Objective: You've been asked to share an explanatory report providing a data-driven strategy for opening their first coffee shop. The investors expressed interest in the following areas, but are open to any additional insights and recommendations you can provide:
- Target audience: What type of customer should we target, and what are their preferences?
- Product offering: What types of coffee beans and drinks should we offer?
- Pricing strategy: How can we align prices with customer value perception.
Data: This data comes from the James Hoffmann YouTube channel where he did a taste test experiment in which participants were shipped 4 different coffees (in the form of frozen coffee extracts), asked to rate the coffees, and answer a few more questions. There were 4,000+ participants from all over the US.
Data Cleaning: After reviewing the questions and responses, I decided to retain certain information, particularly the coffee ratings. Additionally, many questions had responses represented by true or false, which I standardized to 0 for false and 1 for true. I also removed some unnecessary columns.
Subsequently, I unpivoted the data to condense the number of columns and increase the number of rows, where one column represents the survey question, and another column represents the response.
To organize the data systematically, I merged two tables: the survey table and the data dictionary containing the question column. This allowed us to associate each survey response with its corresponding survey section and question number, facilitating a more disciplined arrangement of the data.
Approach: In this Great American Coffee Taste Test, my primary goals was to construct one-page summary that serves as a visual narrative, capturing all Key Metrics overarching responses in Great American Coffee Taste Test. I aimed to spotlight key areas deserving of attention and further analysis.
Let's explore how the project is structured, its building blocks, and the methodology behind this project.
First and foremost, what is this coffee types in this dataset.
Coffee A - Light roasted coffee. It was from Kenya.
Coffee B - Medium roasted coffee. It was a blend.
Coffee C - Dark roasted coffee. It was blend.
Coffee D - Single Estate coffee. Strongly fermented coffee. Unusual variety from Columbia
π‘Key Insights:
Now, based on above analysis will see the business requirements along with answers.
π‘Target audience: What type of customer should we target, and what are their preferences?
Mainly we focus on age group 25-44 age group because 75% of audience in USA preferred to drink coffee than other age groups. Majorly male audience preferred to drink coffee than females. Their preference coffee drink is poureover followed by Latte and Regular drip coffee.
π‘Product offering: What types of coffee beans and drinks should we offer?
As per analysis, we may focus on offering coffee D and coffee A had a vote share of 34.27% and 20.24% accordingly.
π‘Pricing strategy: How can we align prices with customer value perception?
As per analysis of what is the most you've ever paid for a cup of coffee and what is the most you'd ever be willing to pay for a cup of coffee, we can align ideal price for cup of coffee is in between $8 to $10.
The Interactivity
To build the game and the dashboard, I used Power BI Native Visuals and App source visuals, Field Parameters, and (lots of!) Conditional Formatting. In the dashboard, the Matrix chart is only responding to the slicer. In the Matrix, the user can drill up and down between the "questions" and the "response categories".
