The Intensity That Improved and the Total That Rose

How to read this sheet

The three effects sum to the absolute change exactly, and the residual is shown so that can be checked rather than believed - a residual beyond floating-point noise means a product group existed in one period and not the other, which the method cannot weight, and the count of those groups is reported beside it. Everything depends on how the output is split: one row makes the mix effect structurally zero and hands the whole change to volume and efficiency, which is not a finding but an absence of one, and splitting by count band or shade depth rather than by department is what makes the answer worth having. The intensity denominator has to be a real physical quantity that both periods measure the same way - a mill that changed how it counts saleable kilograms between the two years is comparing two different questions, and no decomposition fixes that. The sheet takes energy as given and does not know whether a reduction came from a project, a warmer year, a shut department or a meter that was read differently, so the efficiency effect is a residual in the accounting sense and not a proof of any action. Emission factors move on their own: a grid decarbonising underneath you reduces your reported emissions with no change in the plant at all, which is real for a target and false as an achievement, and a mill reporting against a market-based factor should say which. Finally, an absolute target and an intensity target are different promises and this sheet deliberately refuses to reconcile them - a growing mill can meet the second while failing the first for years, and being told so early is the entire point.