In this study, we propose a novel stratified random sample compromise allocation for the offered product type estimator. This technique involves selecting a sample that, while maintaining costs to a minimum, maximizes the high level of accuracy of a finite population mean. Choosing the right sample size becomes challenging when different characteristics are observed in each unit and when significant measurement costs between units in a stratum affect the cost function. This study formulates the sample allocation problem in integer nonlinear multivariate stratified random sampling using multi-objective mathematical programming with a proposed cost functions. We developed a procedure for the proposed allocation using the integer programming technique, and we compared the proposed allocation with an already-existing compromise allocation. Theoretical observations verified through numerical examples, proposed compromise allocation is more efficient as compared to other competing allocations. The proposed compromise allocation is used in different areas of applications in real life.