This study used an observational design to examine the kinematics of spinal manipulative therapy (SMT) by determining the acceleration characteristics of the manipulative input at the cervical, thoracic, and lumbar spinal regions. Studies of SMT have been restricted to measuring the forces that result from the manipulative input. Several studies have indicated the rate of force development is a key parameter of clinically delivered SMT. Despite this, the movement strategies employed during SMT, including acceleration, have not been directly measured. Participants (n = 29) were recruited from a private practice chiropractic clinic. A wireless accelerometer attached to the clinician's hand was used to characterize the thrust phase of the SMT treatments. Significant differences were found across each spinal region for acceleration amplitude parameters (p < 0.0001). Post-hoc analysis indicated that amplitudes significantly increased in order from thoracic to cervical to lumbar regions (p < 0.0001). Spinal level was also a significant factor in determining the temporal parameters of hand acceleration during SMT (p < 0.0005). This study provides a description of the acceleration properties of clinically delivered SMT. Consistent with that reported for SMT forces, acceleration amplitudes varied significantly across spinal regions with relatively little differences in acceleration latencies. Notably, acceleration amplitudes and latencies were not associated with each other within spinal regions. These findings indicate that changes in acceleration amplitude, rather than latency, are used to tailor SMT to individuals.