The most common mistake I see engineers make with cold-formed steel is treating it like a small hot-rolled section: same instincts, same code assumptions, thinner metal. It is a natural assumption and it is wrong, often in ways that are unconservative. Cold-formed steel is its own discipline, governed by its own specification, and a few of the differences will quietly ruin a design if you import hot-rolled habits.
Different specification, different worldview
Hot-rolled steel is designed to AISC 360, currently the 2022 edition with 2016 and 2010 still in use depending on your adopted code. Cold-formed steel is designed to AISI S100, the North American Specification. These are two different documents built on different premises. AISC 360 assumes sections stocky enough that they can usually reach yield, so local buckling is a special case. AISI S100 assumes the opposite: the plate elements are thin, they buckle before the section yields, and you almost never get the full gross-section capacity. The entire method is organized around that fact.
The behaviors that cause trouble
A handful of behaviors have no real counterpart in routine hot-rolled work.
Local buckling is the normal case in cold-formed design. The thin flats buckle early, so capacity is governed by post-buckling behavior, handled either by reducing the section to effective widths or, in the Direct Strength Method, by an elastic buckling analysis. I wrote separately about the third mode, distortional buckling, which is the one hot-rolled training leaves out entirely.
Web crippling is a real limit state. A concentrated load or a reaction at a support can crush the thin web locally, a failure you essentially never design against in a hot-rolled beam but routinely check in cold-formed members at bearing points and load points.
Cold work of forming changes the steel. The act of bending the corners raises the yield strength locally, and the specification lets you take credit for it in the right circumstances, an allowance that has no equivalent in hot-rolled design.
Connections are different. You are working with screws and the occasional weld or bolt in thin material, with tilting, bearing, and pull-out limit states specific to thin sheet. The bolt-and-weld assumptions of hot-rolled detailing do not apply.
The version question is not academic
Which edition of AISI S100 you are working to can change how you are required to handle these limit states, so it is worth confirming. The Direct Strength Method entered AISI S100 in the 2004 edition as an appendix, became fully integrated into the main body alongside the Effective Width Method in the 2016 edition, and that arrangement continues in the 2020 edition. Your jurisdiction’s adopted building code references a specific edition, so confirm which one governs your project instead of assuming the newest. An old effective-width-only approach and a modern Direct Strength approach with an explicit distortional check can produce different required sections for the same member.
An example from my own work
This is not abstract for me. I built a hat-section calculator that runs the AISI checks across a range of dimensions and loads, precisely so the local and distortional limit states get evaluated correctly every time instead of being estimated by inspection. Cold-formed sections are shaped the way they are, with lips and intermediate stiffeners, specifically to resist the buckling modes described above, and a tool built around that fact produces better results than instinct borrowed from a different material.
If you have cold-formed work, particularly in racking, framing, or any thin-gauge structure, it deserves an engineer who designs it as cold-formed steel on its own terms. That is part of what I do under my structural engineering services.