Alcohol problems are not just an individual concern; they have important negative impacts on romantic relationships. Perceiving one's romantic partner to have an alcohol problem is associated with lower relationship satisfaction and commitment. However, the utility of informant-reports of a partner's alcohol problems in predicting future dyadic conflict remains unknown. Our objective was to test the incremental validity of informant-reports of a partner's alcohol problems in predicting escalations in dyadic conflict over a one-month period beyond the partners' self-reported alcohol problems. One-hundred-eighty-seven opposite-sex couples participated in a one-month longitudinal study involving self- and informant-reports of alcohol problems at baseline and dyadic conflict measures at baseline and one-month follow-up. We hypothesized that, in both sexes, informant-reports of a partner's alcohol problems would predict escalations in dyadic conflict above and beyond the variance explained by self-reports of alcohol problems. This hypothesis was partially supported. Informant-reports of a partners' alcohol problems incrementally predicted future dyadic conflict, but only when women were informants. Women's self-reports of their own alcohol problems also predicted escalations in dyadic conflict whereas men's self-reports did not. Findings suggest that having women report on both their own and their male partners' levels of alcohol-related problems may help identify at-risk couples for early intervention to prevent escalating dyadic conflict. Though self-reported alcohol problems can be accurate, a person's position as expert on his or her alcohol problems can be compromised by biases (e.g., self-deception). Informant-reports may provide a more complete picture of the problem drinker in the romantic relationship context.