DS Sales Forecasting

Overview

This project demonstrates how sales data can be transformed, modeled, and used to generate business insights.
It showcases an end-to-end workflow from raw data to forecasting.

Tech Stack

• dbt: data cleaning + modeling
• Python (pandas, Prophet): forecasting
• Matplotlib/Seaborn: visualization

Workflow

1. Raw sales data → staged in dbt
2. Aggregated into daily sales fact table
3. Forecast next 30 days using Prophet
4. Visualize trends + predictions

Next Steps

• Add anomaly detection for unusual sales spikes
• Deploy model in the cloud (Vertex AI / similar)
• Build dashboard in Tableau/Hex