Purpose The construction industry is one of the largest producers of waste and the EU waste framework directive (2008/98/EC) has established a target of 70% of construction and demolition waste (CDW) to be recycled by 2020. However, except for a few EU countries, only about 50% of CDW is recycled (European Commission, 2018). In order to increase recycling rates, sorting of waste material is of high importance (Kurdve et al., 2019). Still, there are different believes of where the sorting should take place; on-site or upstream the supply chain, e.g. at the waste collection company (CIRF, 2019). The purpose of this study is to increase the understanding of the effects that an increased on-site sorting may have on logistics, measured in terms of load factors and collection costs. Research Approach The study takes a quantitative approach, using empirical data from 22 Swedish construction projects of a construction company during the years 2016-2019. Data was gathered from the databases and bookkeeping of the construction company and the waste collection provider. This was complemented with interviews of personnel knowledgeable of the data from the two companies. Findings and Originality No statistic correlation could be found between on-site sorting level and load factor. Thus, an increased sorting level does not per definition add complexity in logistics. The sorting applied in the projects has in total saved about EUR 300 000 for the construction company, compared to not sort at all. This saving is in terms of lower treatment costs. However, storage and handling costs may increase for higher sorting levels, which is not calculated in this study. The saving in treatment cost also differs between material types, where greatest savings occur for combustible material compared with inert material. Research Impact Since 2000, CDW management has been attracting increasing attention from researchers around the world (Kambiz et al., 2016). Earlier studies in CDW management reports on pros and cons of on-site sorting (e.g. Tam et al., 2009; Wang et al., 2010). To our knowledge this is the first study that tries to empirically determine the effect of on-site sorting on load factors and collection costs. Practical Impact The study gives some guidelines in how construction companies could improve sorting rates and at the same time keep collection costs low. For example, our results indicate that it is positive to sort as much as possible of the chosen material from the mixed fractions and carefully choose the fractions to sort.