Precise cost estimation is a critical component of construction project management with far-reaching consequences for financial planning, resource allocation, and contract negotiations. Traditional cost estimation methods, which are mostly expert opinion and history-based, have a tendency to ignore real-time market forces, labor cost fluctuations, and material price fluctuations. The present study aims to overcome these shortcomings by employing predictive modeling and machine learning (ML) algorithms to enhance the accuracy of cost estimates. A data-driven model is proposed that employ structured datasets in conjunction with advanced analytical models to predict construction costs from the pre-bid stage to contract award. Trained models such as Linear Regression, Random Forest, and Gradient Boosting are employed to evaluate the predictive precision of these models in predicting contract award costs. Model analysis employs assessment metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R2 Score to evaluate the robustness and consistency of these models. The findings confirm that ML models substantially outperform traditional methods, with Random Forest providing the highest level of accuracy by identifying complex, non-linear relationships between cost determinants. The proposed framework enhances decision-making by reducing cost variances and enhancing financial control. The study contributes to the knowledge base on building cost estimation by highlighting the potential of ML-based predictive analytics in minimizing financial risks and enhancing the precision of project budgets. The findings provide a scalable and flexible model for stakeholders to obtain accurate cost forecasts, reducing budget overruns and enabling the effective allocation of resources across different construction scenarios.