A great challenge in climate modelling is how to parametrize sub-grid cloud processes, such as autoconversion and accretion in warm rain formation. In this study, we use ground-based observations and retrievals over the Azores to investigate the so-called enhancement factors, <i>E<sub>auto</sub></i> and <i>E<sub>accr</sub></i>, which are often used in climate models to account for the influences of sub-grid variances of cloud and precipitation water on the autoconversion and accretion processes. <i>E<sub>auto</sub></i> and <i>E<sub>accr</sub></i> are computed at a variety of tempo-spatial scales corresponding to different model resolutions. The calculated <i>E<sub>auto</sub></i> increase from 1.79 (0.5-hr/36 km) to 3.15 (3.5-hr/126 km), and the calculated <i>E<sub>accr</sub></i> increases from 1.25 (0.5-hr/36 km) to 1.6 (5-hr/180 km). Comparing the prescribed enhancement factors to the values from observations shows that GCMs are using a much higher <i>E<sub>auto</sub></i> (3.2) and lower <i>E<sub>accr</sub></i> (1.07). This helps to explain why most of the GCMs produce too frequent precipitation events but with too light precipitation intensity. The ratios of rain to cloud liquid water at <i>E<sub>accr</sub></i> = 1.07 and <i>E<sub>accr</sub></i> = 2.0 are 0.048 and 0.119, respectively, further proving that the prescribed value of <i>E<sub>accr</sub></i> = 1.07 used in GCMs is too small to simulate correct precipitation intensity. Both <i>E<sub>auto</sub></i> and <i>E<sub>accr</sub></i> increase when the boundary layer becomes less stable, and the values are larger in precipitating clouds (CLWP > 75 gm<sup>−2</sup>) than those in nonprecipiting clouds (CLWP < 75 gm<sup>−2</sup>). Therefore, the selection of <i>E<sub>auto</sub></i> and <i>E<sub>accr</sub></i> values in GCMs should be regime-dependent.