Abstract The development of a uniform and modern quality system for the whole of the Finnish Radiation and Nuclear Safety Authority (STUK) began in 1997. The principle of continuous improvement is widely applied by performing annual self-assessments as well as internal and external audits. One of the most important steps in a quality assurance programme is the proficiency of the personnel. Good initiation, non-stop training during work and knowledge of the physical and chemical properties that affect analytical procedures are the requirements for first-rate and motivated working. Working procedures have been planned and followed through in order to prevent any contamination in laboratories. Considerable emphasis has also been placed on the standardisation of uncertainty estimations. This demands good knowledge of the sources of errors in sample preparation, analysis and measurement. Participating in intercomparison exercises and proficiency tests is one way to demonstrate the validity and quality of the methods used. However, a lack of intercomparison exercises and reference material on certain natural radionuclides occurring in waters and certain artificial radionuclides has complicated the validation and proficiency testing of these measurements. Moreover, the arrangement of intercomparisons has not always been optimal.