All comatose patients after circulatory arrest initially have a severely abnormal disturbed electroencephalogram. The speed of normalisation is a robust contributor to prediction of outcome. Differences between patients with poor and good outcome are largest <24 hours after the arrest. Lasting suppression at ≥12 hours or synchronised patterns with >50% suppression at ≥24 hours are invariably associated with poor outcome. This includes burst suppression with identical bursts and generalised periodic discharges on a suppressed background. Recovery towards continuous patterns within 12 hours is strongly associated with a good outcome. Predictive values are highest at <24 hours despite the use of mild therapeutic hypothermia or sedative medication. Additional value of electroencephalography reactivity for the prediction of poor outcome is negligible. Computer-assisted analysis is equally reliable and may facilitate the use of the electroencephalogram at the bedside on intensive care units. Whether or not treatment of electrographic status epilepticus improves outcome is being studied in the Dutch multicentre randomised TELSTAR trial (NCT02056236). © 2021, Netherlands Society of Intensive Care. All rights reserved.