This paper tests how migrants’ willingness to remit changes when given the ability to direct remittances to educational purposes using different forms of commitment. Variants of a dictator game in a lab-in-the-field experiment with Filipino migrants in Rome are used to examine remitting behaviour under varying degrees of commitment. These range from the soft commitment of simply labelling remittances as being for education, to the hard commitment of having funds directly paid to a school and the student’s educational performance monitored. We find that the introduction of simple labelling for education raises remittances by more than 15%. Adding the ability to directly send this funding to the school adds only a further 2.2%. We randomly vary the information asymmetry between migrants and their most closely connected household, but find no significant change in the remittance response to these forms of commitment as information varies. Behaviour in these games is then shown to be predictive of take-up of a new financial product called EduPay, designed to allow migrants to directly pay remittances to schools in the Philippines. We find this take-up is largely driven by a response to the ability to label remittances for education, rather than to the hard commitment feature of directly paying schools.