Aims Early onset neonatal sepsis (EOS) is a rare but significant cause of mortality and morbidity. Babies are treated based on risk factors and clinical indicators, according to National Institute for Health and Care Excellence (NICE) guidelines. An electronic risk calculator has been developed by Kaiser Permanente, providing an estimation of EOS risk for babies ≥34 weeks gestation. Blood cultures are recommended if the risk is ≥1/1000 live births, plus empirical antibiotics if ≥3/1000. The baby can be categorised as Well/Equivocal/Ill and a ‘risk after clinical exam’ score given. We evaluated a large number of EOS cases to determine if it would be safe to introduce the calculator locally. Methods A list of positive blood cultures from babies ≥34 weeks gestation was obtained from two tertiary neonatal units. Study periods were December 2016–November 2017 and July 2016–July 2018 in the first and second trusts respectively. The necessary data for the calculator was obtained from maternal and infant records, and the calculator score recorded. For babies treated based on ‘risk factors’ as per NICE, the score at birth was used. For babies treated according to clinical features, the score after examination was used. The primary outcome was whether the calculator would have recommended empirical treatment in babies who went on to have EOS. Following consultation with the research and development department, there was no requirement for ethical approval. Results There were 21,242 births in the study period and 24 cases of culture-proven EOS. 3 babies were commenced on antibiotics outside of NICE guidance. 11 babies were commenced on antibiotics due to risk factors for EOS; 13 due to clinical indicators. Of the 24 babies with culture-proven EOS, empirical antibiotics were only recommended by the calculator in seven. Of the 11 babies commenced on antibiotics due to risk factors, 10 would have had delayed or missed treatment if the calculator had been used. Conclusion Whilst the calculator has resulted in a substantial reduction in antibiotic use in published studies, we have demonstrated that a large proportion of EOS cases may be missed by the calculator. Currently, the benefit of introducing the calculator does not outweigh the risks.