Policy makers have invested significant resources in financial education to improve financial literacy of the poor, reduce bad financial decision-making, and increase take-up of financial services and products. Yet there is limited evidence on the effectiveness of such interventions, especially in developing countries. This paper provides evidence from a clustered Randomized Controlled Trial (RCT) where a relatively light (a day of training) financial education program was offered to a large sample of women (n=1281) from poor households in non-formal community settings. The educational intervention was a significant departure from the more costly traditional classroom style adult education interventions. It was based on simple ‘Rules of Thumb’ and used a goal-oriented and action-focused approach, targeted at changing behaviors. We find evidence of modest, positive treatment effects for some outcomes including an increase in personal savings, achieved at a relatively low cost of training per participant.