The main challenge of this problem statement is to predicting the next visit of a customer to a store and Product Cannibalization. Product cannibalization refers to sales of one product affects the sales of other product from the same brand. The question of product cannibalization has been a vexing problem since marketing science emerged as a distinct field of study. It deals with the issue of how a new product of a company may draw away sales from its existing products.
Using the raw data, advanced algorithms and unique approaches, models for customer next visit prediction and product cannibalization has been built. The issues for this cannibalization have been properly addressed and help the client to solve the complex and multiple issues. By considering customer demographics and previous transactions, items purchased etc, performing EDA this model has been built. There is a significant improvement with this model.
By analysing data collected from traffic flow analysis, such as customers’ favourite shopping times or days, dwell-time in different store sections and preferred products, AI equips store managers with the insights required to make better, more informed decisions about their inventory, product price and placing merchandise in a store. It helps to manage supply chains, keep stocks to the optimal level and avoiding out-of-stock scenarios.
Our deployed model can process large volumes of data, this optimized model made retail analysis much faster and more effective. It provided store managers with full details of customer behaviour and measures foot traffic in stores, thus helped them make decisions on how to optimize store operations, right stock at right time and improve their customers’ experience.