<p>This article presents a methodological framework for enhancing the estimation of the population mean in stratified random sampling by incorporating auxiliary information through a linear cost function. Recognizing the importance of cost efficiency in survey sampling, we propose a generalized class of estimators that effectively utilizes an auxiliary variable to improve precision without compromising resource constraints. To evaluate the statistical properties of the proposed class, we derive the expressions for bias and mean squared error (MSE) up to the first order of approximation under a linear cost constraint. This theoretical development allows for a comprehensive assessment of the estimator’s performance. Furthermore, a detailed comparative study is conducted against several existing related estimators to demonstrate the relative efficiency and practical advantages of our approach. The integration of a cost function into the estimator’s design ensures that the sampling strategy remains not only statistically sound but also economically viable. It aligns the estimation process with the dual objectives of accuracy and budgetary efficiency, ultimately leading to more reliable and actionable results in practical survey applications.</p>