In my thesis, I use state-of-the-art simulations to model and quantitatively resolve several of the important systematics for cluster cosmology. First, I quantify the impact of triaxiality, that is, the ellipsoidal shapes and orientations of clusters, on cluster observables and explore the relation between triaxiality and other major systematics such as miscentering and projection effects. Second, I quantify and develop a model for the previously unexplored covariance between cluster observables. The covariance study divides into two parts, one focused on the intrinsic covariance of cluster signals due to the underlying formation physics, and the second that includes a realistic cluster finder to quantify the impact of extrinsic terms due to observational effects on the covariance. These results have been included in the analysis of current optical surveys as DES and will be useful for stage IV cluster surveys as LSST and DESI.