To determine whether patients with stroke receiving rehabilitation for upper limb deficits using smart technology (video and reminder functions) demonstrate greater adherence to prescribed home exercise programmes and better functional outcomes when compared with traditional paper-based exercise prescription.Randomized controlled trial comparing upper limb home exercise programmes supported by video and automated reminders on smart technology, with standard paper-based home exercise programmes.A community rehabilitation programme within a large metropolitan health service.Patients with stroke with upper limb deficits, referred for outpatient rehabilitation.Participants were randomly assigned to the control (paper-based home exercise programme) or intervention group (home exercise programme filmed on an electronic tablet, with an automated reminder). Both groups completed their prescribed home exercise programme for four weeks.The primary outcome was adherence using a self-reported log book. Secondary outcomes were change in upper limb function and patient satisfaction.A total of 62 participants were allocated to the intervention ( n = 30) and control groups ( n = 32). There were no differences between the groups for measures of adherence (mean difference 2%, 95% CI -12 to 17) or change in the Wolf Motor Function Test log transformed time (mean difference 0.02 seconds, 95% CI -0.1 to 0.1). There were no between-group differences in how participants found instructions ( p = 0.452), whether they remembered to do their exercises ( p = 0.485), or whether they enjoyed doing their exercises ( p = 0.864).The use of smart technology was not superior to standard paper-based home exercise programmes for patients recovering from stroke. This trial design was registered prospectively with the Australian and New Zealand Clinical Trials Register, ID: ACTRN 12613000786796. http://www.anzctr.org.au/trialSearch.aspx.