How to Showcase Data Analytics Projects on GitHub Properly

retail-sales-analysis/
|-- README.md
|-- requirements.txt
|-- .gitignore
|-- data/
|   |-- README.md           # source link and licence
|   `-- sample_orders.csv   # small sample only
|-- sql/
|   |-- 01_monthly_revenue.sql
|   |-- 02_repeat_customers.sql
|   `-- 03_top_categories.sql
|-- notebooks/
|   |-- 01_cleaning.ipynb
|   `-- 02_analysis.ipynb
|-- images/
|   |-- monthly_revenue.png
|   `-- dashboard.png
`-- dashboard/
    `-- sales_dashboard.pbix

# Python and Jupyter
__pycache__/
.ipynb_checkpoints/
.venv/

# Data: keep only small samples in the repo
data/raw/
*.parquet

# Secrets
.env
*.key

# OS files
.DS_Store
Thumbs.db

# Retail Sales Analysis

Which product categories drive repeat purchases, and when do
sales peak? Analysis of a public retail dataset using SQL,
Python and Power BI.

## Key findings
- Two categories account for most repeat purchases
- Sales peak in the festive season and on weekends
- Repeat rate is lower for customers whose first order was discounted

![Monthly revenue](images/monthly_revenue.png)

## Data
Public dataset (link), 12 months, cleaned as described below.

## Recommendations
1. Focus retention offers on the two strongest categories
2. Test smaller first-order discounts

## Limitations
One retailer, one year; no cost data.

import pandas as pd
import matplotlib.pyplot as plt
from pathlib import Path

Path("images").mkdir(exist_ok=True)
monthly = pd.DataFrame({
    "month": pd.date_range("2025-11-01", periods=12, freq="MS"),
    "revenue": [42, 38, 35, 37, 40, 39, 41, 43, 45, 52, 61, 48],
})
ax = monthly.plot(x="month", y="revenue", legend=False, marker="o")
ax.set_title("Monthly revenue (thousands)")
ax.set_xlabel("")
plt.tight_layout()
plt.savefig("images/monthly_revenue.png", dpi=150)
print("saved images/monthly_revenue.png")

-- sql/02_repeat_customers.sql
-- Purpose: share of customers who ordered more than once, by first-order month
-- Input:   orders(order_id, customer_id, order_date, amount)
-- Output:  first_month, customers, repeat_customers, repeat_pct

WITH firsts AS (
  SELECT customer_id,
         MIN(order_date) AS first_order,
         COUNT(*)        AS orders
  FROM orders
  GROUP BY customer_id
)
SELECT strftime('%Y-%m', first_order) AS first_month,
       COUNT(*) AS customers,
       SUM(orders > 1) AS repeat_customers,
       ROUND(100.0 * SUM(orders > 1) / COUNT(*), 1) AS repeat_pct
FROM firsts
GROUP BY first_month
ORDER BY first_month;

# Run before pushing: executes the notebook top to bottom
# pip install nbconvert nbformat
import subprocess, sys

for nb in ["notebooks/01_cleaning.ipynb", "notebooks/02_analysis.ipynb"]:
    result = subprocess.run(
        [sys.executable, "-m", "jupyter", "nbconvert", "--to", "notebook",
         "--execute", "--inplace", nb],
        capture_output=True, text=True)
    print(nb, "OK" if result.returncode == 0 else "FAILED")
    if result.returncode != 0:
        print(result.stderr[-500:])

[ ] Clear repo name (retail-sales-analysis, not project1)
[ ] One-line description and topics set
[ ] README: question, findings, chart, data, recommendations, limits
[ ] SQL in numbered, commented files
[ ] Notebooks run top to bottom
[ ] Dashboard screenshots in images/
[ ] Data source and licence linked
[ ] No large or private data committed

# Hi, I am <Name>

Aspiring data analyst working with SQL, Excel, Power BI and Python.

## Projects
- **Retail Sales Analysis**: repeat purchases and seasonality (SQL, Power BI)
- **Appointment No-shows**: which slots are missed most (Python)
- **Delivery KPI Dashboard**: on-time performance (Excel, Power BI)

LinkedIn: <link>

Good commit messages for analytics work

add monthly revenue query and chart
clean city names and document steps in README
add repeat customer analysis by first-order month
add dashboard screenshots
fix date parsing in cleaning notebook

Avoid: 'update', 'final', 'changes', 'asdf'