View complementary dashboard (Looker Studio)
Read full article of this project
This data analytics project analyzes more than 246,000 chip purchase transactions across more than 260 stores for a supermarket chain to support category strategy planning and store layout trial evaluation. Customer segments were profiled by lifestage and spending behavior, identifying Mainstream Young Singles and Couples, Mainstream Retirees, and Budget Older Families as the highest-value segments in terms of sales volume and average sale amount per transaction. A control store matching algorithm was created based on correlation coefficient and magnitude distance scoring to identify look-alike benchmarks for three trial store. Using a 90% confidence interval, two of three trial stores were confirmed to have sales and customer uplift during the trial period, supporting a recommendation for a rollout in all stores.