# Systems first *The problem is rarely the person* --- When results disappoint, the instinct is to find the person. Almost always, the better question is what the system around them rewards, permits, and makes easy - because the same people, inside a different structure, produce different results. [[Execution trap]] is the worked case: a team branded an execution failure, fixed without changing a single person, by changing the queue and the sign-offs they worked inside. Treating the system as the unit of diagnosis changes several habits at once. Blaming the individual is usually a diagnosis error: the rep discounting too deeply is responding rationally to a comp plan, the team missing deadlines is carrying invisible work, and firing either leaves the structure that produced them fully intact. Incentives become design material - the cheapest fix is often in the terms rather than the behaviour, as when a contract that pays from signature makes a customer roll out sites that no amount of chasing could ([[Customer-funded growth]] carries that lesson). Feedback makes it harder than it sounds, because systems answer back with a delay and sometimes in the wrong direction first. The right structural fix routinely makes the visible numbers worse before better, which is exactly when the pressure to reverse it peaks - [[Inverse response]] is about surviving that stretch. And a decision that everyone genuinely agreed to still dies if the machinery to carry it doesn't exist; [[Two halves of trust]] takes apart why candour without structure, and structure without candour, each fail on their own. The macro version is the same as the micro: before redesigning the people, the product, or the strategy, ask what the current system is optimising for. It is usually doing exactly what it was built to do. The clearest case I've run was a cluster of software products sharing one management team and one monthly review - one selling at five hundred pounds a year, another at fifty thousand, completely different sales motions run as one thing. We split the teams and split the reviews, so each had its own agenda and its own time, and a product that had declined for years went back to growth. ---