BUSINESS NEED

A small high-end supermarket chain sought to gain a deeper understanding of customer purchasing behaviors to optimize product placement, enhance cross-selling opportunities, and improve promotional strategies. The primary objective was to analyze frequently bought-together products, enabling the supermarket to redesign in-store layouts and create targeted promotions. Additionally, the supermarket aimed to forecast potential sales increases from these strategic layout changes and promotional activities, supporting its broader goal of adopting a data-driven approach to decision-making.

METHODOLOGY

The Analytics Arts team implemented a market basket analysis using transaction data from the supermarket’s point-of-sale system. After cleaning the data, we conducted exploratory analyses to identify frequent item sets and product associations. Association rule mining was applied to uncover actionable insights, using metrics like support, confidence, and lift to validate findings, with feedback from store managers ensuring practicality. The analysis also considered factors like day of the week and holidays to enhance relevance.

OUTPUT

The analysis revealed actionable insights, such as a strong association between pasta, pasta sauce, and cheese purchases, prompting strategic product placement to boost cross-sales. Seasonal patterns, like snack and beverage combinations during sports events and higher product sales on specific days, led to targeted promotions. A simulation model predicted the sales uplift from optimized layouts and bundles.As a result, the supermarket improved product placements, enhanced customer satisfaction, and increased sales, proving the effectiveness of market basket analysis in driving data-driven business decisions.

Recent Case Studies