Every second, large amount of time series data is generated across various organization. Forecasting this data is a critical component of data-driven decision-making, supporting operations in domains such as finance, retail, and supply chain management. It is often difficult for people with no domain knowledge to work with it. The research proposes an automated framework for time series forecasting, designed to assess key data characteristics and select optimal models. This simplifies complex analysis by integrating statistical tests and machine learning techniques, turning it into an accessible abstraction layer. It will then empower user with minimum amount of knowledge to make complex forecasts. The framework then uses Prophet, SARIMA, and ARIMA models in their specific property-tailored application to give more robust predictions with minimal intervention from humans. This approach fills the gap between highly advanced statistical modeling and user-friendly solutions to effectively utilize predictive insight for businesses.