It is important that hospital cafeterias deliver products that create customer satisfaction so that financial objectives are met. An exploratory descriptive survey of 12 selected hospital cafeterias used a self-administered questionnaire to determine how satisfied customers were with services provided. It also asked cafeteria managers to give their perceptions of their customers' relative satisfaction/dissatisfaction with the service. Principal components analysis, followed by varimax rotation, identified four underlying constructs of the 15 pre-selected foodservice characteristics used to measure relative satisfaction. A multiple regression model, controlling for country, hospital size and customer demographics, in which the dependent variable was overall rating, found that the independent variables, the underlying rating constructs--food and service--made a much greater impact on overall rating than environment and accessibility. Most cafeteria managers' predictions about their customers' satisfaction were within two standard deviations of their customers' mean scores of satisfaction. While the managers' close association with their service may have accounted for this, it does not necessarily follow that they have the power to implement policy and product improvements.