In this tutorial, we demonstrate how to use the Synthetic Data SDK in AWS SageMaker to perform conditional generation of partially synthetic datasets. By fixing certain attributes and adjusting others, we can conduct what-if analysis to understand how changes in one variable impact others. Using a gender-balanced dataset with randomized income levels, we explore how removing income gaps affects age distribution and other factors. We also show how to create and share synthetic data securely within a SageMaker environment for privacy-preserving insights. Watch to learn how this technique can be applied in real-world scenarios.