Despite the potential of worksite interventions to boost productivity and save insurance costs, they tend to be costly and tested in nonrandomized trials. The aim of the present study was to test the ability of a very brief worksite intervention based on implementation intentions to improve nutrition among health care workers. Seventy-nine health care workers were randomly allocated to a control condition or to form implementation intentions using standard instructions or with a supporting tool. Fruit intake and metacognitive processing (operationalized as awareness of standards, self-monitoring and self-regulatory effort) were measured at baseline and follow-up. Participants who formed implementation intentions ate significantly more fruit and engaged in significantly more metacognitive processing at follow-up than did participants in the control condition (ds > .70). The findings support the efficacy of implementation intentions for increasing fruit intake in health care workers and preliminary support for the utility of a tool to support implementation intention formation.