All-perovskite tandem solar cells are attracting considerable interest in photovoltaics research, owing to their potential to surpass the theoretical efficiency limit of single-junction cells, in a cost-effective sustainable manner. Thanks to the bandgap-bowing effect, mixed tin−lead (Sn−Pb) perovskites possess a close to ideal narrow bandgap for constructing tandem cells, matched with wide-bandgap neat lead-based counterparts. The performance of all-perovskite tandems, however, has yet to reach its efficiency potential. One of the main obstacles that need to be overcome is theoftentimeslow quality of the mixed Sn−Pb perovskite films, largely caused by the facile oxidation of Sn(II) to Sn(IV), as well as the difficult-to-control film crystallization dynamics. Additional detrimental imperfections are introduced in the perovskite thin film, particularly at its vulnerable surfaces, including the top and bottom interfaces as well as the grain boundaries. Due to these issues, the resultant device performance is distinctly far lower than their theoretically achievable maximum efficiency. Robust modifications and improvements to the surfaces of mixed Sn−Pb perovskite films are therefore critical for the advancement of the field. This Review describes the origins of imperfections in thin films and covers efforts made so far toward reaching a better understanding of mixed Sn−Pb perovskites, in particular with respect to surface modifications that improved the efficiency and stability of the narrow bandgap solar cells. In addition, we also outline the important issues of integrating the narrow bandgap subcells for achieving reliable and efficient all-perovskite double-and multi-junction tandems. Future work should focus on the characterization and visualization of the specific surface defects, as well as tracking their evolution under different external stimuli, guiding in turn the processing for efficient and stable single-junction and tandem solar cell devices.