Extracellular vesicles (EVs) are specialised endogenous carriers of proteins and nucleic acids and are involved in intercellular communication. EVs are therefore proposed as candidate drug delivery systems for the delivery of nucleic acids and other macromolecules. However, the preparation of EV-based drug delivery systems is hampered by the lack of techniques to load the vesicles with nucleic acids. In this work we have now characterised in detail the use of an electroporation method for this purpose. When EVs were electroporated with fluorescently labelled siRNA, siRNA retention was comparable with previously published results (20-25% based on fluorescence spectroscopy and fluorescence fluctuation spectroscopy), and electroporation with unlabelled siRNA resulted in significant siRNA retention in the EV pellet as measured by RT-PCR. Remarkably, when siRNA was electroporated in the absence of EVs, a similar or even greater siRNA retention was measured. Nanoparticle tracking analysis and confocal microscopy showed extensive formation of insoluble siRNA aggregates after electroporation, which could be dramatically reduced by addition of EDTA. Other strategies to reduce aggregate formation, including the use of cuvettes with conductive polymer electrodes and the use of an acidic citrate electroporation buffer, resulted in a more efficient reduction of siRNA precipitation than EDTA. However, under these conditions, siRNA retention was below 0.05% and no significant differences in siRNA retention could be measured between samples electroporated in the presence or absence of EVs. Our results show that electroporation of EVs with siRNA is accompanied by extensive siRNA aggregate formation, which may cause overestimation of the amount of siRNA actually loaded into EVs. Moreover, our data clearly illustrate that electroporation is far less efficient than previously described, and highlight the necessity for alternative methods to prepare siRNA-loaded EVs.