The practice of closing the loop between a decision and what actually happened, so advice can be scored rather than merely admired. Most decision tooling stops at the recommendation, which means no one ever learns whether the reasoning was sound or the advisor was lucky.
A working feedback loop captures the outcome — self-reported, or read from the systems where the result actually lands once they are connected — attributes it back to the decision and the reasoning that produced it, and lets advisor calibration improve from evidence. The loop is also the only honest defense against confident-sounding advice: over enough decisions, a track record replaces a tone of voice.