besides technical parameters, was an information system offering standardized and yet flexible evaluations, possibilities for carrying out analyses in an interactive process and visualization of aggregated results for presentation purposes as well as simple operation. After having tested various systems the choice eventually was a system that proved its practicability when used with external data: the INF* ACT Workstation. In the sales area the simultaneous availability of internal and external data opens up interesting opportunities for analytical work. The situation in the area of production proved particularly complex: a myriad of most diverse parameters, data from various pre systems and lists, partly manual evaluations. In spite of technical difficulties, however, an integrated information system was created in this field also.
In the beginning 80's Henkel developed the idea of real international Marketing (Brands, Concepts), and Sales Managers started to define international concepts for Account Management. To support this strategy, Henkel launched in 1987 the project of an International Marketing Information System. The project scope was to define an integrated information system for decision support that would cover all user needs in terms of data, analysis and tools. Data would have various origins: internal (invoicing, planning) and external (retailer/outlet data collected) including market research data (trade panel, consumer panel, etc.). In terms of analysis functionality, the system would be as simple as possible to give access to the largest number of users, but would also support specific ad-hoc analysis, as well as simulations and forecasts. The first step was to give data access through a very open system of data analysis, Mainly ad-hoc oriented. This step developed skills in terms of data manipulation to support decision-making, from the user side as well as technical EDP support. The second steps, which started in 1991 was to stabilise this "open system", by harmonisation of data, analysis and tools on an European basis. This, of course, was a much more difficult step, that moved the system to a higher level, taking advantage of the exchange of many local experiences. This project was considered as a "must-do" project from a business point of view, for Maintaining a strong competitive position vs. Euro-Accounts and international competitors. This paper will summarise our project concept and implementation scenario, especially/organizational conditions if technical aspects / critical success factors applicable to such projects.
In Market Research, data which is multivariate in nature is often generated. For example, in a Brand Association study the associations of brands with a variety of attributes are measured and in a Usage and Attitudes survey information regarding the products used in different market segments is gained. In order to see the messages in the data clearly and to communicate them to others with maximum impact, it is desirable to display them in a visual form, e.g. by using graphs. However, the commonly-used graphical displays are univariate in nature, i.e. they deal with the dimensions one at a time. This means that some of the information is lost, and also many different graphs may be needed where there are a large number of dimensions. Multivariate mapping techniques to overcome this problem have been available for some years. However, they do not appear to be widely utilised in Market Research, partly because of non-accessibility (suitable computer software may not be immediately available and easy to use) and partly because of lack of confidence on the part of potential users. People may not be confident in using the techniques, and may also fear that the output will not be understood by their target ("non-specialist") audience. This paper shows the authors' experience with two techniques : Correspondence Analysis and Covariance Biplot (Principal Components Plot). The exposition is completely non-mathematical and shows our use of the techniques in some practical examples. In particular we describe some of the enhancements which we have found useful in communication to non- specialist audiences. The aim of the paper is to stimulate discussion with other market researchers, and hopefully to increase the appropriate use of these methods in the industry as a whole. It would also be useful if more user- friendly software might become available as a result of increased interest and use.
Knowledge-based systems (KBS), a branch of artificial intelligence, are designed to replicate the functions performed by a human expert. They enable a user to consult a computer system as he/she would consult an expert advisor to figure out how to solve a problem or to make a decision. Marketing decision-making is complex, situational and unstructured. When making decisions in an unstructured problem situation, marketing managers must use judgement and their general problem-solving abilities. Surprisingly little is known about decision- making by practicing managers, and both research results and theoretical models are underutilized. The complexity and uncertainty in marketing require more advanced, efficient and productive tools for decision-making - there is an obvious need for more sophisticated marketing decision-making. In this paper we will examine the main requirements for a successful use of computer- based systems in marketing decision-making. We will focus on strategic market management in a pharmaceutical industry, outline a conceptual framework for strategic market management and discuss the structure of its knowledge base. Our main argument is that support of a KBS for strategic market management promises to improve the decision-making capabilities of the managers.
Marketing information systems, tools, techniques, and technologies all hold the promise that better market information will become available more quickly in order to make better marketing decisions. In this paper it is argued that the use of information technology applied to Marketing Information Systems can indeed help, i.e. improve marketing decisions as long as implementation is managed in the right way. This paper deals with factors determining success or failure of applying information technology to sales and marketing. It is argued that the ultimate goal of using information technology is to improve marketing and sales productivity. A conceptual framework is provided which facilitates the choice of appropriate information systems and gives guidelines for successful implementation of marketing and sales productivity measures.
Finding the "right" price is one of the main problems of any marketing strategy. To solve this problem it is necessary to have information on the effects of price changes on demand, i.e. on the form of the price response function. To derive this information one needs in general data, methods and models. The main problem, the bottleneck, is the availability of data. This is still TRUE in the age of scanner data. Scanning has improved the conditions for pricing research considerably, but it can never fulfill all data needs. A few general problems are named: insufficient price variation: if the price does not vary, no price effect can be measured; the analysis is restricted to the range of observed values; experimental variation is not always possible. time and obsolescence: to collect sufficient data, scanner based or conventional, takes time (say at least 6 month) and during this time the situation can change. new products: no market data can be available for a new product before it has been launched onto the market. !n all these cases, where market data are not available or are not available fast enough or lack important conditions, laboratory data can be an alternative basis for Price research. The third case, getting-information on price response for new products, is the topic of this paper. In the following we will discuss three approaches for deriving price response functions by laboratory measurement. All these methods have been used within the framework of TESI, the testmarket simulator of GfK, Nuremberg. In particular these methods are simulated shop purchasing with experimental price variation, the TESI-price model for competitive brands, the TESI-price wheel for monadic testing of products. As mentioned above, price response measurement is necessary for improving pricing decisions. But this is not the only benefit that can be gained from such measurements. Price response measurement can also serve for analyzing competitive relations between brands measuring the strength of a brand, i.e. its goodwill or utility, on a monetary scale estimating the demand for a new product.
EDP ( Electronic Data Processing) is not a new thing in Market Research , analysing qualitative research data reaches back to the sixties, SPSS might be still in mind with some researchers. And also for analysing quantitative data, the first systems showed up during that time . But in the late seventies, early eighties, Personal computers started their triumphal run . A small, compact machine , made the work of host systems , which had room size just a few years ago . The mid eighties, and of course now the nineties, brought up even more interesting aspects : Windows, new processor types, LAN , mouse driven menus, spreadsheets , text systems , graphics just to name a few . The effects were tremendous : data management and analysing , project work and visual effects could be produced at the desk , where just a few years ago , large host systems and a large men power were needed . In July 1991 , Jacobs Suchard Germany started realisation of a new Marketing Research Information System . The objective was to increase efficiency of Market Research work , using new EDP technologies and methods. The new system called MRSS ( Market Research Support System ), it's concept and realisation , is subject of this essay.
The present service of market research institutes constists of providing extensive reports and charts for their clients on a periodical basis. The corresponding market research department within the client's company as well as the market researcher of the institute then have to extract all relevant data from this vast pool of information. In most cases the processing of the report data itself requires so much time that the actual market investigation and the development of future strategies are neglected. This paper presents a way how to relieve the market researcher of the time-consuming step of data analysis. In a common project between GfK AG, Nuernberg and the University of Erlangen- Nuernberg an expert system was developed in order to automatically create reports on the change in sales volume of a product category over time. This program systematically searches the data for significant developments by means of a sophisticated method of analysis. A multimedia system is employed to present the results combining verbal descriptions, (statistical) tables, and charts. Drawing from these results the inhouse market researcher is able to easily and swiftly compile "his" own expertise as the basis of a management report.