Extracellular vesicles (EVs) provide a complex means of intercellular signalling between cells at local and distant sites, both within and between different organs. According to their cell-type specific signatures, EVs can function as a novel class of biomarkers for a variety of diseases, and can be used as drug-delivery vehicles. Furthermore, EVs from certain cell types exert beneficial effects in regenerative medicine and for immune modulation. Several techniques are available to harvest EVs from various body fluids or cell culture supernatants. Classically, differential centrifugation, density gradient centrifugation, size-exclusion chromatography and immunocapturing-based methods are used to harvest EVs from EV-containing liquids. Owing to limitations in the scalability of any of these methods, we designed and optimised a polyethylene glycol (PEG)-based precipitation method to enrich EVs from cell culture supernatants. We demonstrate the reproducibility and scalability of this method and compared its efficacy with more classical EV-harvesting methods. We show that washing of the PEG pellet and the re-precipitation by ultracentrifugation remove a huge proportion of PEG co-precipitated molecules such as bovine serum albumine (BSA). However, supported by the results of the size exclusion chromatography, which revealed a higher purity in terms of particles per milligram protein of the obtained EV samples, PEG-prepared EV samples most likely still contain a certain percentage of other non-EV associated molecules. Since PEG-enriched EVs revealed the same therapeutic activity in an ischemic stroke model than corresponding cells, it is unlikely that such co-purified molecules negatively affect the functional properties of obtained EV samples. In summary, maybe not being the purification method of choice if molecular profiling of pure EV samples is intended, the optimised PEG protocol is a scalable and reproducible method, which can easily be adopted by laboratories equipped with an ultracentrifuge to enrich for functional active EVs.