Human-induced vibrations in slender lightweight footbridges have been extensively investigated in the last 30 years. In existing guidelines, the dynamic response of a footbridge can be assessed considering the pedestrians as external non-interacting loads acting on the structure. For a more realistic load model, individuals defined as single Degree of Freedom (DOF) dynamic systems moving along the structure have been employed by researchers and practitioners. Hence, Human-Structure Interaction (HSI) is considered in the dynamic analysis. Adopting this approach to calculate the vibration levels on a footbridge may lead to a costly computational problem when dealing with a crowd due to the large number of DOFs associated to the pedestrians. Recently, a frequency-domain procedure based on a coupled crowd-structure system has been proposed by the authors to overcome the aforementioned issue. To further validate the proposal, the analytical results should be contrasted with experimental measurements. In this sense, this paper presents the dynamic response of two footbridges under the action of streams of walking pedestrians. A laboratory fibre reinforced polymer structure is firstly analysed, and an in-service cable-stayed steel footbridge is secondly studied. The vibration serviceability of both pedestrian structures is assessed considering 0.20 and 0.50 pedestrians/m2. For each bridge, the experimental results are compared with predictions computed through the application of load models from existing guidelines. In addition, the frequency-domain proposal is employed to calculate the steady-state structural response while accounting for HSI. Since dynamic parameters of the human body must be defined for the second approach, different values available in literature are employed. Results demonstrate the overestimation of the response when provisions stated in current design documents are used. Also, the benefits of the novel procedure to consider Crowd-Structure Interaction on lightweight footbridges are highlighted.