AbstractExtended keynote paper of Eurosteel 2021This two‐part paper provides a state‐of‐art report on the most recent findings concerning the behaviour, strength and Direct Strength Method (DSM) design of cold‐formed steel (CFS) columns and beams affected by mode coupling phenomena not adequately covered by the current specifications for CFS members, namely local‐distortional (L‐D), local‐distortional‐global (L‐D‐G), distortional‐global (D‐G) and global‐global (flexural‐torsional/flexural – FT‐F) interaction. The paper addresses experimental tests, numerical simulations and DSM‐based design approaches that are intended to i) acquire in‐depth knowledge on the non‐linear behaviour (elastic and elastic‐plastic), load‐carrying capacity and failure mode nature of the members under consideration, and ii) make use of that knowledge to develop, propose and assess the merit of efficient DSM‐based design approaches to estimate their failure loads/moments. Initially, illustrative column results are briefly presented to help grasp some fundamental concepts, namely the characterisation of i) the aforementioned mode coupling phenomena, ii) different sources of mode interaction that may lead to failure load/moment erosion, and iii) the most detrimental initial geometrical imperfections. The DSM design curves currently codified and two strength curves recently developed for column flexural‐torsional and beam distortional failures are presented next. The two‐part paper then addresses separately each mode coupling phenomenon dealt with, for columns, but only L‐D and D‐G interaction for beams – while L‐D interaction (in columns and beams) appears in Part 1, the remaining column and beam coupling phenomena (all involving global buckling) are dealt with in Part 2 [1]. For columns undergoing L‐D and L‐D‐G interaction, beams experiencing L‐D interaction and angle columns susceptible to FT‐F interaction, the work reported includes experimental studies, numerical simulations and DSM‐based design considerations and/or guidelines. For the remaining coupling phenomena, only numerical results are reported, but they unveil interesting (and unexpected) behavioural features that will help plan future test campaigns and achieve efficient design approaches. Finally, the two‐part paper closes with a few concluding remarks and an outlook regarding future developments in this field.