Managing inventory for a vast number of toy products across different vendors is complex and time-sensitive. Toy N Around was manually forecasting seasonal demand using Excel-based "ladders," which contained product-specific data such as sales history and vendor targets. However, the process was time-consuming and difficult to scale, especially with variables like changing consumer trends, regional seasonality, holidays like Christmas and Easter, and external factors such as viral products or economic shifts.

Eagerworks built a web-based AI platform that automates and enhances the stock prediction process. Users can upload multiple Excel files (ladders) containing product details—such as name, price, and category—and additional files with week-by-week sales targets and relevant calendar events. The platform leverages a machine learning model built with XGBoost, optimized for tabular and numerical data, to generate precise sales forecasts for each product across a semester. Predictions are adjusted based on contextual variables, helping Toy N Around refine import schedules and stay ahead of demand fluctuations.

The platform also features a seamless interface for uploading data, managing forecasts, and downloading complete prediction files for further planning.

The AI-powered platform significantly reduced the time needed to produce accurate sales forecasts, particularly for products with rich historical data. It enabled Toy N Around to partially automate weekly stock updates and improved decision-making around import logistics. The system delivered more reliable projections, reduced human error, and increased efficiency in the prediction workflow. Ultimately, this collaboration helped Target and its partners enhance retail operations in a dynamic and data-rich market like the United States.