Navigating retail inflation in Brazil: A machine learning and web scraping approach to the basic food basket

Abstract

In response to the escalating challenges of global inflation, particularly in developing countries like Brazil, this study combines web scraping and machine learning to analyze inflation dynamics within the retail sector. By systematically real-time pricing and product data from a sponsor company and its four main competitors, we focus on Brazil's most consumed staple foods—beans, rice, sugar, and coffee. Our analysis reveals critical insights into how inflation impacts consumer choices and supply chain operations, highlighting the effectiveness of this approach in providing strategic solutions for managing retail sectors under economic stress. The findings highlight the effectiveness of this approach in providing strategic solutions for managing retail sectors under economic stress. Notably, we observed a 400% increase in sales volume for beans following a 50% price reduction and discovered coffee's price stability as a competitive advantage. Additionally, managerial insights emphasize the importance of diversified sourcing and strategic inventory management to mitigate the adverse effects of inflation.