INTRODUCTION: The evidence base for the treatment of strabismus (squint) is poor. Our main aim is to improve this evidence base for the treatment of a common type of childhood squint {intermittent exotropia, [X(T)]}. We conducted an external pilot study in order to inform the design and conduct of a future full randomised controlled trial (RCT). METHODS: Children of between 6 months and 16 years with a recent diagnosis of X(T) were eligible for recruitment. Participants were recruited from secondary care at the ophthalmology departments at four UK NHS foundation trusts. Participants were randomised to either active monitoring or surgery. This report describes the findings of the Pilot Rehearsal Trial and Qualitative Study, and assesses the success against the objectives proposed. RECRUITMENT AND RETENTION: The experience gained during the Pilot Rehearsal Trial demonstrates the ability to recruit and retain sites that are willing to randomise children to both trial arms, and for parents to agree to randomisation of their children to such a study. One child declined the group allocation. A total of 231 children were screened (expected 240), of whom 138 (60%) were eligible (expected 228: 95%) and 49 (35% of eligible) children were recruited (expected 144: 63% of eligible). Strategies that improved recruitment over the course of the trial are discussed, together with the reasons why fewer children were eligible for recruitment than initially anticipated. Attrition was low. Outcome data were obtained for 47 of 49 randomised children. TRIAL PROCESSES AND DATA COLLECTION: The Trial Management processes proved effective. There were high levels of completion on all of the data collection forms. However, the feedback from the treatment orthoptists revealed that some modifications should be made to the length and frequency of the health service assessment and travel assessment questionnaires, thus reducing the burden on participants in the main trial. Modifications to the wording of the questions also need to be made. MONITORING OF BIAS: Children who recruited to the trial were older and had more severe strabismus than those children eligible but declining participation. Strategies to account for this in a full trial are proposed. REASONS FOR PARTICIPATION OR DECLINING STUDY: These were identified using qualitative interviews. The principal reasons for declining entry into the study were strong preferences for and against surgical treatment. HARMS: There were no serious unexpected adverse events. Two children had overcorrection of their X(T) with reduction in binocular vision following surgery, which is in line with previous studies. No children in the active monitoring arm developed a constant strabismus although two showed some reduction in control. CONCLUSIONS: The SamExo study has demonstrated that it is possible to recruit and retain participants to a randomised trial of surgery compared with active monitoring for X(T). For longer-term full RCTs, in order to maximise the generalisability of future studies, consideration needs to be given to planning more time and clinic appointments to assess eligibility and to allow consideration of participation; the greater use of research nurses for recruitment; and accommodating the strong preferences of some parents both for and against surgical intervention. TRIAL REGISTRATION: Current Controlled Trials ISRCTN44114892. FUNDING: This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 19, No. 39. See the NIHR Journals Library website for further project information.