Evidence Chain

Every fact travels through about 11 handoffs between the real world and your decision, and the Evidence Chain is the practice of walking those handoffs backward to find the one that is broken.

It is Monday. Your dashboard says customer satisfaction fell from 84% to 71% in three weeks, and you are one meeting away from pulling two engineers off the roadmap to fix it. Nobody in the room doubts the number. Then someone mentions the survey moved from email to an in-app popup last month. The room goes quiet. You were about to spend a quarter of your team's time acting on a chart whose meaning nobody had checked.

The Evidence Chain is the list of 11 places a claim passes through between the world and your action: world, trace, source or sensor, sample, measure, data, inference, belief, decision, action, downstream outcome. The cue is a number that feels solid and is about to move real money or people. Name the action it would trigger, walk backward one link at a time asking what must be true at that link, and stop at the first one you cannot vouch for. That link is what you go check. Here it is the measure: the popup changed who answers, not how happy they are. What changes is that the room stops arguing about the conclusion and starts testing one link. Skip it when the decision is cheap and easy to reverse.