The aim of the study was to investigate the accuracy of Sleep On Cue: a novel iPhone application that uses behavioural responses to auditory stimuli to estimate sleep onset. Twelve young adults underwent polysomnography recording while simultaneously using Sleep On Cue. Participants completed as many sleep-onset trials as possible within a 2-h period following their normal bedtime. On each trial, participants were awoken by the app following behavioural sleep onset. Then, after a short break of wakefulness, commenced the next trial. There was a high degree of correspondence between polysomnography-determined sleep onset and Sleep On Cue behavioural sleep onset, r = 0.79, P < 0.001. On average, Sleep On Cue overestimated sleep-onset latency by 3.17 min (SD = 3.04). When polysomnography sleep onset was defined as the beginning of N2 sleep, the discrepancy was reduced considerably (M = 0.81, SD = 1.96). The discrepancy between polysomnography and Sleep On Cue varied between individuals, which was potentially due to variations in auditory stimulus intensity. Further research is required to determine whether modifications to the stimulus intensity and behavioural response could improve the accuracy of the app. Nonetheless, Sleep On Cue is a viable option for estimating sleep onset and may be used to administer Intensive Sleep Retraining or facilitate power naps in the home environment.