Method research
The methods of research of marketing remains focused on evaluating different methods of collection, evaluation and processing of data. A number of the methods have now established themselves as standards in scientific research.
Procedure
- Observation: Mystery shopping refers to the execution of sham or test purchases by trained employees of market research companies through covertly participating observation. It is used to check previously defined quality features on the basis of established international rules of conduct for market and opinion research . The main goals are to identify weak points in the company, but also to raise employees' awareness and motivation for customer-oriented behavior.
- Internet survey : The Internet is now an everyday item and is increasingly being used to collect data. The advantages of the online survey are the increased flexibility , location and time independence. In particular, the lack of direct contact between the interviewer and the respondent ensures an improved interview situation. Many market research companies use this instrument exclusively or as a supplement to traditional surveys. A possible disadvantage of Internet surveys is still the lack of or at least dubious representativeness . One tries to remedy this by setting up online panels that can adapt demographic characteristics, but the extent to which this also applies to psychographic characteristics remains uncertain.
- Conjoint measurement : The procedure tries to calculate the meaning of individual characteristics of a product or a service indirectly through the different preference for specific service bundles that differ with regard to certain characteristics.
Market segmentation
In order to be able to work the market more efficiently, it is helpful to subdivide the overall market and specifically address the resulting clusters with suitable products / services. The needs of these clearly defined customer groups can thus be satisfied with a certain marketing mix . In order to achieve the desired differentiation, the groups should be homogeneous in themselves, i. H. the interests and desires of the people in a group should be the same or similar. In order to be able to segment the market, it is necessary to define criteria for subdivision. Often these are socio-demographic, psychographic or purchasing behavior characteristics . Once the criteria have been established, statistical methods are used to form segments. Particularly suitable methods are cluster and discriminant analysis, as well as logistic regression and the Chi-Square-Automatic-Interaction-Detector (CHAID).
- Regression analysis
- Root cause analysis: To investigate which marketing variables have the greatest impact on sales
- Impact prognosis: Clarifies how the turnover changes if you change a variable, for example what happens to the turnover if the price increases or more money is spent on advertising.
- Time series analysis: This method is used to calculate trends. A typical question is: "How is demand for a product likely to develop over the next three years under the given circumstances?"
- Causal analysis: The causal analysis is a suitable instrument for very complex models where the interdependencies between individual dimensions have to be made visible.
As part of the causal analysis, dimensions are first measured, in which, for example, the consumer evaluates products, which are then calculated in the analysis to form complex dependency structures and finally displayed visually.
Forecasts
A distinction is made between short-term and long-term forecasts. While the short-term forecasts are mainly used for short-term changes and decisions (e.g. questions of optimal storage), the task of long-term forecasts is to uncover differences between targets and the expected development of this size and to identify consumer needs relatively early in order to identify any trends and to identify possible potential markets with promise.
literature
- G. Schweiger G., G. Schrattenecker: Advertising , 6th edition, Stuttgart: Lucius and Lucius 2005