Background Therapeutic hypothermia is known to improve outcomes in neonates with hypoxic ischemic encephalopathy (HIE). There are no studies that have compared servo controlled cooling (SCC) versus manually controlled cooling (MCC) methods in HIE. Aim To compare the outcomes of SCC versus MCC in neonates with HIE. Methods and outcome measures Between Jan 2008 and May 2011, MCC with cool-gel packs was used to achieve rectal temperatures of 33.5 to 34.5 °C in our units. Subsequently, we changed to SCC to achieve rectal temperatures of 33 to 34 °C. 105 neonates received SCC whereas 95 received MCC. Retrospective study with multivariate analysis was conducted comparing thermoregulation (primary outcome) and neurodevelopmental outcomes of SCC versus MCC. Results In the SCC group, 72.3% had stage 2 or 3 HIE versus 77.9% in the MCC. The remaining had stage 1, but attending neonatologists decided to provide cooling. Prescribed rectal temperatures were achieved in 80.5% (5768/7168) instances in SCC versus 72.9% (4449/6108) in MCC (p < 0.0001). There were no significant differences in the incidence of ‘death or moderate-severe disability’ [SCC 26/85 vs MCC 26/87, adjusted odds ratio: 1.29, 95% CI: 0.48, 3.42; p = 0.614]. The results were similar after excluding stage 1 HIE [SCC 25/66 vs MCC 26/69, adjusted odds ratio: 1.01, 95% CI: 0.50, 2.02; p = 0.981]. Conclusions SCC resulted in better thermoregulation in neonates undergoing therapeutic hypothermia. There were no significant differences in neurodevelopmental outcomes, but the study was underpowered to answer this. Randomized trials are needed to fine-tune various aspects of TH in neonates with HIE.