The surface radiation budget (SRB) strongly controls the summertime evolution of sea ice and, therefore, plays a key role for the ongoing transformations of the Arctic climate system. Clouds can have a significant impact on the SRB, which is quantified by the cloud radiative effect (CRE). Consequently, continuous, Arctic-wide monitoring of clouds and further factors governing the CRE, including surface and thermodynamic properties, is required. These persistent observations can only be provided by passive remote sensing instruments aboard polar-orbiting satellites. However, cloud detection deficiencies and the lack of accurate surface albedo data over heterogeneous sea ice limit the precision of satellite products and subsequent CRE estimates. Therefore, this study quantifies the accuracy of satellite cloud products, the surface albedo assumed therein, thermodynamic analysis data, and the resulting CRE simulations. To isolate the contributions of individual parameters to the CRE bias, satellite-derived simulation input is consecutively replaced with collocated aircraft observations that were collected over sea ice north of Greenland during the Arctic Radiation–Cloud–Aerosol–Surface Interaction Experiment (ARCSIX) between May and August 2024. It is concluded that clouds warm the surface according to simulations initialized with aircraft measurements, whereas satellite-based CRE estimates suggest a cooling effect. This discrepancy is primarily caused by a negative bias in the assumed surface albedo. Substantial biases are also identified for cloud height and low-level air temperature, but compensating effects and a relatively weak sensitivity of thermal-infrared radiation to these biases mitigate their impacts on the CRE.