This paper presents an alternative approach to estimating software project effort using analogy, which is compared with traditional algorithmic methods such as stepwise regression. The analogy-based method, known as ANGEL, uses a case-based reasoning approach to find similar completed projects and uses their known effort values to predict the effort for a new project. The method is validated on nine industrial datasets, totaling 275 projects, and is shown to outperform algorithmic models in all cases. The key idea is to characterize projects using features such as the number of interfaces, development method, and size of the functional requirements document. These features are standardized and used to calculate similarity in n-dimensional space. The known effort values of the nearest neighbors are then used to predict the effort for the new project. The method is automated and has been shown to be effective even in cases where traditional algorithmic models fail, such as when the dataset contains only categorical data or when no statistically significant relationships exist between the independent variables and effort. The paper also discusses the sensitivity of the analogy method to the size and homogeneity of the dataset, and concludes that estimation by analogy is a viable technique that can complement traditional estimation methods. The paper highlights the advantages of analogy-based estimation, including its intuitive nature and ability to handle poorly understood domains. It also notes that while analogy-based estimation can be effective, it is not without limitations, and that multiple methods should be used to improve prediction accuracy. The paper concludes that analogy-based estimation is a promising technique for predicting software project effort, particularly in situations where traditional algorithmic models are not applicable.This paper presents an alternative approach to estimating software project effort using analogy, which is compared with traditional algorithmic methods such as stepwise regression. The analogy-based method, known as ANGEL, uses a case-based reasoning approach to find similar completed projects and uses their known effort values to predict the effort for a new project. The method is validated on nine industrial datasets, totaling 275 projects, and is shown to outperform algorithmic models in all cases. The key idea is to characterize projects using features such as the number of interfaces, development method, and size of the functional requirements document. These features are standardized and used to calculate similarity in n-dimensional space. The known effort values of the nearest neighbors are then used to predict the effort for the new project. The method is automated and has been shown to be effective even in cases where traditional algorithmic models fail, such as when the dataset contains only categorical data or when no statistically significant relationships exist between the independent variables and effort. The paper also discusses the sensitivity of the analogy method to the size and homogeneity of the dataset, and concludes that estimation by analogy is a viable technique that can complement traditional estimation methods. The paper highlights the advantages of analogy-based estimation, including its intuitive nature and ability to handle poorly understood domains. It also notes that while analogy-based estimation can be effective, it is not without limitations, and that multiple methods should be used to improve prediction accuracy. The paper concludes that analogy-based estimation is a promising technique for predicting software project effort, particularly in situations where traditional algorithmic models are not applicable.