BackgroundPerforming electrocardiography is common in general practice, but the quality of indication setting and diagnostic accuracy have been disputed.ObjectivesTo assess the competence of general practitioners (GPs) in their decision-making process with regard to recording and interpreting an electrocardiogram (ECG) and evaluating the relevance of the result for management.MethodsAn online case vignette survey was performed among GPs and cardiologists (in 2015). Nine cases describing situations for which Dutch clinical guidelines recommend or advise against recording an ECG were presented. In each case, the participant had to make choices on recording an ECG, interpreting it, and using the result in a management decision. The reference standard for each ECG diagnosis was set by the expert author team.ResultsFifty GPs who interpret ECGs themselves, eight GPs who do not and 12 cardiologists completed the survey. Adherence to guidelines recommending an ECG was high for suspected atrial fibrillation, suspected arrhythmia present during consultation, including bradycardia, but much lower for progressive heart failure and stable angina. Diagnostic accuracy of GPs was best in atrial fibrillation (96%), sick sinus syndrome (85%) and old myocardial infarction (82%), but poor in left anterior fascicular block (16%) and incomplete right bundle branch block (10%). GPs often acknowledged the low relevance of the results of a non-indicated ECG.ConclusionGPs do not fully adhere to Dutch cardiovascular guidelines on indications for recording ECGs. Diagnostic accuracy was high for atrial fibrillation, sick sinus syndrome and old myocardial infarction and poor for left anterior fascicular block and incomplete right bundle branch block.Electronic supplementary materialThe online version of this article (10.1007/s12471-018-1124-2) contains supplementary material, which is available to authorized users.