The current study examined the effectiveness of a social network intervention to promote physical activity among adolescents. Social network interventions utilize the effects that peers have on each other to change behavior throughout the social network. These interventions typically identify a small number of individuals within social networks (i.e. influence agents), and train them to promote specific behaviors. The current intervention is characterized by two innovative aspects; The selection and the training of the influence agents. In previous studies, the selected influence agents were those in the network who received the most nominations. For friendship nominations, popular adolescents will receive more nominations from their peers. However, it is argued that popular adolescents often depend on the social norms of the network to remain popular, and therefore may be reluctant to change their behavior. Selecting adolescents who have close ties with all classmates might be more strategic. Second, previous studies used face to face training, which makes the intervention difficult to carry out on a large scale. Therefore, this study contacts and trains de influence agents via smartphones, and keeps contact with the influence agents during the intervention. A total of 190 adolescents (46.32% boys; M age = 12.17, age range 11-14 years) were randomly allocated to either the intervention or control condition. Participants received a research smartphone to measure sociometrics and covariates, and an accelerometer to measure physical activity (steps per day) at baseline and during the intervention one month later. In the intervention condition, the influence agents were trained via the smartphone how they could promote physical activity among their classmates.A multi-level model tested the effectiveness of the intervention, controlling for clustering of the data within participants and days. No intervention effect was observed (b= .11, t(599.46) = 1.10, p =.27).This was the first study to test whether physical activity could be promoted via influential peers by using smartphones. Unfortunately, the intervention could not increase physical activity in adolescents. Next to other discussion points, the consequences of the smartphone based training and the selection criterion for the influence agents are discussed.