Evaluating the success of a biometric pilot has often been limited to producing a statistical analysis of common metrics, such as enrolment and matching rates or system throughput. In practice, biometric systems are used in ways that neither their manufacturers nor their integrators had envisaged. In addition, there is much evidence to suggest that other, non-quantitative factors should be considered when gathering data and assessing system performance. This article proposes several methods by which evaluators can better assess the overall effectiveness of a biometric pilot.