The electrocatalytic reduction of CO2 (CO2RR) into value-added hydrocarbons is limited due to high limiting potential (UL) and competing hydrogen evolution reaction (HER). To find the best catalyst for CO2 reduction the concept of hydrogen poisoning was not considered in the catalyst screening process. Herein, we present a simple screening method and graphical construction using multiparameter optimization for the design of highly active and selective single-atom catalysts (SAC) using density functional theory calculations. A series of SAC namely, MN4, MBN3 and H@MBN3 (M: metal) are investigated for CO2RR. Our results revealed that MN4 and MBN3 SAC are not favorable for CO2RR due to high UL > −0.85 V and hydrogen poisoning (ΔGH* < 0), respectively. H@MBN3 SAC (stable compounds forming H–B bonds) are identified as efficient catalysts with a low value of UL and significantly hinder the competitive HER. Among these, H@CoBN3 and H@FeBN3 SAC show excellent CO2RR activity with limiting potential −0.30 and −0.44 V respectively for CH4 production and no chance of HER. Scaling relations reveal the importance of *COOH/*CHO binding energy (Eb) as an energy descriptor to evaluate the catalytic performance. This work provides a new theoretical perspective to design a highly selective catalyst for CO2RR. Display omitted •Simple graphical construction is provided to screen the catalyst for selective CO2RR.•The hydrogen poisoning issue is addressed, which blocks the active site.•Hydrogen assisted catalysts (H@MBN3) are designed for highly efficient CO2RR.•The binding energy of *COOH is defined as an energy descriptor for SAC catalyst.