A data-driven report developed by Carllos Watts-Nogueira as part of the Artificial Intelligence & Machine Learning program at the University of San Diego / Fullstack Academy (Section: 2504-FTB-CT-AIM-PT), this capstone project translates raw sales data into strategic insights for executive decision-making.
-
- Start date: May/2025
-
- End Date: June/2025
Objective:
Analyze the fourth-quarter sales performance of AAL, a prominent clothing brand in Australia, across states and demographics. This includes revenue insights, time-of-day breakdowns, and strategic recommendations for future growth.
AAL, a leading clothing brand in Australia, is undergoing aggressive expansion and seeks data-driven insights to optimize its sales strategy.
This project analyzes Q4 sales performance across states and age demographics, with the goal of identifying high-performing regions and addressing underperforming markets.
Through detailed data wrangling, normalization, and advanced visualization techniques, I deliver actionable recommendations to support strategic planning.
Dataset: AusApparalSales4thQrt2020.csv
A synthetic sales dataset representing fourth-quarter transactions across Australian states segmented by demographic group and time-of-day.
Tools Used:
- Python 3
- JupyterLab Notebook
- Pandas, NumPy, SciPy
- Seaborn, Matplotlib
Key Highlights:
- State-wise revenue comparison
- Demographic trends (Kids, Women, Men, Seniors)
- Time-based sales performance (peak vs. off-peak)
- Normalization techniques for clean data analysis
- Strategic dashboard for executive decision-making
Bonus Features:
- Box plots & distplots for statistical behavior
- Markdown-enhanced reporting for clarity
- Actionable recommendations to optimize regional performance
- Aggregated temporal analysis (daily/weekly/monthly/quarterly breakdowns)
- GroupBy-driven demographic insights for performance comparison