We explore the effects of a randomly assigned conditional cash transfer in Honduras (Bono 10000) on early childhood development. We find significant impacts on cognitive development in children 0-60 months, with an average effect size of 0.13 SD. We show differential impacts by type of transfer: 0-5-year-old children from families receiving the "health" transfer, which targeted families with 0-5-year-old children only, benefited significantly from the program, whereas 0-5 year-olds in families receiving the "education" transfer, which targeted 6-18 year-olds, perceived no benefit. In comparison with other programs, the effect of this impact is sizeable (0.34 SD on average). Although the overall program appears to have slightly changed some behaviors that might affect children (i.e. decreased probability of maternal employment, and increased maternal self-esteem), we did not find heterogenous impacts of the Bono across these variables. Results are explained mainly by differences in conditions: while the "education" component imposed conditions only on children of schooling age, the "health" transfer required regular health checkups of 0-5 year old children. The "health" transfer families were more likely to attend health checkups, which may have induced behavior changes that improved children's health and cognitive development, including purchasing more nutritious food. These results imply that cash without well-targeted conditions attached, might not be as effective for the development of young children.