The value of model‐based translation in drug discovery and development is now effectively being recognized in many disease areas and among various stakeholders. Such quantitative approaches are expected to facilitate the selection on which compound to prioritize for successful development, predict the human efficacious dose based on preclinical data with adequate precision, guide design, and de‐risk later development stages. The importance of time‐dependencies, which are typically species‐dependent due to different turnover rates of biological processes, is, however, often neglected. For bacterial infections, the choice of dosing regimen is typically relying on preclinical pharmacokinetic (PK) and pharmacodynamic (PD) data, because the bacterial load and disease severity, and consequently the PK/PD relationship, cannot be quantified well on clinical data, given the low‐information end points used. It is time to recognize the limitations of using time‐collapsed approaches for translation (i.e., methods where targets are based on summary measures of exposure and response). Models describing the full time‐course captures important quantitative information of drug distribution, bacterial growth, antibiotic killing, and resistance development, and can account for species‐differences in the PK profiles driving the killing. Furthermore, with a model‐based approach for translation, we can take a holistic approach in development of a joint model for in vitro, in vivo, and clinical data, as well as incorporating information on the contribution of the immune system. Such advancements are anticipated to facilitate rational decision making during various stages of drug development and in the optimization of treatment regimens for different groups of patients.