Uncovering Revenue Trends: Predictive Analytics with Linear Regression and Time Series in Coffee Shop Transactions

This research uses Linear Regression and Time Series Analysis, with the use of Prophet, in analyzing revenue patterns in coffee shop transactions. Using over 149,000 transaction records, simple and multiple regression models were conducted to find out the effect of major predictors like transaction volume, store location, and product category. Prophet helped to conduct accurate 30-day predictions of revenues, drawing attention to seasonal and temporal trends. The results show that adding different features significantly increases the accuracy of regression analyses, while the Prophet model strengthens these findings by making actionable forecasting possible. These observations give support to approaches in inventory optimization, targeted marketing, and operational planning, demonstrating the usefulness of predictive analytics in revenue management over small and large retail businesses.