Background Recently, patient-specific guides (PSGs) have been introduced, claiming a significant improvement in accuracy and reproducibility of component positioning in TKA. Despite intensive marketing by the manufacturers, this claim has not yet been confirmed in a controlled prospective trial. Questions/purposes We (1) compared three-planar component alignment and overall coronal mechanical alignment between PSG and conventional instrumentation and (2) logged the need for applying changes in the suggested position of the PSG. Methods In this randomized controlled trial, we enrolled 128 patients. In the PSG cohort, surgical navigation was used as an intraoperative control. When the suggested cut deviated more than 3°from target, the use of PSG was abandoned and marked as an outlier. When cranial-caudal position or size was adapted, the PSG was marked as modified. All patients underwent long-leg standing radiography and CT scan. Deviation of more than 3°from the target in any plane was defined as an outlier. Results The PSG and conventional cohorts showed similar numbers of outliers in overall coronal alignment (25% versus 28%; p = 0.69), femoral coronal alignment (7% versus 14%) (p = 0.24), and femoral axial alignment (23% versus 17%; p = 0.50). There were more outliers in tibial coronal (15% versus 3%; p = 0.03) and sagittal 21% versus 3%; p = 0.002) alignment in the PSG group than in the conventional group. PSGs were abandoned in 14 patients (22%) and modified in 18 (28%). Conclusions PSGs do not improve accuracy in TKA and, in our experience, were somewhat impractical in that the procedure needed to be either modified or abandoned with some frequency. Level of Evidence Level I, therapeutic study. See instructions for authors for a complete description of levels of evidence.