Key points Neurostimulation is a rapidly emerging approach to swallowing rehabilitation, but cerebellar stimulation has not been explored as a treatment. Such proposed therapies for post‐stroke dysphagia have required confirmation of physiological effects and optimisation of parameters in healthy humans prior to translational progression into patient groups. There is strong evidence for a role of the cerebellum in swallowing physiology, but this relationship has been under‐explored. Recently, single pulses of cerebellar magnetic stimulation have been shown to directly evoke responses from pharyngeal musculature and produce short‐term enhancement of cortico‐pharyngeal motor evoked potentials, suggesting the feasibility of a cerebellar approach to neurostimulation in the swallowing system. We therefore examined multiple parameters of repetitive cerebellar magnetic stimulation and have described the optimal settings to provoke longer‐lasting changes in swallowing neurophysiology. Based on evidence from the post‐stroke dysphagia neurostimulation literature, these changes may have a therapeutic potential for swallowing rehabilitation. Abstract Brain neurostimulation has been shown to modulate cortical swallowing neurophysiology in post‐stroke dysphagia with therapeutic effects which are critically dependent on the stimulation parameters. Cerebellar neurostimulation is, however, a novel, unexplored approach to modulation of swallowing pathways as a prelude to therapy for dysphagia. Here, we randomised healthy human subjects (n = 17) to receive one of five cerebellar repetitive TMS (rTMS) interventions (Sham, 1 Hz, 5 Hz, 10 Hz and 20 Hz) on separate visits to our laboratory. Additionally, a subset of subjects randomly received each of three different durations (50, 250, 500 pulses) of optimal frequency versus sham cerebellar rTMS. Prior to interventions subjects underwent MRI‐guided single‐pulse transcranial magnetic stimulation (TMS) to co‐localise pharyngeal and thenar representation in the cortex and cerebellum (midline and hemispheric) before acquisition of baseline motor evoked potential (MEP) recordings from each site as a measure of excitability. Post‐interventional MEPs were recorded for an hour and compared to sham using repeated measures ANOVA. Only 10 Hz cerebellar rTMS increased cortico‐pharyngeal MEP amplitudes (mean bilateral increase 52%, P = 0.007) with effects lasting 30 min post‐intervention with an optimal train length of 250 pulses (P = 0.019). These optimised parameters of cerebellar rTMS can produce sustained increases in corticobulbar excitability and may have clinical translation in future studies of neurogenic dysphagia.