Despite the large number of publications, membrane fouling still is not well understood due to the complexity of the interacting phenomena and the multitude of module and reactor configurations as well as wastewaters and operating conditions. To reduce the number of influencing factors, often lab trials are carried out where only the parameter of interest is to be varied. These are either filtration experiments (e.g., filtration mechanisms, fouling rate), biological investigations (e.g., soluble microbial products (SMP) occurrence), a combination of both (e.g., fouling propensity of SMP formed under different conditions) or concern suited cleaning protocols. As the published results of such investigations differ significantly, the question arises if lab investigations are useful for full scale plants. Due to distinct and inherent differences between lab and full scale operation the transferability is considerably limited, e.g., fouling rates measured in lab scale are inappropriate to describe long-term full scale operation. In general, lab scale experiments provide valuable methods for fundamental research but for full scale operation they can only be applied for filtration characterisation on a short-term basis. If done properly, i.e., under comparable hydrodynamic and operating conditions and with fresh sludge they still yield valuable information. This paper discusses options and limitations of lab trials and gives "do's and don'ts" deduced from own experiences in lab, pilot and full scale and from literature examples.