The Company Competitive Advantages: Santa Fe - Case Study Example

SANTA FE CASE Executive Summary A questionnaire was prepared and distributed to determine the psychographic profile and the factors that influence people to visit either Santa Fe Grill or Jose’s Southwestern Café. 405 respondents were found to have qualified for inclusion into the data. An exploratory factor analysis was used to determine the factors that influenced their decisions to choose one competitor over the other and a multiple regression was conducted to quantify and to predict the associations between these factors and the decision to choose one restaurant. It was found that the customers who picked Santa Fe tended to have more children staying in their homes. Also, Santa Fe’s advertising was found to be more memorable. The freshness and taste of the food, and the friendliness, speed of service and efficiency of staff were found to be better at Jose’s Southwestern Café. Therefore, it is recommended that Santa Fe consider increasing their staff training to provide better service, improving the menu to give a better taste to its customers and improving the supply chain to enable fresher ingredients to be used. Introduction In the preparation of this report, the researcher has conducted a questionnaire to determine Santa Fe’s competitive advantages, identify and address areas for future improvement, and suggest recommendations for further improvement. This report will begin with a comprehensive description of the methodology used throughout the survey. The report will then discuss the results found after statistical analysis of the survey data. Finally, the report will discuss the implications of the results for management of Santa Fe Grill and suggest some recommendations for the company’s management. Methodology The researcher began the study by randomly sampling a number of people at a location that was deemed to be representative of the general population. The researcher screened the people for suitability for inclusion into the research by asking several screening questions. The people who responded positively to the four screening questions were included in the survey. The screening questions asked pertained to whether the individual regularly dined out at casual dining restaurants, whether they have eaten at more than one Mexican restaurant in the last six months, whether they have a gross annual household income of $15,000 per year or more, and whether they have most recently eaten at either Santa Fe Grill or Jose’s Southwestern Café. These questions were matched with the normal customer profile at both Mexican restaurants and since the purpose of the survey was to question the average diner at Santa Fe Grill or its competitor, these questions were necessary to ensure that the questionnaire was given to Santa Fe’s target market. After screening, 405 respondents were identified. The respondents were then asked a series of questions which were aimed at finding out information about their lifestyle, perceptions of the restaurant, relationship with the restaurant, reasons for why the restaurant was chosen, and classification questions aimed at further determining the customer profile. The following information shows the types of data assessed for each factor: Lifestyle – Scale data assessed on a Likert scale with 1 being “strongly disagree” and 7 being “strongly agree”. Perceptions – Scale data assessed on a Likert scale with 1 being “strongly disagree” and 7 being “strongly agree”. Relationship - Scale data assessed on a Likert scale with 1 being “strongly disagree” and 7 being “strongly agree”. Selection – The questions in this section are ranked from 1 to 4 and each attribute must be ranked with a different number. Classification – Nominal data. After the results of the survey were obtained, these were entered into SPSS version 14.0 for Windows. All statistical analysis of thedata was performed using SPSS. When analyzing the data, since the first three factors were assessed using scale data, a reliability measure was used to test whether the replies to the questions were internally consistent. Cronbach’s alpha is typically used to test these relationships. Table 1 below displays the value of Cronbach’s alpha calculated with SPSS for the first three factors: Lifestyle Perceptions Relationship Cronbach’s alpha 0.482 0.550 0.913 Table 1. Calculated Cronbach’s alpha for each factor Although there are no statistically strong tests for values of Cronbach’s alpha, a usual rule of thumb used is that Cronbach’s alpha must be greater than 0.7 to show that the data is internally consistent. In the sample, only the relationship data showed internal consistency when measured against that benchmark while lifestyle and perceptions showed a figure less than 0.7. However, data that is not internally consistent is not necessarily wrong and may indicate that the respondent simply agree with some of the statements that constitute an aspect of lifestyle measured, for example, but not the others. As another example, there might not be a direct relationship between whether the respondent “eats balanced and nutritious meals” and whether the respondent is “self-confident about themselves and their future”. Having checked for internal consistency, an exploratory factor analysis with varimax rotation is performed. The purpose of conducting exploratory factor analysis is to determine the underlying structure when a large set of variables are presented. The factor analysis process is designed to reduce the number of variables into a smaller set of variables which adequately describe the data on the dimensions that are explored in the questionnaire. Varimax rotation is a specific technique within exploratory factor analysis that aims to determine a set of underlying factor variables that maximize the variances found within the data; in other words, the factors that have been extracted should have large and small loadings of the underlying variables and the varimax rotation process will make it easier to identify which variables are identified with the extracted factors (Johnson & Wichern, 2006). The exploratory factor analysis will be conducted within SPSS with the following procedures: The exploratory factor analysis is an iterative process in which various statistical characteristics in each step will be analysed and the underlying variables removed from consideration if they are found to violate any one of the statistical principles. The factor analysis will then be performed until the extracted results conform to the principles and these factors will then be described. In the first step of the factor analysis, the data is examined to see if any correlations between the variables included in the analysis are greater than 0.30. If a number of variables are found to be correlated with a correlation coefficient greater than 0.30, the analysis is found to be significant and it is therefore a valid procedure for the study. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy is used to estimate whether the overall sample was adequate for the purposes of factor analysis. A Kaiser-Meyer-Olkin Measure of Sampling Adequacy is deemed to be sufficient if the measure is found to be greater than 0.5 for the overall variables. The Measures of Sampling Adequacy are examined for each of the individual variables and the same standard is used – any variables that are found to have a measure lower than 0.5 are eliminated from the analysis since there is insufficient sample data for that variable to be rigorously analysed – and the factor analysis process is restarted until all individual Measures of Sampling Adequacy were found to be higher than 0.5. Bartlett’s Test of Sphericity is used to test the relationship between the variable. In essence, Bartlett’s Test of Sphericity tests the null hypothesis that the variables in the population correlation matrix are not correlated with one another. A high level of significance in Bartlett’s Test of Sphericity rejects the null hypothesis and concludes that the relationships between the variables are strong. If this is the case, then the factor analysis of the data is probably valid. The latent root criterion was used to determine the number of factors to be used in further analysis. Under the latent root criterion, any of those factors which have eigenvalues less than one are eliminated from further analysis. After the factors have been eliminated, the communalities for the variables are examined. Communalities represent the proportion of the variance in the original variables accounted for by the factor solution. The factors should explain at least half of each variable’s variance, so the communality value for each variable should be at least 0.5. Any variables that do not comply with this criterion are removed from analysis and the factor analysis is conducted once again excluding that variable. This is repeated until all the variables satisfy the communality requirement. The remaining variables are then checked for complex structure. Complex structure is found when one variable has high loadings of 0.4 or greater on more than one component. Any variable found to have a complex structure is removed from the analysis and after removal, the factor analysis is conducted once again without the excluded variable. Then, the remaining variables are checked to see if there are any single variable factors. Single variable factors are those variables that load on only one of the factors and if this is found, that variable is removed from the analysis and the factor analysis conducted without the excluded variable. Finally, the factor analysis is conducted to ensure that all the variables conform to the requirements listed out in the previous steps and when this is confirmed, the factor analysis is considered to be complete. Once the factors have been extracted, the number of variables has been reduced to the point that it has become feasible to conduct a multiple regression analysis on the data. In the case of Santa Fe, the dependent variable used for analysis will be the variable pertaining to which of the following Mexican restaurants the respondent has eaten at most recently. The purpose of the multiple regression analysis is therefore to determine which of the factors described in the questionnaire would have the strongest predictive power on whether they would be most likely to eat at Santa Fe Grill or its competitor, Jose’s Southwestern Café. Therefore, in the multiple regression analysis conducted in this survey, the multiple regression equation can be generally defined as: where Y is the dependent variable (whether the respondent ate at Santa Fe Grill or Jose’s Southwestern Café), x1, x2…, xn are the independent variables which correspond to the values calculated from the exploratory factor analysis process, β1, β2, …, βn are the coefficients of regression that are obtained from the multiple regression, and εn is the residual obtained from the multiple regression. The process of conducting multiple regression generates several test statistics that are of interest. The R2 of the model determines the goodness-of-fit of the model to the data, that is to say that the value of R2 shows the percentage of the model which is explained by the independent variables included. Low levels of R2 suggest that there could be other factors not tested for in the regression equation which explain the changes in the dependent variable whereas high levels of R2 implies that the model is a good fit for the data and therefore the independent variables considerably explain the variations in the dependent variable. The F-test is used to determine the validity of the model and should test the null hypothesis that the regression model proposed does not explain variations in the dependent variable. For individual predictors, the t-test is used to determine if the independent variable has a significant predictive effect on the dependent variable and if the calculated score from the t-test is too low, it is presumed that the specific independent variable may not have a significant effect on the dependent variable. Following the results of the multiple regression, a check will be performed to examine if the variables suffer from multicollinearity. Multicollinearity describes the situation in which some of the independent variables exhibit unacceptably high correlation with each other. Although the presence of multicollinearity does not necessarily mean that the multiple regression model itself should be invalidated, the presence of multicollinearity suggests that the individual predictors may not be a good predictor as a whole and small changes in the predictors may lead to large changes in the dependent variable. The test for multicollinearity that is used in this study is to calculate the variance inflation factor (VIF) where a VIF greater than 10 suggests that the model suffers from a multicollinearity problem. Results The exploratory factor analysis was conducted and resulted in the following variables: Three factors that account for 73% of the variance in the Lifestyle dimension. The first factor accounts for the questions “People come to me”, “friends come to me” and “self-confident”. The second factor accounts for the questions “buy new products” and “try new brands”. The third factor accounts for the questions “avoid fried foods” and “careful about what I eat”. The other questions in the questionnaire for this section were eliminated during the factor analysis process. Two factors that account for 83% of the variance in the Perceptions dimension. The first factor accounts for the questions “friendly employees”, “knowledgeable employees” and “speed of service”. The second factor accounts for the questions “fresh food” and “excellent food taste”. A single factor accounting for 80% of the variance in the Relationships dimension. All questions were included. A single factor accounting for 71% of the variance in the Selection dimension. The factor accounts for “price” and “food quality”. A single factor accounting for 66% of the variance in the Classification dimension. The factor accounts for “ad recall” and “number of children at home”. The factors were then regressed against the dependent variable – where was the last place that the respondent visited: Santa Fe Grill or San Jose Southwestern Café. The output from SPSS for this is attached as Appendix 1. The R2 of the model was 0.814 indicating that the model accounted for 81.4% of the variation. The F-statistic was 97.247(p-value=0.000) which suggests that the model is highly significant and therefore the model is very likely to describe the factors that influence the customers’ decisions on where to eat. Using a 2% significance interval, only the two factors on the Perceptions dimension and all the factors for Relationships, Selection, and Classification were found to be significant. The regression equation is therefore the following: Multicollinearity was found to be an issue with some of the variables with a VIF of greater than 5. Discussion and Recommendations The results from the regression suggest the following: The customers perception about the employees is important. Specifically, the employee’s friendliness, being knowledgeable and their ability to provide faster service are important. The respondents to this survey have generally found that San Jose Southwestern Café has been generally better at this than Santa Fe Grill and this had affected their decision to eat at the competitor’s restaurants. However, the customers who have gone to Santa Fe Grill have reported that they are generally satisfied with their restaurant experience and would definitely consider recommending it to their friends or making a return visit to Santa Fe Grill. The results from this study suggest that Santa Fe Grill may be better at servicing regular customers compared to San Jose Southwestern Café. This relationship was not as strong as the results for the Perception dimension and therefore Santa Fe Grill’s management should consider maintaining their relationships with their existing customers and encouraging them to return, although it is far more important to improve the perceived quality of service to attract new customers. The results on the Classification suggest that Santa Fe Grill generally attracts a family crowd with more than 2 children at home. The respondents have also found that they appear to have recalled seeing an advertisement in the last 60 days for Santa Fe Grill which suggests that the advertising measures taken by Santa Fe Grill have been more memorable than the competitors. The results found suggest that the customers who visit Santa Fe are generally families with two or more children living at home. This researcher believes that the decision to eat in Santa Fe Grill could be motivated more by pleasing the children when considering that the respondents remembered Santa Fe Grill’s advertisements better than the competitor. However, the study has a limitation – the questionnaire asked if the respondent had two or more children living at home who were under the age of 38. Further research should be done to determine the age group of the regular patrons of the restaurant.- For example, whether Santa Fe Grill appeals more to young children, teenagers or young adults. This research would be useful in helping the company design more targeted advertising to appeal to that particular age group. Those customers who have visited either Santa Fe Grill or San Jose Southwestern Café have generally found two factors in the Perceptions dimension to be particularly significant: the overall experience with the employees, and the freshness and taste of the food. Santa Fe Grill appears to be lagging behind its competitor on this measure and the results of the study showed that San Jose Southwestern Café was perceived to have better tasting and fresher food, and more efficient employees. This researcher therefore believes that Santa Fe Grill management should consider increasing their efforts in training their employees to work more efficiently and approach the customer with a friendlier attitude. Slightly more significant was the effect of the freshness and the taste of the food in which Santa Fe Grill is also lagging behind San Jose Southwestern Café. The implications for this finding is that management should consider ways to improve the taste of the food, this could include measures such as revamping the ingredients of the food, and improving supply chains so that the ingredients could be brought to the restaurant earlier and in a fresher state. The researcher notes that there are, however, several limitations with the regression model used to make these recommendations. Most notably is the presence of multicollinearity in some of the independent variables of the regression equation. This suggests that there is a non-trivial chance that the independent variables actually affect one another and while this doesn’t necessarily reduce the explanatory power of the model as a whole, it suggests that more research is needed to determine whether there are any statistically significant relationships between the independent variables used. Bibliography Johnson, R. A. & Wichern, D. W., 2006. Applied Multivariate Statistical Analysis. 5th ed. ed. New York: Prentice-Hall. Appendix 1 Read More