top of page

Revenue Forecasting & Optimization

This project simulates a real-world business scenario where I designed and implemented a complete data analytics pipeline for an e-commerce platform. The goal was to accurately forecast future revenue, analyze the effectiveness of marketing promotions, and simulate pricing strategies that balance profitability with customer demand.

​

I used a combination of statistical modeling (Prophet, ARIMA), advanced visualization, and business-focused simulation to uncover trends, seasonality, and the true impact of discount campaigns on revenue. The final output is a fully interactive Streamlit dashboard that provides actionable insights for business stakeholders. This project showcases my ability to bridge data science and business strategy, delivering not just predictions—but decisions.

  • Python: Core language for data manipulation, modeling, and automation

  • Pandas & NumPy: Data cleaning, transformation, and numerical computation

  • Prophet: Time series forecasting with seasonality and holiday effects

  • ARIMA (pmdarima): Statistical modeling for trend-based revenue prediction

  • Matplotlib, Seaborn, Plotly: Data visualization libraries for trend, seasonality, and forecast analysis

  • Scikit-learn: For regression modeling and simulation of pricing impact

  • Streamlit: Web framework to deploy interactive dashboards with filters, KPI tiles, and simulations

  • Google Colab: Cloud-based development environment for notebook execution

  • Git & GitHub: Version control and open-source project hosting

Tools & Technologies Used

  • Cleaned and engineered features from real-world e-commerce datasets

  • Built time series models and compared MAE, RMSE, MAPE metrics

  • Visualized forecast vs. actual revenue with confidence intervals

  • Simulated revenue impact for different discount strategies

  • Developed a Streamlit dashboard with KPI tiles, forecast visualizations, and pricing controls

What I did

download.png
download (1).png
download (3).png
download (4).png
download (5).png
Screenshot (1098).png

Key Highlights

  • Time series forecasting using Prophet and ARIMA/SARIMA

  • STL decomposition for seasonality and trend analysis

  • Promotion impact modeling using regression analysis

  • Dynamic pricing strategy simulation

  • Streamlit dashboard for interactive business reporting

Links

Github: Revenue Forecasting Optimisation

Live App: revenue-forecasting-optimisation-app

bottom of page