A weather forecast tells you it will rain ten millimetres on Wednesday. What it cannot tell you is where that water goes — which valleys fill, which roads close, whose house floods. That depends on the terrain. The forecast and the terrain answer different questions, and neither makes the other redundant.
The forecast we have built
An individual risk assessment is a forecast. Probability, severity, the expected range of impact — for any given risk these can be characterised with real discipline, and most mature risk functions do it credibly. We have spent decades improving these forecasts: probability distributions, sensitivity analysis, quantitative assessment. What almost no one has built is a map of the terrain those risks sit in.
The terrain we have not
The terrain is the structure: the connections between risks, the pathways effects travel along, the points where separate flows converge. It exists whether or not anyone has mapped it. When the World Economic Forum dropped “ineffectiveness of multilateral institutions” from its 2024 taxonomy, it stopped reporting on the risk. It did not close the pathways — multilateral breakdown still reaches armed conflict, tariff wars and supply-chain stress. Reality does not change because you stop looking at it.
What the terrain makes visible
Once the landscape is modelled, things the forecast cannot surface come into view. Some risks sit where several pathways converge — streams of consequence flowing into one point, each assessed in isolation. Some sit between parts of the system that would otherwise be unconnected, carrying effects across a boundary a list cannot see. A risk that appears moderate in individual assessment may sit where four pathways converge. A risk that appears critical may function in isolation. The forecast cannot tell you which is which. The structure can.
Both questions matter
This is not an argument against assessing individual risks — keep doing it. It is the observation that an individual assessment tells you what each risk is, while the network tells you what the landscape does with it. Most organisations are only asking the first. The structure is already there, already load-bearing. The only question is whether you find it through analysis, or later through impact.