Background Many drugs are used off-label in term and preterm neonates, and dosing recommendations for many drugs are lacking in the Dutch Paediatric Formulary (DPF). This results in widely varying dosing regimens used across neonatal intensive care units (NICUs) in the Netherlands. The Neodose pilot project aimed to develop best-evidence national dosing recommendations for (pre)term neonates. Because scientific evidence is scarce, a consensus-based approach was used. Methods A priority drug list, containing the most frequently used drugs for neonates, was drafted. From this list 22 drugs were selected for further research within the Neodose pilot project. The pilot utilized a two-step approach: First, consensus was established with all Dutch NICUs for neonatal dosing recommendations. Local treatment protocols were retrieved, compared and discussed, leading to consensus-based dosing recommendations. Secondly, we aimed to develop best-evidence dosing recommendations for the following five drugs: acyclovir, ganciclovir, ibuprofen, hydrocortisone and dexamethasone. Results For 21 of 22 drugs, local dosing guidelines differed significantly. Mostly concerning total daily dose, dosing frequency and route of administration. Little or no distinction is made between treatment of preterm and term neonates. Approximately half of the consensus-based dosing recommendations (45%) differ in some degree from all local protocols. Comparing the consensus-based dosing recommendations with the available evidence, almost half of the consensus doses were adjusted. The grounds on which dosing recommendations were adjusted differed. Acyclovir-dosing adjustment was based on pharmacokinetics. Hydrocortisone-dosing was adjusted due to new insights after the evidence has been put together. For dexamethasone-dosing, the consensus dose was eventually chosen, because every available trial used a different dosing regimen. Conclusion This pilot showed that, when evidence is inconclusive, consensus on dosing regimens in neonates can be obtained ...