The analysis and design of structures sensitive to second order effects require the consideration of initial imperfections. Scaffold structures, due to their low stiffness, are very sensitive to second order effects. The accuracy of the design of these structures relies in part on the estimation of initial imperfections affecting the structure. In the 1960s, the C.E.C.M., the European Convention of Steel Construction, carried out a series of experimental tests to determine the degree of initial imperfections of each type of cross-section. Five buckling curves were established. The European design code for steel structures, Eurocode 3 Part 1-1, classifies hot-rolled hollow tubular sections as profiles with few imperfections, i.e. buckling curve ‘a’, and cold-formed sections as sections with many imperfections, i.e. buckling curve ‘c’. A test campaign was conducted to analyse the experimental buckling behaviour of hot-rolled and cold-formed hollow sections of scaffold structures. Twenty-nine buckling tests on simply supported members and nine buckling tests on scaffolding ladders, from various European suppliers, were carried out. Statistical studies of the experimental results point out that the estimate of the initial imperfection currently defined for the hot-rolled and cold-formed tubular sections is far too penalizing for sections used in scaffolding.