Industry: Retail Service

Service: Data Engineering, Machine Learning

Technology: Amazon Web Services

Client Overview

US-based Furniture retail stores that sell furniture and cooperate with multiple brands, providing a platform to sell products in multiple categories. Seeking a third-party company that could build – future sales forecasting system to forecast which products are to be available for purchase at the right store at the right time.

Business Need

  • Leverage AWS services to build a fast and effective ML model for forecasting future sales to enhance the existing furniture business with efficient inventory planning and to avoid products being out of stock
  • The business goal was to use the Amazon Sagemaker service to build an ML model to forecast future sales

How Did We Help

  • CloudJournee’s professionally certified Data Engineers conducted continuous discussions with client staff and Data Analysts to understand and analyze the current data and its features
  • CloudJournee fully understood the existing complexities of the client’s project at hand and quickly advised on the Time series forecasting system
  • Analyzing and cleaning the unstructured data was done through the EDA process using Amazon Sagemaker Jupyter notebook and the preprocessed data was then used to build, train, test, and deploy the ML model using the ML algorithm that would forecast the future sales
  • As a result, future sales were predicted for relevant categories of products, which helped in planning the inventory and making sure of product availability


  • Future forecasting helped the client to plan inventory that Increased sales by 12%
  • Reduction in Out of stock by 10 %
  • Identify Seasonal Patterns (Egg -Sales pattern of month & year) and plan better product availability
  • Future sales forecasting helped the client in creating opportunities to improve business