Purpose: Session rating of perceived exertion (sRPE) is an often-used measure to assess athletes’ training load (TL). However, little is known about which factors could optimize the quality of data collection thereof. The aim of this study was to investigate the effects of the survey methods and the time points when sRPE was assessed on the correlation between subjective (sRPE) and objective (heart-rate training impulse; TRIMP) assessment of TL. Methods: In the first part of the study, 45 well-trained subjects (30 men and 15 women) performed 20 running sessions with a heart-rate monitor and reported sRPE 30 min after training cessation. For the reporting, the subjects were grouped into 3 groups (paper–pencil, online questionnaire, and mobile device). In the second part of the study, another 40 athletes (28 men and 12 women) performed 4 × 5 running sessions with the 4 time points to report the sRPE randomly assigned (directly after training cessation, 30 min postexercise, in the evening of the same day, and the next morning directly after waking up). Results: The assessment of sRPE is influenced by time point, survey method, TRIMP, sex, and training type. It is recommended to assess sRPE values via a mobile device or online tool, as the paper survey method displayed lower correlations between sRPE and TRIMP. Conclusions: Subjective TL measures are highly individual. When compared with the same relative intensity, lower sRPE values were reported by women for the training types representing slow runs and for time points with greater duration between training cessation and sRPE assessment. The assessment method for sRPE should be kept constant for each athlete, and comparisons between athletes or sexes are not recommended.