OBJECTIVE: The aim of this study is to evaluate the impact of the Health-Promoting Communities: Being Active Eating Well (HPC:BAEW, 2007-2010) initiative, which comprised community-based multi-component interventions adapted to community context in five separate communities. The intervention aimed to promote healthy eating, physical activity and stronger, healthier communities. METHODS: A mixed method and multilevel quasi-experimental evaluation of the HPC:BAEW initiative captured process, impact and outcome data. The evaluation involved both cross-sectional (children and adolescents) and longitudinal designs (adults) with data collected pre- and post-intervention in intervention (n=2408 children and adolescents from 18 schools, n=501 adults from 22 workplaces) and comparison groups (n=3163 children and adolescents from 33 schools, n=318 adults from seven workplaces). Anthropometry, obesity-related behavioural and environmental data, information regarding community context and implementation factors were collected. The primary outcomes were differences in anthropometry (weight, waist, body mass index (BMI) and standardised BMI (BMI z-score)) over time compared with comparison communities. Baseline data was collected 2008/2009 and post-intervention collected in 2010 with an average intervention time frame of approximately 12 months. RESULTS: The strategies most commonly implemented were related to social marketing, stakeholder engagement, network and partnership development, community-directed needs assessment and capacity building. Analysis of post-intervention data showed gains in community capacity, but few impacts on environments, policy or individual knowledge, skills, beliefs and perceptions. Relative to the comparison group, one community achieved a lower prevalence of overweight/obesity, lower weight, waist circumference and BMI (P<0.005). One community achieved a higher level of healthy eating policy implementation in schools; two communities achieved improved healthy eating-related behaviours (P<0.03); one community achieved lower sedentary behaviours; and one community achieved higher levels of physical activity in schools (P<0.05). All effect sizes were in the small-to-moderate range. CONCLUSIONS: This was a complex and ambitious initiative, which attempted to expand a previously successful community-based intervention in Victoria into five new contexts and communities. Overall, project success was quite inconsistent, and some significant differences were in the unanticipated direction. However, there are many important learnings that should inform future health-promotion activities. The heterogeneity of outcomes of HPC:BAEW communities reflects the reality of life whereby effectiveness of intervention strategies is dependent on individual and community factors. Future health promotion should consider a systems approach whereby existing systems are modified rather than relying heavily on the addition of new activities, with longer time frames for implementation.