Background: Diarrhoeal diseases are an important cause of morbidity and mortality in children under five years old in South Asia and Sub-Saharan Africa. Handwashing with soap after contact with faeces is a cost-effective way of preventing diarrhoea. However, handwashing with soap frequencies are low in many settings. Aim: The study aimed to design and evaluate two interventions to increase handwashing with soap after using the toilet, in housing compounds in Koumassi commune, Abidjan, Côte d’Ivoire. Methods: We randomly assigned 75 compounds to one of three arms in a 1:1:1 ratio. One arm received an intervention package comprising disgust-inducing handwashing messages, designed using the Theory of Normative Social Behaviour, and a handwashing station (TNSB); the second arm received only the handwashing station (HWS-only); the third arm served as a control group. The primary outcome was the proportion of occasions when hands were washed with soap after using the toilet, measured at the one-month and five-month post-intervention delivery. Results: One month post intervention, handwashing with soap somewhat increased from a baseline frequency of 3% to 9% in the HWS-only intervention group, and from 7% to 24% in the TNSB-based handwashing intervention group. In the control group, handwashing with soap changed little (from 6% at baseline to 5%). There was strong evidence (P<0.0001) that handwashing with soap frequencies varied between arms: HWS-only intervention versus control, OR=2.00 (95% CI: 1.03- 3.90); TNSB-based intervention versus control OR=7.17 (95% CI: 3.91-13.12). Five months post intervention, the corresponding ORs were 1.01 (95% CI: 0.50-2.04) and 3.11 (95% CI: 1.62-6.00) respectively. Conclusion: A handwashing intervention combining disgust-inducing messages with supply of a handwashing station was effective at increasing handwashing with soap after using the toilet. The provision of a handwashing station-alone had little impact. Future studies should investigate whether the same approach, when delivered via mass media, can have the same effect as the face-to-face delivery used in this study.