Abstract Background Patients with cervical dystonia (CD) typically require regular injections of botulinum toxin to maintain symptomatic control. We aimed to document long-term patient satisfaction with CD symptom control in a large cohort of patients treated in routine practice. Methods This was a prospective, international, observational study (NCT01753349) following the course of adult CD treated with botulinum neurotoxin type A (BoNT-A) over 3 years. A comprehensive clinical assessment status was performed at each injection visit and subjects reported satisfaction in two ways: satisfaction with symptom control at peak effect and at the end of treatment cycle. Results Subject satisfaction remained relatively stable from the first to the last injection visit. At 3 years, 89.9% of subjects reported satisfaction with symptom control at peak effect and 55.6% reported satisfaction with symptom control at end of treatment cycle. By contrast, objective ratings of CD severity showed an overall reduction over 3 years. Mean ± SD Toronto Western Spasmodic Rating Scale (TWSTRS) Total scores (clinician assessed at end of treatment cycle) decreased from 31.59 ± 13.04 at baseline to 24.49 ± 12.43 at 3 years (mean ± SD reduction from baseline of − 6.97 ± 11.56 points). Tsui scale scores also showed gradual improvement; the percent of subjects with a tremor component score of 4 reduced from 12.4% at baseline to 8.1% at 3 years. Conclusions Despite objective clinical improvements over 3 years, subject satisfaction with symptom control remained relatively constant, indicating that factors other than symptom control also play a role in patient satisfaction.