This paper examines the "three factors" approach previously presented by the senior author for determining the net section efficiency of a bolted cold-formed steel open profile. One objective is to ascertain that the net section efficiency is governed by three factors: the in-plane shear lag associated with stress concentration around a bolt hole that is also present in a flat sheet, the out-of-plane shear lag that is also present in a bi-symmetric I-section bolted at the flanges only, and the bending moment arising from the connection eccentricity with respect to the neutral axis. This paper presents the test results of 55 single and back-to-back channel braces bolted at the web including those connected with one row of bolts perpendicular to the axial load. The test results affirm the three factors approach, and it was found that the back-to-back channel braces were affected by local bending even though the connection eccentricity was nominally zero. The paper asserts the need to avoid snug-tightening laboratory test specimens and the importance of identifying the failure modes accurately.