A High on two of a hundred and twenty is a Low

An impact in Aspen is one typed impact, rated Low, Medium or High at one org node, with a from and to window. That is the only thing an analyst enters. Every group-level figure in every heatmap is derived from those ratings, and the derivations are worth stating plainly, because they are where a change tool either earns trust or loses it.

Roll-up

Roll-up answers "what does this impact look like from one level up". It weights severity by the share of headcount it touches. A High rated on a role held by 2 people, inside a group of 120, is a High for those two and a small fraction of the group's load. At group level it reads Low.

The alternative, propagating the maximum rating upward, is how most spreadsheets do it, and it is why a portfolio heatmap built that way is red everywhere and tells nobody anything. Weighting by headcount keeps the group-level figure proportional to the group.

Stacking

Stacking answers "what happens when several impacts converge on one team in one window". Three separate Mediums from three separate projects, all landing on the same team in the same month, are a different experience from one Medium. Aspen combines them by a mapping the organisation configures, so the rule that "three Mediums in a month read High" is yours to set and is applied consistently.

The heatmap cell shows the stacked rating and, in small type, the number of impacts that combined to make it, so the reader can tell a single large impact from a pile of small ones.

Inheritance

Inheritance runs the other way. A rating on a group flows to its descendants: rate a Medium on Finance and every team and role beneath Finance carries that Medium until someone says otherwise. A per-child override records where one team's experience differs from its parent's, and the override is itself a rated impact with its own history.

This is what lets an analyst start coarse and refine. Early in a change, one rating at the org-unit level is honest. Later, as the impact assessment matures, the overrides carry the detail.

Pressure

Pressure is the forward projection: for each team, the impact load over time, computed from every impact window across every programme and initiative. It is the derived view that makes convergence visible before the month it happens.

Why none of it is hand-entered

A derived figure that can also be typed in stops being derived. The moment a group-level rating can be edited directly, it can disagree with the ratings beneath it, and nobody can say which is true. Aspen holds the leaf ratings and the configured mappings, and computes the rest. Reassessments are appended, so the history of what was believed at each point is kept.