Dosing of antibiotics in neonates requires finding a delicate balance between maximal efficacy and minimal toxicity. There is a lack of data on efficacy of currently used antibiotics in neonates, and rational dosing therefore needs to be based on gestational- and postnatal-age-dependent pharmacokinetics in combination with surrogate markers. These surrogate markers are: (i) the area-under-the serum concentration time curve to minimum inhibitory concentration ratio (AUC/MIC); (ii) peak concentration to MIC ratio (Cmax/MIC); and (iii) the time the concentration remains above the MIC (T>MIC). Whereas the efficacy of beta-lactam antibiotics (including carbapenems) depends on T>MIC, the efficacy of most other antimicrobials (including aminoglycosides and fluoroquinolones) is related to AUC/MIC and Cmax/MIC. Most modern dosing regimens are adequate when these concentration effect relationships are taken into account. Dosing adjustments in neonates are suggested, based on these relationships. Several antimicrobial combinations for treatment of meningitis and necrotizing enterocolitis exist. Empiric treatment should be based on efficacy, concerns about resistance as well as information from institutional microbiological surveillance.