Consumer research is a process of collecting and analyzing consumer-related information, which can help companies understand consumer characteristics, needs, preferences, behaviors, satisfaction and loyalty, so as to formulate effective product, price, channel and promotion strategies. There are a variety of data sources for consumer research, such as questionnaires, in-depth interviews, focus groups, observational methods, and experimental methods. The data types of consumer research can be divided into quantitative data, which can be expressed in numbers, and qualitative data, which can be described in words. The data analysis of consumer research refers to the process of sorting, processing, summarizing, interpreting and evaluating the collected data. Its purpose is to find meaningful patterns, relationships, trends and differences from the data, so as to obtain valuable insights andConclusion。

The science of data analysis in consumer research is mainly reflected in the following aspects:

The goal of data analysis. The goal of data analysis is to refer to the specific problems and expected results of data analysis. It should be consistent with the goals and objectives of consumer research, and should also be in line with the company's strategy and marketing goals. The goal of data analysis should be clear, specific, measurable, achievable and meaningful, and it can be tested using the SMART principle. For example, the goal of a data analysis can be "to analyze consumers' perceptions, attitudes and willingness to buy a new product, as well as the key factors that affect these factors, to provide a basis for product pricing, packaging and promotion".

Methods of data analysis. The method of data analysis refers to the specific steps and techniques of data analysis, which should be selected according to the type, quantity, quality of the data and the objectives of the analysis. The methods of data analysis can be divided into descriptive analysis, exploratory analysis, inferential analysis, and predictive analysis, which are used to describe the basic characteristics of the data, discover the underlying structure of the data, test the assumptions and relationships of the data, and predict future changes in the data. The methods of data analysis can also be divided into statistical analysis, factor analysis, cluster analysis, discrimination analysis, regression analysis, association analysis, text analysis, sentiment analysis, etc., which are used to analyze the distribution, dimension, grouping, classification, relationship, rules, content, emotion, etc. The method of data analysis should be suitable, effective, reliable and scientific, and it can be guided by CRISP-DM models. For example, a data analysis method can be "descriptive statistical analysis of quantitative data from questionnaires, textual and emotional analysis of qualitative data from in-depth interviews, and then segmentation of consumers by factor analysis and cluster analysis, and prediction and interpretation of consumers' cognition, attitude and purchase intention by discriminant analysis and regression analysis".

Results of data analysis. The result of data analysis refers to the output and output of data analysis. It should be able to answer the goals and questions of data analysis, and it should also be able to provide support for the decision-making and actions of the enterprise. The results of data analysis should be clear, accurate, complete, useful and persuasive, and it can be presented using KISS principles. For example, the result of a data analysis can be "consumers' perception, attitude and willingness to buy new products are affected by the product's function, quality, price, packaging and word-of-mouth, of which function and quality are the most important factors, price and packaging are secondary factors, and word-of-mouth is the least important factor. Consumers can be divided into four segments: high-end loyalists, mid-range followers, low-end tenters, and non-interested, which account for 15%, 25%, 35%, and 25%, respectively. High-end loyalists have the highest awareness, attitude and willingness to buy new products, followed by low-end tries, middle-end followers again, and uninterested ones are the lowest. It is suggested that enterprises should adopt different marketing strategies for different market segments. For example, for high-end loyal people, they can adopt high-price and high-quality strategies, emphasizing the function and quality of products, and improving the popularity and loyalty of products; for middle-end followers, they can adopt medium-price and medium-quality strategies to balance the function, quality, price and packaging of products, and increase the attractiveness and competitiveness of products. For low-end try, the strategy of low price and low quality can be adopted to reduce the price and cost of the product and expand the coverage and trial of the product; for those who are not interested, the strategy of giving up can be adopted, no longer investing resources and energy, and diverting attention and focus".

Conclusion

The data analysis of consumer research is a science, which requires the use of scientific methods and techniques to extract valuable insights from the vast amount of data. This paper introduces the basic principles and steps of data analysis of consumer research, as well as some commonly used data analysis tools and methods. This article also combines some consumer research cases provided by Shangpu Consulting to show how to use the science of data analysis to provide useful suggestions and solutions for enterprises. It is hoped that this paper can be helpful to the practice and understanding of data analysis of consumer research.