For almost all quality parameters of a cigarette the methods of measurement have been developed in laboratories. Opinions differ on definitions and the best way of measurement. Thus many long lasting discussions and meetings were necessary before standards could be settled. If we have to solve the problem of measuring a certain quality parameter on-line, we often find that the laboratory method cannot be transfered to process conditions. So for on-line measurement a different method has to be chosen. However, this sometimes causes differences in the results, which again often causes long lasting discussions or other negative economic consequences due to wrong decisions. For the measurement of weight, moisture, firmness, ventilation etc. we find that different methods are in use for laboratory and on-line-measurement. In this paper ventilation measurement is taken as an example to explain this fundamental problem in detail. It is shown that the values of on-line-measurement have an as-good-or-better correlation with the values of smoke analysis as the laboratory values. The author suggests that for laboratory routine control measurements a method be used which can be adapted to process conditions.