The performance of bulk heterojunction (BHJ) organic solar cells can be affected by a range of factors including the materials combination, processing solvent, post deposition annealing, and/or whether they are used in a conventional or inverted architecture. In this study we compared conventional and inverted BHJ solar cells composed of a non-polymeric donor (5 Z ,5′ Z )‐5,5'‐[(5‴,5‴''''‐{4,8‐bis[5‐(2‐ethylhexyl)‐4‐ n -hexylthiophen‐2‐yl]benzo[1,2‐b:4,5‐b']dithiophene‐2,6‐diyl}bis{3′,3'',3‴‐trihexyl‐[2,2':5′,2'':5'',2‴‐quaterthiophene]‐5‴,5‐diyl})bis(methanylylidene)]bis[3‐ n -hexyl‐2‐thioxothiazolidin‐4‐one] (BQR) and [6,6]-phenyl-C 71 -butyric acid methyl ester (PC 71 BM) as the acceptor. It was found that the conventional device structure had power conversion efficiencies nearly two and a half times that of the inverted device, 9.0% versus 4.0%. Through a combination of Shockley equivalent circuit fitting, optical modelling, and light intensity dependent photocurrent measurements we identified that the origin of the power losses for the inverted architecture relative to the conventional device structure arose from a larger component of bimolecular recombination. Neutron reflectometry measurements showed that the origin of the larger bimolecular recombination losses for the inverted device was due to the PC 71 BM phase separating, with a PC 71 BM rich layer located near the anode reducing the hole extraction efficiency. • BQR and PC 71 BM conventional and inverted BHJ solar cells are compared. • The conventional device structure was two and a half times more efficient. • Increased bimolecular recombination identified as the origin of the power losses for the inverted architecture. • A PC 71 BM rich layer near the anode reduced the hole extraction efficiency in inverted device.