The Youden square design is widely used in experimental research to control two sources of variability, but missing data can compromise the results. Addressing missing data is critical to maintaining the integrity and reliability of such experiments. This paper proposes to adapt the Yates method to handle missing data specifically in Youden square designs. We begin by outlining the structure of the Youden square design and the challenges posed by missing data. The Yates method, known for its robustness in estimating missing data, is adapted to fit this design. We demonstrate its effectiveness through simulations and real-world case studies. The simulation involved generating experimental data with one missing value, and the case study analyzed chemical process research with critical missing data points. The results show that the Yates method maintains statistical validity and improves data completeness compared to traditional methods. Its advantage lies in utilizing Youden's quadratic structure for more accurate estimation. This study highlights the Yates method as a solution to handle missing data, improving the quality and reliability of experimental research.