This project builds a drug store sales forecasting model to predict future pharmaceutical revenue trends using 3 years of historical sales data across 1,000+ stores. A linear regression model was trained on monthly aggregated sales to forecast the next 6 months of revenue.
Pharmaceutical companies and retail drug store networks need accurate sales forecasts to optimize inventory, plan staffing, and allocate marketing budgets. This project simulates the kind of commercial forecasting analysis consultants at firms like ZS Associates perform for pharma clients.
- Clear seasonal sales patterns visible across all years — peaks in mid-year (July/August) and year-end
- Sales show a gradual upward trend from 2013 to 2015
- Forecast predicts stable revenue of ~193M per month for the next 6 months based on historical trend
- Source: Rossmann Drug Store Sales (Kaggle)
- 1M+ rows of daily sales data across 1,000+ stores
- Aggregated to monthly level for trend analysis
- Loaded and cleaned raw sales data using Python/pandas
- Removed closed store days (Sales = 0)
- Aggregated daily sales to monthly totals
- Engineered time index feature for modeling
- Trained Linear Regression model on 31 months of data
- Forecasted next 6 months of sales
- Visualized actual vs forecasted sales in Tableau Public
Built in Tableau Public. Includes:
- Line chart: actual vs forecasted sales trend
- Bar chart: monthly sales breakdown by year
Click the link to view dashboard: https://public.tableau.com/views/Pharmasalesforecastingdashboard/Dashboard1?:language=en-US&:sid=&:redirect=auth&:display_count=n&:origin=viz_share_link
- Python (pandas, matplotlib, scikit-learn)
- Jupyter Notebook (VS Code)
- Tableau Public
- GitHub