The embodied carbon decisions that are already made by stage 2

Life cycle assessment has a timing problem, and it is not a technical one. The methodology is mature. The data is adequate. The problem is that LCA is most often commissioned as evidence, at the point where evidence is required, and it is most valuable as a design instrument, at the point where the design can still move.

On a typical commercial building, the substructure, superstructure and facade dominate embodied carbon. Those three element groups are also the ones that get resolved earliest and become most expensive to revisit. By the time a project is assembling its certification submission, the frame material is decided, the grid is decided, the facade specification is largely decided, and the honest scope for reduction has narrowed to finishes and substitutions at the margin.

This is how a project ends up with an accurate assessment and a disappointing number. The assessment is not wrong. It arrived after the decisions it should have informed.

What changes the outcome is running the assessment while options are still genuinely open — and accepting that an early assessment is coarser. At concept, you do not have a bill of quantities. You have areas, a structural strategy and a facade intent. That is enough to compare a steel frame against concrete, a fully glazed envelope against a punched-window one, or a new build against a retrofit, with an error band wide enough to be honest about and narrow enough to separate the options.

The useful question at that stage is not "what is our number?" It is "which of these choices moves the number most, and is that choice still available to us?" A comparison with a known margin of error, delivered while the design is fluid, beats a precise figure delivered when nothing can change.

There is a commercial argument here too, and it is the one that tends to land. Carbon reduction identified early is usually cheaper, because it is a specification decision rather than a redesign. Carbon reduction identified late is a variation. If value engineering is going to happen anyway — and it is — then having a carbon column in that conversation alongside the cost column is the difference between informed trade-offs and accidental ones.

So commission the assessment earlier than the certification timetable demands, ask for option comparisons rather than a single figure, and require the output to name which decisions are still reversible. The submission-grade calculation can follow later; it is the cheaper half of the work.