In modal testing, the quality of vibration signals and clarity of peak points to extract the natural frequencies, corresponding mode shapes and damping ratio depend on the number and location of the sensors. High measurement costs are required for structural identification and long-term structural health monitoring in large structures, which require a high number of sensors. Therefore, the minimum number of sensors should be placed at appropriate locations on the system during experimental measurements to ensure both information of sufficient quality and cost reductions. The aim of this study is to perform cost-efficient non-destructive modal tests for a laminated composite and steel cantilever beams using an optimal sensor placement approach. Finite element models of the sample beams are constituted in ANSYS (R) software to determine the initial candidate set of sensor locations. Then, ambient vibration tests are conducted. Based on the experimental modal amplitudes and mode shapes, optimal sensor locations are determined using the effective independence method, and measurements are repeated. The study shows that there is good agreement between the natural frequencies and mode shapes obtained from the initial measurements and those obtained using the limited number of sensors.