In this paper, we use a laboratory experiment to analyze the effect of social preferences in a coordination game with Pareto-ranked equilibria. Inequality is increased by increasing the coordination payoffs of some subjects while the coordination payoffs of others remain unchanged. Theoretically, in this setting, inequality aversion may lead to a negative relationship between inequality and coordination success, while total payoff motivations lead to a positive relationship. Using a within-subject experimental design, we find that more inequality unambiguously yields a higher level of coordination success. Furthermore, this result holds even for subjects whose payoffs remain unchanged. Our results suggest that total payoff motivations drive the positive relationship between inequality and coordination success found in this experiment. Moreover, our data highlight that the order of treatment matters. Groups facing over time a reduction in inequalities reach the efficient outcome more often, over the entire experiment, compared to groups facing over time an increase in inequalities. This study thus contributes to understanding whether social preferences and variations in inequality affect the outcome of coordination problems.