You are currently browsing the monthly archive for June 2005.
Apple is a very good innovator of new products and a leading product champion in the technology industry. They are known to introduce state-of-the-art products long before their rivals can even imagine about those ideas. But they do not follow-through by improving their processes and make these products available to the common user. Apple has to work on improving their business processes and find effective ways to deliver these products through efficient distribution channels and marketing approaches. The only way Apple can prevent the market share being taken away by its rivals who copy Apple’s innovations is to emerge as a process champion as well and work on reducing the cost of their products so that the customers do not turn away for a cheaper product. They should also patent all of its ideas so that they can share the revenues with their rivals who use Apple’s patented ideas.
The core principle of Apple is innovation. They promote creativity and innovation at all levels of their business. Employees are given incentives for coming up with new product ideas and given the freedom to implement prototypes at the expense of the company. They conduct workshops to develop the creative minds of employees and spreads the awarness that creativity and innovation are the keys to success in their industry. Apple is a creative product champion and comes up with new products way before its rivals. The competitors of Apple, like Dell, Acer, HP, IBM, follow the creative ideas of Apple and develop their own products using these ideas. They distribute these products more effectively and gain the market share with a more competitive pricing model and improved business processes.
The differential responses by a consumer in purchasing decisions is used by many firms in market segmentation and design advertising, products etc., based on consumer characteristics. Such segmentation usually involves purchasing rates as the major factor influencing the buyer’s behavior and targets such groups as “heavy users”, “brand loyals”, and similar consumer groups. This study, however, stresses on the importance of other parameters like “awareness”, “general liking”, and “intention to purchase” in defining consumer groups, along with purchase rates. The research is based on the hypothesis that a single segmenting variable like purchase rate is not enough to define consumer characteristics, but require an extended regression model, which includes associated parameters like customer’s awareness to the product features and competing products, his/her likes and dislikes on a particular product group and the degree of intention to purchase.
The major concepts used in the study are the consumer’s overt behavior, learning and cognition. The concept of overt behavior is measured using the variable – purchase rate. Learning can be measured by variables like advertising awareness, motivation, and various preconceptions. Cognitive ability of a consumer is measured using the variables such as his/her attitudes, comprehension, intention to purchase and level of satisfaction. In addition, the study also uses demographic variables related to the consumer’s household. Following is a comprehensive list of variables identified in this study:
Endogenous Variables: Attention, Perceptual Bias, Stimulus Ambiguity, motive, Overt Search, Attitude, Intention to Purchase, Brand Comprehension, Confidence, Purchase and Satisfaction Level.
Exogenous Variables: Exposure to Media, Word-of-Mouth Conversation Activity, Receipt of Samples/Coupons and Price Paid.
Socio-demographic Variables: Mealtime Group, Homemaking Skill, Leadership, Household Size, Size of City, Age of Housewife, Housewife’s Education, Housewife’s Employment, Household Income, Housewife’s Time-Per-Week in Kitchen, Size of Meal and Prior Purchase Behavior.
The study assumes multiple-variable relationship among criterion variables and includes demographic descriptors of the decision units. Explicit functional relationships are defined using flow charts and general functional equations relating the variables, leading to preliminary testing of linear specifications of the complex relationships among different elements of the system.
The research is designed using a statistical model, which evolves under certain benchmarking criteria. The starting point of the design is a linear model under the hypothesis that no segments exist. The design then expands the equation by searching for regression coefficient inequalities and suggests a causal and correlational study among the concepts that are identified for the study.
The study measures different variables using the natural rank order within each of the explanatory variables, all of which are discrete category measures. For example, “income” takes on values from 1 to 5, corresponding to increasing income levels. The variables are measured by splitting the values monotonically into smaller and smaller divisions.
The research was conducted on a consumer panel, which was established in a test market that consisted of 300,000 households and 1,000,000 people. A sample of 70,000, selected systematically (probabilistic systematic sampling + judgment) from area telephone directories, received recruitment letters and screening questionnaires. About 8,300 people responded and 1,100 were selected for the panel. After the screening questionnaire, members reported every 2 weeks on their purchases of fife food product classes identified for the study. The eating habits, involvement in homemaking, impulsiveness, gregariousness, time pressures, and concerns with nutrition were measured using mailed questionnaires. Respondents were also asked whether they had ever heard of brands in the related product class. And for each brand they had heard of, they were asked about the usage of these products, their attitudes towards that brand, importance of these attitudes in their purchase decisions, and the likelihood of their purchasing the brand in the next month. Another phase of telephone interviews was conducted over the telephone and only those families who completed all phases of data collection procedure were used for analysis. After screening out a handful of other data points due to missing data, produced a final sample of 693 households used for the study.
The results are provided as tables and cross-tabs and analyzed equation by equation. First, attitude is analyzed as a dependent variable and the results show a significant increase in the coefficient of determination of this segmented specification. The study finds that the coefficient of previous purchase divides for respondents whose intention level is smaller than 2 and is raised significantly when intention is greater than or equal to 2.0, indicating that previous purchase is a strong segmenting variable. Analysis of intention to purchase as a dependent variable show that communication activities like word-of-mouth, company at meal etc., have a significant impact on buyer’s behavior. Heavy purchasers had a lower attitude coefficient, and homemaking skills turned out to be significant for people with measured confidence of 4.0 or higher. Analysis of purchase as the dependent variable show that the sole endogenous variable had a statistically significant coefficient, but re-specification of the variable to include a broader set of exogenous variables, divided it into several categories of dummy variables, each of which implies positive relationships between intention and purchase.
The research ends with a discussion of results and a summary. Market segments for a convenience food product were defined in terms of parametric relationships between three criterion variables (attitude, intention to purchase, and purchase) and a variety of causal factors including endogenous behavioral measures and exogenous socioeconomic variables. Goodness-of-fit measures and tests of significance on coefficients were used to detect different interrelationships. The analysis show that the segment identified turned out to be composed of relationships among endogenous variables, a disappointment to the focus of study, because these variables are not subject to direct manipulation but rather are phenomena, which intervene in the decision process. Sociodemographic measures used in the study provided a basis for segment identification. The study, however, does not clearly summarize the results and say about whether the hypothesis is supported or not. It just concludes by indicating that more time and money are required for a segmentation study like this.
The research is well designed and focuses on the hypothesis. The sampling procedure and measurements are well suited for the field of study. But the results are not discussed well, and often tend to be too technical and difficult to understand. I would recommend interpreting the mathematical results to plain English so that it is easier to understand by companies/individuals who are looking at the results of similar research, to improve their market segmentation strategies. A longitudinal study, including the same panel, on the behavioral changes would also help in getting an accurate picture of the buyer behavior model.
