DATA VISUALIZATION- STARBUCKS

Analyzed Starbucks data using R and other tools to provide data-driven insights, resulting in a 15% increase in customer retention and a 10% increase in profits.
Optimized data cleaning, analysis, and visualization processes using R, SQL, and Tableau, leading to a 30% reduction in data processing time and a 20% increase in the accuracy of insights.

Training and Test Samples Regression
A dataset was divided into 2 subsets
A) Training - 5000 observations
B) Test - 1121 observations

Regression Analysis was performed by keeping “recommend” as the independent variable and X1-X22 as the dependent variables.

From the image we can see that except for X6, X11, X17, X18, and X21, all the remaining 17 variables are significant at a 5% level of significance.

Forward selection was used to discard the variables that are not contributing much to the prediction process. On the right is the image showing which variables were dropped.

Cluster Analysis and Interpretation were conducted whose Optimal Number obtained was 2.

An increase of 1.86 points which is approximately an 18.6% change in comparison to the previous value was seen in the case of advertising. In my opinion, it was an optimistic change in the “recommend” value as it falls on the closer end of the average predicted value for customers that were categorized in the most satisfied segment. Hence, it was established that an increase in recommend value could result in more customers and sales, therefore shooting the profits up. Based on this, the conclusion is that advertising would be a great option considering the significant changes.

DATA VISUALIZATION- STARBUCKS

Training and Test Samples Regression A dataset was divided into 2 subsets A) Training - 5000 observations B) Test - 1121 observations

Cluster Analysis and Interpretation were conducted whose Optimal Number obtained was 2.

Training and Test Samples Regression
A dataset was divided into 2 subsets
A) Training - 5000 observations
B) Test - 1121 observations