In this talk we analyse the claim that supersymmetry (SUSY) naturally accounts for the observed dark matter density. In many cases, it is necessary to tune the parameters of a SUSY model to fit the WMAP data. We provide a quantitative analysis of the degree of tuning required for different annihilation channels. Some regions are natural, requiring no tuning at all, whereas others require tuning at the 0.1\% level.

PACS numbers: 11.30.Pb