--- - |2+ Background The therapeutic alliance is an important factor in psychotherapy, affecting both therapy processes and outcome. Therapy transfers may impair the quality of the therapeutic alliance and increase symptom severity. The aim of this study is to investigate the impact of patient transfers in cognitive behavioural therapy on alliance and symptoms in the sessions after the transfer. Method Patient‐ and therapist‐rated therapeutic alliance and patient‐reported symptom severity were measured session‐to‐session. Differences in the levels of alliance and symptom severity before (i.e., with the original therapist) and after (i.e., with the new therapist) the transfer session were analysed. The development of alliance and symptom severity was explored using multilevel growth models. Results A significant drop in the alliance was found after the transfer, whereas no differences were found with regard to symptom severity. After an average of 2.93 sessions, the therapeutic alliance as rated by patients reached pretransfer levels, whereas it took an average of 5.05 sessions for therapist‐rated alliance levels to be at a similar level as before the transfer. Inter‐individual differences were found with regard to the development of the therapeutic alliance over time. Conclusions Therapy transfers have no long lasting negative effects on either symptom impairment or the therapeutic alliance. - Clinical Psychology & Psychotherapy, EarlyView.