Nowadays Retailers are into a digital transformation inside the physical store. They are looking for new processes in order to implement a Technology Revolution at the Instore experience level. Retail Reply, together with Amazon Web Services, is designing an end-to-end journey with several use case, from the customer engagement outside the store to the payment processing.
The aim is dual: on one side to evolve the actual customer experience into a digital human experience, on the other side to provide more specific data (customer behavior, heat map, etc.) in order to improve the customer satisfaction.
Retail Reply solution is based on standard AWS services such as Panorama, Sagemaker and Rekognition. Starting from them, we are developing algorithms on Machine Learning, image recognition and chat-bot to realize innovative use-cases.
Set outdoor, it can monitor the behavior of passersby comparing to the ones entering the store with the others standing in front of store window.
Benefits: provide data analysis to improve the store window to bring more customers to enter.
• Number of people transited outside
• Number of people stopped in front of the windows • Number of people entered • Average time before to enter • Average time before to go away
• Rekognition people pathing
Set indoor, combined with visual recognition can record employees time of arrival at work and of the end of work.
Benefits: check automation and immediate data analysis.
Expected Outcome: • Identification of employee via uniform, badge or image
Inside the dressing room it will be possible to mirror yourself in an Interactive and Smart Mirror.
This mirror is able to recognize the garment and to propose the following: • Information (material, price) • Size availability • Outfit Visual AI algorithm identifies items within any image, breaks them down to precise visual attributes, and suggests products from your inventory to complete the look.
Benefits: cross and up selling, customer experience improvement.
Expected Outcome: • Item recommended linked with the customer profile and appearance • Possibility to create a “palinsesto” for the waiting time
• Rekognition • Sagemaker
Cameras spread inside the store to monitor any movements of customers and employees. In particular, to monitor the customers’ preferred shelves and those one less frequented; to monitor any fraudulent behaviour by customers or people from the staff.
Benefits: analyze strategic points of the store, store layout improvement and goods restocking forecast + reduce fraudulent behaviors both from customers and employees’ side.
Expected Outcome: • Heatmap in order to understand the better item positioning and the best store layout
• Possibility to verify the inside layout if it matches or not with the standard template
• Rekognition • Sagemaker