Background: A total of 68% of pre-school children with cleft palate have speech problems requiring speech therapy. There is a lack of access to regular targeted therapy. Parent training leads to positive outcomes in early communication skills in cleft palate and non-cleft speech disorders. Connected health has been used to address inadequate access to therapy, providing intervention to those who would not otherwise receive therapy. Aims: To evaluate the speech, activity and participation outcomes of Parent Led, Therapist Supervised, Articulation Therapy (PLAT) compared with routine speech therapy intervention in parent-child dyads. Methods & Procedures: A total of 44 children, aged 2.9-7.5 years, were included in a two-centre, two-phase randomized controlled trial. Informed consent and assent were obtained. Participants and speech and language therapists (SLTs) were unblinded to the groups. Parents, in the parent-trained group (n = 23), attended 2 days' training, received a detailed speech therapy programme, and undertook intervention over 12 weeks supported by the cleft specialist SLT using FaceTime and one face-to-face session. In the control arm (n = 21), parent-child dyads received six therapy sessions over 12 weeks with a research SLT, comparable with usual care. Speech recordings were undertaken pre- and post-intervention. Percent consonant correct (PCC) was analysed by external SLTs blinded to the time and group. Activity and participation were measured using the Intelligibility in Context Scale (ICS) and Focus on Outcomes for Children Under Six (FOCUS) questionnaire. Outcomes & Results: There was no evidence of an interaction between Time and Group or an overall statistical difference between groups for PCC scores. There was a statistically significant difference over time for both groups (words: p < 0.002; confidence interval (CI) = 9.38-16.27; d = 0.57; sentences: p < 0.002; CI = 16.04-25.97; d = 0.23). Effect sizes were medium for words and small for sentences. For intelligibility and participation, there was no evidence of an interaction between Time and Group or an overall statistical difference between groups. A statistically significant difference over time was found for intelligibility (F = 29.97, d.f. = 1, 42, p < 0.001, 95 % CI = 1.45--3.15 d = 0.46) and for participation (F = 14.19, d.f. = 1, 41, p < 0.001 95% CI = 7.63--25.03; d = 0.36) with FOCUS results indicating clinically meaningful (parent-led group) and significant (control group) change in participation. Conclusions & Implications: PLAT can be as effective as routine care in changing speech, activity and participation outcomes for children with cleft palate, when supported by a specialist cleft SLT using connected health.