The conversion of abundant small molecules to value-added products serves as an attractive method to store renewable energy in chemical bonds. A family of macrocyclic cobalt aminopyridine complexes was previously reported to reduce CO2 to CO with 98% faradaic efficiency through the formation of hydrogen-bonding networks and with the number of secondary amines affecting catalyst performance. One of these aminopyridine macrocycles, (NH)1(NMe)3-bridged calix[4]pyridine (L5), was modified with a nitrophenyl group to form LNO2 and metalated with a cobalt(II) precursor to generate CoLNO2, which would allow for probing the positioning and steric effects on catalysis. The addition of a nitrophenyl moiety to the ligand backbone results in a drastic shift in selectivity. Large current increases in the presence of added protons and CoLNO2 are observed under both N2 and CO2. The current increases under N2 are ∼30 times larger than the ones under CO2, suggesting a change in the selectivity of CoLNO2 to favor H2 production versus CO2 reduction. H2 is determined to be the dominant reduction product by gas chromatography, reaching faradaic efficiencies up to 76% under N2 with TFE and 71% under CO2 with H2O, in addition to small amounts of formate. X-ray photoelectron spectroscopy (XPS) reveals the presence of a cobalt-containing heterogeneous deposit on the working electrode surface, indicating the addition of the nitrophenyl group reduces the electrochemical stability of the catalyst. These observed catalytic behaviors are demonstrably different relative to the tetra-NH bridged macrocycle, which shows 98% faradaic efficiency for CO2-to-CO conversion with TFE, highlighting the importance of pendant hydrogen bond donors and electrochemically robust functional groups for selective CO2 conversion.