Grocery Sales Analysis
About Project
This project presents an end-to-end data analytics solution, built to analyze and visualize historical grocery sales data from Favorita stores in Ecuador. Using Python for data preprocessing and Power BI for interactive dashboards, the analysis uncovers sales patterns, highlights the impact of promotions and holidays, and forecasts future trends.
With over 125 million transaction records, the dashboard delivers insights into unit sales, product demand, and store performance. It also demonstrates how external factors—such as oil prices, transferred holidays, and national events like the 2016 Ecuador earthquake—influence consumer behavior.
The goal of this project is to simplify complex retail data into actionable intelligence, empowering businesses to make smarter decisions on inventory management, pricing strategies, and marketing campaigns.
Recommendations
1. Optimize Inventory Planning
Leverage historical sales trends, seasonality, and regional consumption patterns to forecast demand more precisely. Align inventory levels with predicted peaks (e.g., holiday seasons) and dips to prevent both overstocking and stockouts.
Expected Outcome:
10–15% reduction in excess inventory costs.
Improved shelf availability leading to 5–8% higher sales during peak periods.
Reduced product wastage and holding costs.
2. Leverage Promotions Strategically
Analyze past promotional performance to identify high-impact periods and product categories. Schedule promotions around holidays, paydays, and high-traffic weekends to drive incremental revenue.
Expected Outcome:
12–20% sales uplift during promotional cycles.
Higher customer engagement and repeat purchase rates.
Optimized marketing spend with measurable ROI improvements.
3. Enhance Regional Strategies
Segment performance by region and store type (urban vs. suburban, large vs. express stores). Customize pricing, product mix, and marketing efforts to match local preferences and buying power.
Expected Outcome:
8–10% improvement in regional revenue contribution.
Increased customer satisfaction and loyalty due to localized assortments.
Stronger regional market share.
4. Monitor External Factors
Integrate external datasets — such as inflation rates, fuel prices, holidays, and national events — into predictive models to capture real-world influences on consumer behavior.
Expected Outcome:
More accurate demand forecasts (up to 90–95% accuracy).
Enhanced responsiveness to market shifts and economic conditions.
Improved strategic planning and budgeting accuracy.
5. Expand Dashboard Usage
Roll out interactive Power BI dashboards to regional managers and category heads for real-time performance monitoring. Enable drill-through analytics for SKU, store, and promotion-level insights.
Expected Outcome:
30–40% faster decision-making cycle.
Improved collaboration between operations, sales, and marketing teams.
Data-driven culture fostering continuous improvement and agility.
Summary
The Grocery Sales Analysis project successfully transformed raw transactional data from Favorita stores into actionable business insights. By leveraging Python for data preparation and Power BI for interactive dashboards, the project uncovered trends in unit sales, product demand, store performance, and the influence of promotions, holidays, and external events. The analysis of over 125 million records not only highlighted key sales drivers but also provided a foundation for accurate forecasting and strategic decision-making. The dashboard is available upon request.