Retail Analytics Project

I have a time-sensitive retail analytics assignment that has to be wrapped up within 3 to 4 days. Once we start, I’ll share the raw transactional files and a short requirements brief. You’ll first wrangle the data in SQL , then move into Python for any advanced calculations or feature engineering, and finally present the insights in an interactive Power BI dashboard.

Core focus
• Clean and model the data so that it is query-ready
• Produce a concise Python notebook that documents every transformation
• Build a visually polished Power BI report with slicers, drill-downs, and KPI cards

Acceptance criteria
1. SQL scripts execute without errors on my side.
2. Python notebook runs end-to-end via a single “Run All” and outputs the final dataset.
- Engineer features such as Recency, Frequency, and Monetary value.
- Train baseline and advanced models (Logistic Regression, Random Forest, Gradient Boosting).
- Perform hyperparameter tuning and evaluate using Accuracy, Precision, Recall, F1, and ROC-AUC.
- Interpret model outputs and identify top churn predictors.
3. Power BI file shows no broken visuals and refreshes from the SQL views.
4. A brief Markdown hand-off note summarises assumptions, column definitions, and refresh steps.

If everything clicks, I have a queue of similar analytics mini-projects lined up, so this could be the start of an ongoing collaboration.

1. Jupyter Notebook (Single Or Phase-wise)
File(s):
• Phase1.ipynb, Phase2.ipynb, Phase3.ipynb, Phase4.ipynb
Must include:
• Clean, commented code
• Plots + metrics
• Short insights written in Markdown

2. Power BI Dashboard
File: RetailSmart_Dashboard.pbix
Should show:
• KPIs
• Customer Segments
• Churn insights
• Forecast trends
• Any other interesting insights you want to highlight

3. Exported CSV Outputs (for Power BI)
Folder: data_visualization/Includes (only the essentials):
• model_input.csv
• cluster_summary.csv
• forecast_results.csv
• association_rules.csv

4. Final Report (Short Business Summary)