End-to-end retail sales analysis across three Lagos branches, covering six months of transaction data. The analysis logic is tool-neutral because the same business concerns are addressed in three tools: Power BI for the interactive dashboard, Python for visualization, and SQL for querying. As part of my portfolio, I examined six months' worth of retail sales data from a store in Lagos. Finding out which divisions and goods were genuinely generating profit rather than just income was the aim. The biggest surprise for me was seeing how much margin Clothing products had compared to Electronics, despite the latter dominating revenue. That gap is what provoked the opportunity matrix chart.

KPI Summary	Monthly Trend

Branch Performance	Opportunity Matrix

1. How much revenue and profit did the business generate across Jan–Jun 2025?
2. Which branch is underperforming — and is it a volume or product mix problem?
3. Which products deserve more investment based on margin, not just revenue?
4. What caused the February dip and March surge?
5. Where is the biggest untapped margin opportunity in the portfolio?

This repository contains an end-to-end analytic workflow from raw data to business recommendations
retail-sales-analysis/
│
├── data/
│   └── retail_sales_cleaned.xlsx      # Source: 294 transactions, 11 columns
│
├── sql/
│   └── retail_sales_analysis.sql      # 8 sections: QC → KPIs → trends →
│                                      # branches → categories → products →
│                                      # advanced (CTEs, window functions, ABC)
│
├── python/
│   └── visualizations.py              # 8 charts — run once to regenerate all visuals
│
├── powerbi/
│   ├── dashboard_setup_guide.md       # Step-by-step: load data, DAX measures, layout
│   └── retail_sales_dashboard.pbix    # Power BI file (open in Power BI Desktop)
│
├── visuals/                           # All chart outputs (PNG, 150 DPI)
│   ├── 01_kpi_cards.png
│   ├── 02_monthly_trend.png
│   ├── 03_branch_performance.png
│   ├── 04_category_donut.png
│   ├── 05_product_revenue.png
│   ├── 06_margin_by_product.png
│   ├── 07_stacked_branch_monthly.png
│   └── 08_opportunity_matrix.png
│
├── docs/
│   └── analysis_summary.md            # Written narrative: findings + recommendations
│
├── requirements.txt
├── .gitignore
└── README.md

No missing values. No duplicates. Date range: 2025-01-01 to 2025-06-30.

1. Top-line performance ₦200.5M revenue · ₦32.0M profit · 30.2% average gross margin · 294 transactions

2. Yaba is underperforming — but it's a product mix problem, not footfall Yaba has 79 transactions vs Ikeja's 129, but its average order value is also lower. The SQL branch deep-dive (Section 4c) shows Yaba is over-indexed in Groceries and has almost no Electronics presence — exactly where the revenue is concentrated.

3. Clothing has a 43% average margin but only 3.9% revenue share T-Shirt sits at 52.5% margin. Jeans at 42.9%. Both are in the "High Margin, Low Revenue" quadrant of the opportunity matrix — prime candidates for promotional push.

4. Rice and Laptop together = 72% of revenue at under 14% margin The business is heavily dependent on two thin-margin products. This is a concentration risk.

5. March revenue was ₦42.5M — 117% higher than February's ₦19.6M The steepest MoM swing in the dataset. The SQL MoM query (Section 3b) surfaces this. Understanding what drove it is the single highest-value question for the business.

Load data/retail_sales_cleaned.xlsx into any SQL client (DB Browser for SQLite, DBeaver, pgAdmin, etc.) then run sql/retail_sales_analysis.sql section by section.

# Clone
git clone https://github.com/YOUR_USERNAME/retail-sales-analysis.git
cd retail-sales-analysis

# Install dependencies
pip install -r requirements.txt

# Generate all charts
python python/visualizations.py

Open powerbi/retail_sales_dashboard.pbix in Power BI Desktop. If starting fresh, follow powerbi/dashboard_setup_guide.md — it includes every DAX measure and the full layout spec.

These are the four tools I use day-to-day. Every aggregation in Python has a SQL equivalent in the .sql file, and every chart has a Power BI counterpart in the dashboard, intentionally, to show the analysis logic is consistent across environments.

Data analyst with hands-on experience in SQL, Python, and Power BI. Seeking to contribute to a data-driven team where insights drive real business decisions

Open to data analyst, BI analyst, and junior analytics engineer roles.

• 📧 kayodeosuya@gmail.com
• 🐙 GitHub

⭐ If this project was useful, a star helps other analysts find it.