Customer Shopping Behavior with Apriori Analysis

Shopping malls often run promotional campaigns during special holidays, offering customers the chance to win prizes or enter raffles in exchange for submitting their purchase invoices. These campaigns provide a valuable source of transactional data, as this information is typically only accessible by the stores. By collecting these invoices, we can generate unique insights about customer spending patterns and movement across multiple retailers.
Using Python and the mlxtend library, I applied Market Basket Analysis techniques with the Apriori algorithm to identify frequent store pairs that customers visited during the same trip. Metrics like support and confidence highlighted the most common and impactful store combinations.
To present the findings, I created a Chord diagram with Python that visualized store relationships and their influence on total visits and upselling opportunities. These insights helped the operations and marketing teams optimize the store mix and design strategies to increase spending in future campaigns.
You can read more about this analysis in the Investor Relations report this project was showcased in the Earnings Release 4Q20 (p. 14) [direct link to the PDF].