ABSTRACTIntroductionThe value of information obtained from a metabolomic study depends on how much of the metabolome is present in analysed samples. Thus, only a comprehensive and reproducible extraction method will provide reliable data because the metabolites that will be measured are those that were extracted and all conclusions will be built around this information.ObjectiveTo discuss the efficiency and reliability of available sample pre‐treatment methods and their application in different fields of metabolomics.MethodsThe review has three sections: the first deals with pre‐extraction techniques, the second discusses the choice of extraction solvents and their main features and the third includes a brief description of the most used extraction techniques: microwave‐assisted extraction, solid‐phase extraction, supercritical fluid extraction, Soxhlet and a new method developed in our laboratory – the comprehensive extraction method.ResultsExamination of over 200 studies showed that sample collection, homogenisation, grinding and storage could affect the yield and reproducibility of results. They also revealed that apart from the solvent used for extraction, the extraction techniques have a decisive role on the metabolites available for analysis.ConclusionIt is essential to evaluate efficacy and reproducibility of sample pre‐treatment as a first step to ensure the reliability of a metabolomic study. Among the reviewed methods, the comprehensive extraction method appears to provide a promising approach for extracting diverse types of metabolites. Copyright © 2014 John Wiley & Sons, Ltd.