Pizza Restaurant

Executive Summary

Our initial goal was to provide recommendations for business improvements based on the sales data. However, during our analysis, we identified several data discrepancies that need to be addressed before moving forward. Key issues include:

• Unusually Low Monthly Sales Variance: The sales data shows a surprisingly uniform monthly variance ranging from only $12 to $280, which is atypical for a retail operation and raises concerns about data accuracy.
• Limited Pizza Variety: The data indicates that only one type of pizza was sold in XL or XXL sizes, which is unusual for a business that typically offers a variety of options.
• Topping Distribution: It was found that only 50% of pizzas were sold with a cheese topping, and garlic was listed as the most popular topping. These patterns are inconsistent with standard consumer preferences and further suggest potential data integrity issues. Given these anomalies, it’s crucial to pause on making any business recommendations. I recommend scheduling a meeting with the business owner to review these findings and determine if they align with their expectations. If discrepancies are confirmed, we should collaborate with the IT department to investigate and rectify the data issues before proceeding with our analysis. This will ensure that any future recommendations are based on accurate and reliable data.

Project Summary

The initial objective of this project was to analyze sales trends over time, focusing on different types of pizzas down to the ingredient level. The insights gathered would have been utilized for forecasting ingredient usage and optimizing stocking levels.

Data Validation

During the data build-out, I noticed that the data appeared unusually uniform. This prompted a deeper investigation, revealing several anomalies that raised concerns about the accuracy of the data:

• Size Discrepancy: Sales of XL and XXL pizzas were exceptionally low, with only one type of pizza sold in these sizes.
• Cheese Topping Issue: Surprisingly, cheese, a common topping, was only present on 50% of the pizzas, which is highly irregular.

Data Preparation

In preparation for a more detailed analysis, I aimed to explore the topping distribution in depth. To achieve this, I:

• Duplicated the table and removed unnecessary rows.
• Split the ingredients by commas and unpivoted the data to tie it back to the main table.

This process revealed another inconsistency: garlic emerged as the most frequent topping, which was unexpected.

Next Steps

Given these discrepancies, the following actions are recommended:

• Business Consultation: Engage with the business owner to verify whether these data patterns were expected.
• IT Collaboration: If the data is found to be inaccurate, collaborate with IT to identify the source of the issue and implement necessary fixes.

Learnings

This experience reinforced the importance of trusting instincts when data appears abnormal. Continued diligence in investigating anomalies ensures the integrity of the analysis.

• 2024 7
• 2023 1