An analysis is exactly as honest as its setup. That is why half the work happens before pressing Run.
I define the analysis setup: fixtures, applied forces, self-weight, safety factors
Loads come from codes, not guesswork: snow and wind to EN 1991, or to DBN V.1.2-2 for Ukrainian projects
I mesh with refinement at joints and check that further refinement no longer changes the result
I calculate von Mises stress, displacement and factor of safety against the steel's yield strength
I give a verdict — holds or not — and specific section changes where needed