Density functional calculations were preformed to investigate whether adding Ni into a Cu surface (denoted as Cu/Ni) or adding Cu into a Ni surface (Ni/Cu) is more efficient for catalyzing the water-gas shift (WGS)? The reactions of water dissociation and monoxide dissociation were selected to assess the activity and selectivity towards WGS, respectively. Our results show that Ni-atom modification of surfaces is thermodynamically favorable for both reactions. Kinetically, compared with pure Cu, water dissociation is greatly facilitated on Ni-modified surfaces, and the activity is insensitive to the Ni concentration; however, monoxide dissociation is not well-promoted on one Ni-atom-modified surfaces, but two Ni-atom modification can notably decrease the dissociation barriers. Overall, on the basis of these results, we conclude that 1) the catalytic performance of bimetallic metals is superior to monometallic ones; 2) at the same Ni concentration on the surface, Cu/Ni and Ni/Cu alloys have almost the same performance towards WGS; and 3) to acquire high WGS performance, the surface Ni atoms should either be low in concentration or highly dispersed.